Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Allocation of resources and educational adequacy: case studies of school-level resource use in southern California Title I Program Improvement middle schools
(USC Thesis Other)
Allocation of resources and educational adequacy: case studies of school-level resource use in southern California Title I Program Improvement middle schools
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
ALLOCATION OF RESOURCES AND EDUCATIONAL ADEQUACY:
CASE STUDIES OF SCHOOL-LEVEL RESOURCE USE IN SOUTHERN
CALIFORNIA TITLE I PROGRAM IMPROVEMENT MIDDLE SCHOOLS
by
Deborah Martinez Granger
A Dissertation Presented to the
FACULTY OF THE ROSSIER SCHOOL OF EDUCATION
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF EDUCATION
May 2009
Copyright 2009 Deborah Ann Martinez Granger
ii
Dedication
This is for the incredible people who formed my past, enrich my present, and
provide hope for the future. You teach me something new every day. Your patience and
support while I’ve spent the past few years chasing this rainbow have meant the world to
me. I hope to be there for you as you pursue brass rings of your own. Thank you for
your endless grace in allowing me to think I have something to share.
I love you, Mom and Dad, Denise and Michael, Rudy and Kris – and Dylan
(Li’l D) and Hunter (Li’l H). You are blessings in my life. I love and admire my
amazing sons, Chris and Ryan, for your brilliance and passion for making the world a
better place. My appreciation for your graciousness while my nose has been buried in the
books is second only to my excitement that we all will be graduates in 2009! Manuel, I
know you’ve been meaning to thank me for all the extra time I’ve given you to perfect
your golf game. Please allow me to thank you first for so graciously supporting me while
I ran loose for a bit.
Everything good in me is because of you. Thank you all for believing in
possibilities.
iii
Acknowledgements
Meditation necklaces in some cultures include a 109
th
bead that is distinguished
from the rest. Upon reaching this bead, one is called to remember and thank his or her
teacher. What a gift it is to be surrounded by wise, committed and caring people who
teach me constantly! Words do not convey the boundless appreciation for the “109
th
beads” who helped to make this dream a reality and taught me so much along the way. I
mention a few people here knowing there are so many others who were behind the scenes
and to whom I am eternally grateful.
Deep thanks to my committee members: Dr. Larry Picus, your unflagging support
provided me the opportunity to be on the cutting edge of educational change. Thank your
for entrusting us with your “baby!” Dr. John Nelson, your unblinking commitment to our
group inspired me to persevere. Dr. Guilbert Hentschke, you stepped in at the 11
th
hour
and saved the day. Dr. Dennis Hocevar, you helped set the foundation for this endeavor.
Dr. Sheri McDonald, you have been my sounding board through thick and thin –
thank you for your friendship and letting us carve this path together! I neither would nor
could have done this without you! Dr. Linda MacDonell provided the mentorship that
inspired this adventure. I cannot thank my family enough for your support during this
adventure – as always, you were my safety net and reminder of what matters.
Finally, thank you to the principals of the eight middle schools described in this
project – your commitment to your students and teachers, your capacity to rally the
troops, and your ingenuity are inspirational.
iv
Table of Contents
Dedication ii
Acknowledgments iii
List of Tables vii
List of Figures ix
Abstract x
Chapter 1: Introduction to the Problem of School Finance Adequacy 1
Broad Issues in School Finance 2
Statement of the Problem 13
Purpose of the Study 16
Research Questions 17
Summary of Methodology 18
Limitations 19
Delimitations 20
Assumptions 21
Definitions 21
Chapter 2: Literature Review 29
Educational Resource Use 31
Resource Use for Instructional Improvement 36
Educational Adequacy 43
Costing Out Adequacy 48
Reform – Research and Recommendations 59
Professional Development 66
An Environment Conducive to Adequacy 70
Impact of Litigation, Legislation and Policy on Educational 76
Adequacy in California
Essentials for School Improvement 82
Summary 83
Chapter 3: Methodology 85
Overview 85
Description of the Research Design 87
Data Collection 103
Data Analysis 106
v
Chapter 4: Results of the Study 110
Restatement of Research Questions 110
The Sample Schools 111
Presentation of Findings 139
School Resource Use 144
Summary and Discussion 197
Chapter 5: Summary, Conclusion, and Recommendations 208
Background 208
Delimitation of the Problem 212
Research Questions 215
Discussion of Findings 216
Question One: What are the Current Instructional 216
Improvement Strategies at the School Level?
Question Two : How are Resources Used to Implement 219
the School’s Instructional Improvement Strategies?
Question Three: How are the Actual Resource Patterns 226
at the School Sites Aligned with or Different from the
Resource Use Strategies that are Used in the Evidence-Based Model?
Question Four : How does the Availability of Resources Affect 252
the Development and Implementation of the Instructional
Improvement Plan?
Guiding Forces for California Middle Schools Alongside the 254
Evidence-Based Model
Emerging Insights 257
Considerations for Future Research 261
Concluding Remarks 262
References 263
Appendices
Appendix A: Informed Consent 278
Appendix B: Invitation to Participate 279
Appendix C: School Visit/Interview Dates 280
Appendix D: List of Documents and Artifacts Provided by 281
Participating Schools
Appendix E: Data Collection Codebook 282
Appendix F: Open-Ended Data Collection Protocol 293
Appendix G: Data Collection Protocol 295
Appendix H: Analysis Process 306
Appendix I: Case Study – Botticelli Junior High School 310
Appendix J: Case Study – Chagall Junior High School 349
Appendix K: Case Study – Degas Intermediate School 381
Appendix L: Case Study – Picasso Middle School 413
Appendix M: Case Study – Renoir Intermediate School 451
vi
Appendix N: Case Study – Rodin Intermediate School 480
Appendix O: Case Study – Van Gogh Middle School 513
Appendix P: Case Study – Warhol Junior High School 550
vii
List of Tables
Table 2.1: Key Schoolwide Expenditure Elements Comprising the 53
School-level Expenditure Structure
Table 3.1: Evidence-Based Adequate Staffing and Resource 97
Components for K-12 Schools – Core Staffing and Specialist
Teachers
Table 3.2: Evidence-Based Adequate Staffing and Resource 97
Components for K-12 Schools – Extended Support
Table 3.3: Evidence-Based Adequate Staffing and Resource 99
Components for K-12 Schools – Intensive Professional
Development
Table 3.4: Evidence-Based Adequate Support and Resource 99
Components for K-12 Schools – Additional Resources
Table 3.5: Technology, Equipment, and Instructional Materials 100
Table 4.1: Student Demographic Groups Comparison 136
Table 4.2: 2008 API by Student Group 137
Table 4.3: 2008 Program Improvement Status and Prior Years 137
in Program Improvement
Table 4.4: Average (Mean) Annual API Growth Before and 138
After Attaining Program Improvement Designation
Table 4.5: 1999-2008 Overall API – Case-Study Schools and 139
California Middle Schools Overall
Table 4.6: Core Class Size Analysis 145
Table 4.7: Elective/Specialist Classes as a Percentage of 146
Core Classes
Table 4.8: Library Staffing Analysis 149
Table 4.9: Librarian- and Library Staff-to-Student Ratios 150
viii
Table 4.10: Summer School Instructional Minutes and 162
Session Length
Table 4.11: Extra Help for Struggling Students 165
Table 4.12: Plans for Regular School Day Professional Collaboration 170
and Learning
Table 4.13: Allocation of Resources for Professional Development 176
Table 4.14: Allocation of Guidance Counselors 180
Table 4.15: Student Support Resource Allocation 184
Table 4.16: Allocation of Principals and Assistant Principals 185
Table 4.17: Secretary and Clerk Allocations 187
Table 4.18: Student Activities 189
Table 4.19: Allocation of Computers 193
Table 4.20: Aggregate Average Resource Use in Full 194
Time Equivalencies
Table 4.21: Average Resource Use in Dollars 195
Table 5.1: Warhol JHS – Percentage of Proficient and Advanced 218
Scores on the 2008 Grade Eight English Language Arts CST
by Student Subgroup
Table 5.2: English Learner 2008 CST Proficiency Percentages 219
by Subject Area
Table 5.3: Resources for School Improvement Strategies 224
Table 5.4: Average (Mean) Class Sizes at Schools Using 243
Extraordinary Resources to Reduce Class Sizes
Table 5.5: 2007 – 2008 Achievement Gains at Schools with 245
Reduced Class Sizes
Table 5.6: Comparison: Taking Center Stage – Act II, Evidence 255
Based Model, and Essential Features of Improving Schools
ix
List of Figures
Figure 2.1: The Evidence-Based Model of school finance 58
adequacy: middle school staffing elements.
Figure 2.2: Costs and benefits of an excellent education 61
per high school graduate.
Figure 3.1: Evidence-Based Model for adequate resource 96
allocation at the middle school level
x
Abstract
State and federal legislation and litigation reflect the belief and hope that public
education may eradicate poverty, sustain a democratic society, and maintain a powerful
economy. The current standards and accountability movement provides a stage for
examining the interaction between educational resources as inputs in relation to student
achievement outcomes.
Researchers currently explore connections between instructional practices
associated with dramatic student achievement gains and necessary personnel and
materials to implement those practices. Their findings hold the potential for identifying
the true costs of adequately educating K-12 students.
At a time of fiscal stress in California, this study supports the search for the best
possible educational outcomes given limited financial resources. After examining
evidence-based instructional practices associated with student achievement as well as
several “costing out” approaches, the investigator will conduct a series of case studies of
Title I Program Improvement middle schools to describe efforts to raise student
achievement when the stakes are high, and to compare these efforts to the Evidence-
Based Model for costing out an education in which essentially all students attain
academic proficiency.
1
CHAPTER ONE
Introduction to the Problem of School Finance Adequacy
How do schools faced with increasing scrutiny over providing students with an
adequate education allocate scarce instructional resources to deliver the best possible
educational outcomes? How might the ingredients associated with research-based best
practices for school improvement be transformed into viable school budget items?
California’s Serrano v. Priest decisions in 1971 and 1976 (EdSource, 2008a) initiated a
national trend toward school finance litigation based on the equal protection clauses in
state constitutions (Kauffman, 2004). The focus of state-level school finance litigation
across much of the United States since the 1990s has shifted from fairness via equal
distribution of funds across schools to fairness via provision of the funds necessary for
schools to adequately educate students per state and federal expectations. Establishment
of legitimate student academic attainment expectancies redirected school finance
litigation away from simple equal distribution of dollars toward provision of necessary
resources for students to attain educational outcomes demonstrating knowledge and skills
necessary for personally satisfying lives and capacities to contribute to society.
A number of school finance adequacy studies have followed court decisions
declaring state school finance systems unconstitutional. While California has not been
subject to such litigation, its legislators, educators, and community members grapple with
a complex school finance system through which schools are funded well below the
national average (Chambers, Levin, & DeLancey, 2006; Kocivar, C. & Halladay, S.,
2005; National Center for Education Statistics, 2005). A growing number of schools
2
struggle to meet educational objectives and operate within a political funding paradigm
that is unresponsive to schools that state: “This is what we need to adequately educate our
students and this is what it costs.” The sections that follow illustrate the broad issues in
school finance pointing to the potential of school finance adequacy as a mechanism for
rethinking allocation of scarce instructional resources.
Broad Issues in School Finance
Fairness at the Federal Level
Seminal federal attempts at educational fairness have led to significant school
finance implications. A number of examples illustrate this point. In the Brown v. Board of
Education Supreme Court decision of 1954, the Court concurred with the plaintiff’s
contention that “segregated public schools are not ‘equal’ and cannot be made ‘equal,’
and that hence they are deprived of the equal protection of the laws” (The National
Center for Public Policy Research, 2008). As a result of the Brown v. Board decision
large urban schools districts including Los Angeles Unified School District in California
faced lawsuits of their own and were compelled to develop mandatory busing programs
to reduce de facto segregation within the school district (U.S. Supreme Court Center,
n.d.). Ongoing efforts to desegregate schools have included development of magnet
schools and a fleet of buses that numbered 2,600 in 2001 (South Coast Air Quality
Management District, 2001). Expenditures for student transportation and for
development of specialized schools reflect local school finance implications of decisions
made at the national level.
3
Passage of the Elementary and Secondary Education Act (ESEA) in 1965
reflected a national attempt to eliminate poverty through education. Title I of the ESEA
directed $1 billion toward compensatory education programs for school-age children
living in poverty. In a press release from the U.S. Department of Education dated
February 6, 2006, the President announced a fiscal year 2007 budget request including
$12.7 billion for Title I programs (U.S. Department of Education, 2006). The share of
federal funds to provide educational services for educationally vulnerable children
increased from 5.7% in 1990-91 to 8.3% of the total spent on public education in 2004-05
(U.S. Department of Education, 2007). In ESEA’s current incarnation as the No Child
Left Behind Act (NCLB) of 2001, acceptance of federal Title I funds by school districts
requires acceptance of federal student performance expectations and sanctions to schools
and districts for unacceptable levels of student achievement (U.S. Department of
Education, 2005). In a letter to the state superintendent of public instruction in Utah in
2004, the acting deputy secretary of the U.S. Department of Education indicated that
turning down Title I funds would result in forfeiture of funds for Reading First,
Educational Technology State Grants, Safe and Drug Free Schools, 21
st
Century
Community Learning Centers, Comprehensive School Reform, Even Start, and Education
for Homeless Children and Youth (Hickock, E., 2004, Letter to Stephen O. Laing). In
addition to earmarked programs, use of Title I funds is limited to specific categories of
support to specific students. Proscriptions, expectations, and sanctions associated with
federal Title I funding under NCLB exert a powerful influence upon school finance at the
school, district, and state levels.
4
Enacted by Congress in 1975 and updated in 2004, the Individuals with
Disabilities Education Act (IDEA) outlines expectations for education and inclusion of
the nation’s more than six million eligible students (National Dissemination Center for
Children with Disabilities, n.d.). This federal promise included a commitment to
underwrite 40% of the average costs incurred in providing equal opportunity to special
education students (National Education Association, 2008). The National Education
Association reported that in 2004, in addition to the locally allocated per-student funding,
the average cost per special education student was $9,369. Of this, the federal
government paid approximately 20%, resulting in a $10.6 billion shortfall to states and
school districts. Underfunded mandates such as the IDEA encroach upon the general
fund dollars intended for all students.
At the federal level, the notion of fairness changed with the June 2007 U.S.
Supreme Court decision that affirmative action plans may not include classification of
students by race. As a result, school district enrollment plans including race/ethnicity
quotas to support desegregated campuses became unconstitutional. Rebell (2008)
celebrates this decision in that it directs attention away from the color of students’ skins
and toward educational adequacy. By pursuing educational adequacy instead of racial
integration, Rebell suggests the possibility of achieving the equal protection goals of
Brown v. Board of Education, an idea also advanced by the late Albert Shanker
(Kahlenberg, 2008). The trend favoring plaintiffs in educational adequacy cases since
1989 requires “that all schools must receive the resources necessary to provide students
with the opportunity for a meaningful education that enables them to meet challenging
5
new state standards” (Rebell, p. 432). The federal Supreme Court decision to prohibit
race- or ethnicity-based equal protection efforts compels schools, districts, and states to
reconsider how to use their resources to best meet all students’ educational needs.
Federal actions to increase educational fairness have resulted in financial
consequences for states, districts, and individual schools. At the same time, school
revenues have increased steadily, outpacing inflation, and securely placing education
revenues above those of the highest-earning private companies in the nation. The
continually rising costs of education lead educators, researchers, and policymakers to
wonder if the results justify the costs. The section that follows illuminates the state of
education spending and the controversy surrounding it.
Education Spending and Controversy
A review of education spending by Hanushek and Rivkin (1997) shows that from
1890 to 1990 real dollar expenditures per pupil increased approximately 3.5% annually.
Researchers have wondered why student achievement outcomes appear paltry relative to
the public investment. As a result, they have analyzed national school expenditure data in
attempts to understand what happens to the money that flows into education. A number
of findings consistent across studies have emerged.
Odden, Monk, Nakib, and Picus (1995), Picus, Tetreault and Murphy (1996) and
others report that roughly 60% of education expenditures per pupil are directed toward a
large category called “Instruction.” Odden and Picus (2008) summarize the findings and
outline common expenditure distribution patterns. The 40% of funds outside instruction
are distributed across six expenditure categories: (a) instructional support – 5%,
6
(b) student support – 5%, (c) site administration – 5%, (d) district administration – 5%,
(e) district-run maintenance and operations – 10%, and (f) transportation, food, and
miscellaneous other items – 10%. Regardless of actual dollars spent, the distribution
remains consistent across expenditure categories. This finding led Hanushek and Rivkin
(1997) to peek inside the seemingly black box of instruction to analyze how funds
earmarked for instruction actually are distributed. Their analysis concluded that high-
interest, non-core courses have been added to the offerings within secondary schools as
funds continually increase.
Additionally, academic core courses have not received a relatively greater number
of additional teaching staff although these are the areas in which student performance is
assessed for state and federal school achievement purposes. Adding compensatory
education and special education specialists also increases the number of credentialed staff
members without reducing class sizes or boosting student achievement. These findings
suggest evaluation of resource allocation based on effectiveness.
Effective resource allocation occupies a prominent position on the education
landscape as NCLB matures and target percentages of proficient students continue to
grow. Additionally, the notion of fairness as it applies to education funding has changed
to the point that school finance litigation across the country has shifted its focus away
from finance equity. A result of this litigation and related legislative action has been
concerted attention directed to effective resource allocation as described in the next
section.
7
Education Finance Litigation and Resource Allocation Decision-making
Education has a long history of being considered a local concern, evidenced by
the 16,025 reported public school districts spread across the United States during the
2003-04 school year (National Center for Education Statistics, n.d.). Until the 1970s,
local control included a preponderance of funding from local sources, typically property
tax revenues. This meant that children from property-rich communities attended better-
funded schools. The resource differential between students in property-wealthy versus
property-poor communities was considered so inequitable that it sparked a California
Supreme Court decision. The public as well as the courts understood that resources
matter.
In declaring the Serrano v. Priest decisions of 1971 and 1976, the California
Supreme Court determined that the state’s education finance system was unconstitutional.
Equal protection, the foundation for the case, was defined as equal distribution of
education funds across all school districts in the state. California’s relatively stable local
financing system gave way to a more volatile one and the state became the chief
underwriter of public education. The school finance equity movement included litigation
and legislation across the U.S. resulting in states assuming both the burden and control of
public education.
Arguments for fairness in school finance allocation shifted from equity to
adequacy beginning with New York’s groundbreaking Campaign for Fiscal Equity case.
The Court of Appeals decided in 1995 that the state was obligated to provide all students
with a sound education (Campaign for Fiscal Equity, Inc., 2008). This and subsequent
8
litigation across the states left state supreme courts to grapple with defining a decent
education. Garfield, Garfield, and Willardson (2003, p. 200) offer a definition of
adequacy that includes the primary purpose of public education:
Democracy is best served by extending to all children an equal
opportunity to attend schools that are adequate for the achievement of self-
realization, economic sufficiency, civic responsibility, and satisfactory
human relationships. Spending the same number of dollars on each
student is evidence of equality, but it may not be equitable.
Courts interpreted previously vague education clauses of state constitutions and
provided preliminary language to describe expected student outcomes. The educational
standards movement, resulting in state-approved and measurable descriptions of what
students may be expected to know and be able to do relieved the courts from the task of
delineating adequate education specifics.
Now that states know what the educational objectives are, the question is how
might they best attain those achievement outcomes? A number of researchers have
explored the relation of educational inputs to student performance outcomes in efforts to
answer the “how” question. Rebell (2007) indicates that adequacy studies have taken
place in 30 states with the intention of determining “objectively the amount of funding
needed to provide all students a meaningful opportunity for an adequate education” (p.
1303, italics in original). Recent finance adequacy studies in California (e.g, Chambers,
Levin, & DeLancey, 2006) used the state’s 2011-2012 NCLB performance targets as
adequacy objectives. Chambers et al. then addressed the question of what resources
public schools in California need in order to ensure an adequate educational opportunity
for their students. Four models describe approaches to determining the necessary
9
resources to obtain adequate educational outcomes. Identification of the necessary
ingredients to achieve learning objectives is the first step in determining costs and
building budgets.
“Costing out” models provide practical guidance to fiduciary decisions. These
models include the (a) Expert/Professional Judgment Model, (b) Cost Function Model,
(c) Successful Schools Model, and (d) Evidence-Based or State-of-the-Art Model
(Hanushek, 2006; Odden & Picus, 2008). A brief orientation to each model assists in
distinguishing each as well as identifying its strengths and challenges.
The Expert/Professional Judgment Model relies upon a panel of education experts
to determine the resources needed for all students to achieve at acceptable levels.
Challenges with the Expert/Professional Judgment approach include biases held by panel
members and a lack of expertise with the specific issues, programs, and populations for
whom the panel is making recommendations.
The intent behind Cost Function analysis is to “determine not only the levels of
present spending that are associated, on average, with a specific set of outcomes, but also
how those levels of spending may vary for districts of different characteristics that serve
different student populations” (Rebell, 2007, p. 1312). This approach depends upon an
extensive range of data as well as complex analyses. Its complexity makes it difficult to
explain to the public and policy makers. Hanushek (2006) criticizes the Cost Function
approach explaining that unlike industry where a drive to be ever more efficient allows a
separation between what is spent and what should be spent to achieve a given goal,
education requires no such attempt at efficiency.
10
At least 10 states have attempted “Successful Schools” studies. Schools that meet
success criteria are analyzed in terms of “resource inputs, student test scores, and other
precisely defined outcome measures” (Rebell, 2007, p. 1311). Critics such as Hanushek
(2006) refer to this as the “Successful Students” model as it offers no explanation of what
leads students to do well. Recent adjustments to control for student academic
vulnerability characteristics have improved this approach to costing out adequacy.
Varying definitions of success result in different resource allocation recommendations.
The Evidence-Based or State-of-the-Art approach to finance adequacy is
grounded in current research and proven best practices in instruction. This empirical base
separates the Evidence-Based approach from the Expert Judgment model. Based on
evidence informing the nature of high-quality instructional programs, Odden, Archibald,
Fermanich, and Gross (2003) classified expenditures across instructional and
noninstructional expenditure categories. Odden and Picus (2008) subsequently developed
resource-based expenditure structures for prototypical elementary, middle, and high
schools predicated on student characteristics and needs. This Evidence-Based model
allows a school budget to be built from the ground up. While this approach has informed
legislative school funding decisions, it faces the challenge of never having been fully
implemented. Being untested, its critics question its viability as an effective model.
However, the alignment of this model with NCLB scientifically-based research as well as
proven best practices confers credibility and justifies exploration of its potential to
dramatically improve student performance. As such, the Evidence-Based Model forms
the foundation for this study.
11
The approaches to school finance adequacy identified above describe potential
mechanisms for costing out the essential ingredients for dramatically improving schools.
Recognition that resources make a difference energized early school finance equity
litigation. Yet, showing how those resources affect student learning has been a difficult
task. Additionally, tightly linking two, differently organized accounting systems, that of
expenditures and that of staff allocations, challenges researchers, educators, and policy
makers. Given the budget crisis facing California in 2008, recognizing the resources
essential to student achievement, and therefore essential budget items, assumes especially
significant proportions. The story of attempts to attach dollars to school improvement
follows in the next section.
Coupling Dollars with School Improvement in Lean Times
The sheer number of students enrolled in California schools (eight % of the
nation’s students) lends significance to state issues when considering the costs of
adequately educating students. Not only is California a large and populous state, but it
also is home to many children who come to school from diverse linguistic backgrounds
and who live in poverty. The education system faces huge demands. In a set of 20
connected studies, Chambers, Levin, and DeLancey (2006) costed out educational
adequacy for California students. Using an Expert Judgment approach, they determined
that the additional costs necessary to achieve adequate student performance outcomes
would increase the state’s school finance budget by 53% to 71% of fiscal year 2005
12
expenditures. Court decisions in adequacy suits in other states along with research
findings indicate that quality and quantity of resources matter. These resources come
with price tags.
California’s 2008 budget shortfall casts a bright light on the question of money as
it relates to educational resource allocation and student achievement outcomes. The
governor announced in February 2008 that all state-funded public agencies must trim
10% from their budgets, eliminating $4 billion from the K-12 public education coffers
(Williams, 2008). Dollars were translated to human resources when California state
schools chief, Jack O’Connell announced in March 2008 that 20,000 teachers, librarians,
nurses, and support staff had received pink slips as a result of the budget crisis (California
Department of Education, 2008b). Alongside this threat of severe staffing cutbacks,
NCLB expectations for California student academic performance simultaneously have
begun a seven-year steady and steep climb toward the 100% target proficiency mark in
2013-14. Which might be the highest-leverage resources in such a bricks-without-straw
landscape? To what extent do teachers and other educational resources matter? Answers
to these questions are critical as funding diminishes.
A number of researchers have debated the question of the influence of resources
on levels of student learning. Results have been mixed, as have been methodologies,
analyses, and variables. While Hanushek’s findings (1989) indicate no positive
correlation between resource levels and student learning, subsequent reviews of the same
data by Hedges, Laine, and Greenwald (1994) and Greenwald, Hedges, and Laine (1996)
show significant positive effect sizes suggesting that resources indeed matter. Archibald’s
13
(2006) research concludes that quality teachers influence student achievement, again
indicating that resources make a difference. In the midst of a state budget crisis in which
resources appear likely to be diminished, the production function debate seems especially
significant. Which resources offer the most potential for which students is a question
researchers and educators strive to understand more thoroughly.
This study explores use of instructional resources at the school level during the
2008-2009 school year in California. With the intent of contributing to the field of school
finance and instructional resource allocation, research conducted for this study reflects an
attempt to connect two critical educational input factors: a) instructional improvement
strategies, and b) instructional resources to implement improvement strategies. It
addresses the problem faced by states and the nation in making good on a commitment to
educate the nation’s children for the subsequent benefit to the individuals as well as to the
prosperity of the country. The financial implications of a national movement toward
educational adequacy have not yet unfolded, but the idea of providing children with what
they need to meet a specified educational expectation suggests a reformation of school
finance policy at the local, state, and national levels.
Statement of the Problem
Schools have not yet explicitly attached dollars to resources that best influence
student achievement. As the stakes rise and resources decline, identification of high-
impact school improvement strategies has become increasingly significant.
Six years into NCLB, the education community struggles to increase the number
of students with proficient and advanced standards test results. A number of effective
14
practices identified in research have been employed in comprehensive school reforms
with high levels of success. Academically successful “beating the odds” schools serving
high needs populations across the nation have employed such practices as smaller schools
(Huebner, Corbett & Phillippo, 2006; Lawrence et al., 2002), and class sizes (Health &
Education Research Operative Services, Inc., 2003), teacher collaboration (Schmoker,
2004), effective professional development (McREL, 2005), technology to enhance
learning (The Education Trust, 2005), and immediate support for struggling students
(Brown, Morris, & Fields, 2005) . These schools have attained exceptional success levels
with their students.
However, many schools serving the nation’s most vulnerable students do not
appear to provide the type, quality or quantity of resources considered necessary for an
adequate education. Litigation resulting in dramatic infusions of funds (e.g., Abbott
districts in New Jersey) has not always led to use of effective instructional practices,
much less improved student achievement. Equivocal results such as these have led both
critics and proponents of school finance adequacy to recommend that school revenues be
spent more intentionally in order to dramatically and positively influence student
performance (Picus, 2000; Hanushek, 2006). Odden, Monk, Nakib, and Picus (1995)
further explain that schools need to be producers of achievement, not merely consumers
of resources.
Odden et al., 2003 conclude that school-level data are essential to understanding
the resource-achievement connection. Teaching and learning occur at the school site as
opposed to the district office or state department of education. Additionally, the school
15
site is where resource allocation directly affects students. Yet, within the fields of
education, research, and policy very little is known about which resources are used in
California public schools, much less about how those resources are distributed to school
clients--students. Identification of instructional programs and strategies in current use,
along with the resources allocated for their implementation has the potential to illuminate
how educators attempt to provide the array of inputs necessary for student achievement
outcomes.
Both state and national corrective actions for schools failing to make adequate
annual progress toward expected achievement goals delineate specific expectations for
rethinking instructional time, professional development, and instructional materials.
Even within these federal Program Improvement (PI) and California School Assistance
and Intervention Teams (SAIT) schools, it is unclear how the mandated reform elements
lead to schools that embrace six core strategies for dramatic instructional improvement
resulting in student achievement gains as described by Odden, Picus, Mangan, and
Fermanich (2006):
1. recalibrating expectations for student learning,
2. reengineering to “deploy more powerful instructional strategies and use
resources more productively” (Odden, Picus, Goetz, Mangan, & Fermanich,
2006, p. 4),
3. redesigning professional development for teachers,
4. reinforcing “must learn” content and skills with students,
5. retooling technology to best utilize on-line resources, and
16
6. restructuring compensation protocols to reward effective professionals.
The purpose of this study is to shed light on how current reform efforts compare
with a scientifically based, highly intentional, and measurable model such as the
Evidence-Based approach. Examination of resource allocation at a variety of schools
within the state offers a starting point for continued research in the area of funding school
improvement.
School finance research findings (Picus, 1993) remind practitioners and scholars
that increased funding does not translate to high-leverage reallocation of resources. When
more money is available, approximately 60% continues to be dedicated to instruction.
Within that 60%, more money does not result in more teachers for high-stakes courses.
Instead, funding increases often support additional electives and non-instructional staff
members. The potential for intentional redirection of funds toward effective practices
seems enormous and timely.
Purpose of the Study
The purpose of this study is to initiate exploration into resource allocation vis-à-
vis the Evidence-based Model as developed by Odden and Picus (2008). Discovering
instructional improvement strategies currently being implemented in southern California
schools supports development of a high-impact resource allocation baseline. Explication
of how resources are utilized at Title I Program Improvement middle schools as well as
comparison of resource use to the Evidence-Based model permits a deeper understanding
of intentionality and comprehensiveness of school improvement efforts.
17
Research Questions
This study endeavors to reveal, through analysis of school-level data, how
California schools currently allocate resources in an environment in which state-approved
standards and performance expectations represent educational adequacy. Answers to the
following research questions will contribute to the foundation of a California database,
which may support future research endeavors as well as inform resource and funding
allocation decisions from the school site to the state capitol:
1. What are the current instructional improvement strategies at the school-level?
2. How are resources used to implement the school’s instructional improvement
strategies? (instructional vision, plan, goals, etc.)
3. How are the actual resource patterns at the school sites aligned with or different
from the resource use strategies that are used in the Evidence-Based Model?
4. How does the availability of resources affect the development and
implementation of the instructional improvement plan?
Questions are derived from a review of literature affiliated with educational
adequacy: resource use and reallocation, improvement of school performance, leadership
for sustained improvement, professional development for sustained improvement, the
education change process, and effects of school finance reform efforts. While the first
three research questions support exploration of use of current resources, the final question
addresses the issue of contemplating and working toward a long-term vision for student
achievement with limited resource options. Miles (2000) speaks to a manifestation of this
in explaining that schools “often make short-term budgeting and organizational decisions
18
without considering a long-range plan to better match their complete set of resources to
their educational strategy” (p. 13).
Chapter Two, the literature review provides the lenses from which the research
questions were developed. Chapter Three delineates the methodology of the study. It
describes the research design, population, sample, instruments, procedures, data analysis
approaches, and ethical issues. Chapter Four presents the findings. Chapter Five
summarizes the study and offers considerations for future endeavors in the study and
provision of an adequate education to California’s children.
Summary of Methodology
This study consists of a modified case study based on a purposeful sample of
eight southern California Title I Program Improvement middle school principals and
other key information providers identified by the principals. Selection of schools was
based upon Program Improvement status as of September 2007. Principals were asked a
series of open- and close-ended interview questions and provided the investigator with
public-access documents for triangulation of data. The author contributed to customizing
the existing Evidence-Based Code Book previously used in finance adequacy research in
Arkansas to reflect the California context. The Code Book provided the interview
questions and protocol. Questions and documents reviewed pertained directly to the four
research questions.
Data were entered on-line to a password-protected Excel database containing data
from this and other studies using the California Evidence-Based Code Book, and
interview protocol. The researcher analyzed the data from this study to identify current
19
instructional improvement strategies and their associated resource allocations. These
were compared with the Evidence-Based model. Finally, the researcher sought patterns,
trends, and themes regarding availability of resources and decisions about program and
strategy implementation.
Limitations
This study was based on a purposeful sample of eight Title I middle schools with
Program Improvement status as of September 2007. Student demographics varied
somewhat from school to school and data collection occurred between September and
November 2008. Statistical generalizability does not extend beyond the schools studied.
However, patterns and trends may serve to inform schools with similar demographics and
sanctions status. Attempts at accuracy in data included triangulation through open-and
close ended interview questions and review of related documents. However, data
accuracy depended to some extent upon interviewee memory. This study represents a
snapshot of eight schools during a uniquely challenging point in the state economy in
California as opposed to a longitudinal perspective of school reform efforts. Short-term
budget issues may influence school-level decision-making regarding resource allocation
in ways unknown to the researcher.
Additionally, sampling issues included the issue of bias inherent in researching a
sample of volunteers. Additional issues emerged with accurate data collection in the area
of summer school. Centralization of summer school resulted in principals not having this
information easily available and some district office reports were not organized to
provide the type of data (e.g., number of teachers from a given school teaching in the
20
district summer school program, number of students from a given school attending the
district summer school program). While sampling issues limit generalization to the
schools analyzed, the data nonetheless will provide a valuable first step toward
understanding the resource-achievement connection in California.
Delimitations
This study entailed a review of instructional strategies and resources allocated to
implement them at a purposeful sample of eight Title I Program Improvement middle
schools in southern California in Fall 2008. It included open enrollment, neighborhood
public middle schools and did not include privately managed, charter, magnet, or voucher
schools. By virtue of their Program Improvement designation and the timing of this
study, participating middle schools had not met federal NCLB adequate yearly objectives
for a minimum of two consecutive years. The researcher sought schools at various levels
of the Program Improvement sanctions process.
This study excluded non-school specific resources such as student transportation,
building maintenance and operations, district office operations, and food service.
Expenditures related to debt service and facility construction were beyond the scope of
this study, which emphasized school-level instruction-related resources and their
allocation.
Study participants primarily included the principal at each school. Participants
were asked a series of questions in a specific sequence from the Evidence-Based Code
Book as well as a set of open-ended questions also based on the Evidence-Based Model.
Participants provided documents pertinent to resource allocation (e.g., master schedule,
21
school vision, & bell schedule) for data triangulation. The researcher was trained several
months prior to conducting the interviews in the content and protocols of using the
Evidence-Based Code Book and open-ended interview questions.
Assumptions
By emphasizing the Evidence-Based model, the author assumes that a focus on
instructional practices grounded in current school improvement research offers a
meaningful comparison point for intentional and high-leverage use of scarce resources to
maximize student learning. The author additionally assumes that participants responded
honestly and to the best of their abilities.
Definitions
Significant contributors to the field of school finance have developed definitions
as part of an ongoing effort toward commonly held understandings of terms – both for
budget purposes as well as to standardize jargon used in the field. Such researchers and
expert sources are cited to clarify a number of terms used in this study.
1. (Educational) Adequacy: Student academic performance expectations as determined
by a state’s curriculum standards, NCLB, and measured by a state’s testing system
with a goal that all but the most severely disabled students will achieve at proficient
or advanced levels – and the funding to support necessary resources to achieve these
outcomes (Odden & Picus, 2008).
2. Adequate Yearly Progress: “A goal of the 2001 federal law No Child Left Behind
(NCLB) that requires schools and districts to measure and report students’ annual
22
progress toward proficiency in English/language arts and mathematics by 2013-
2014.
3. Progress is based on whether the school or district met its Annual Measurable
Objectives and demonstrated 95% participation on standardized tests, achieved its
target on the Academic Performance Index and, for high schools, met target
graduation rates” (Ed-Data, 2008).
4. Advancement Via Individual Determination (AVID): National program targeting
underachieving middle- and high-school students to prepare them for college
admissions and achievement.
5. Annual Measurable Objective (AMO): Target percentage of students whose state
test results identify them as proficient or advanced each year as determined by
NCLB.
6. Average Daily Attendance: “The total number of days of student attendance divided
by the total number of days in the regular school year. A student attending every
school day would equal one ADA. Generally, ADA is lower than enrollment due to
such factors as transience, dropouts, and illness. A school district’s revenue limit
income is based on its ADA. The state collects ADA counts at the district but not the
school level” (Ed-Data, 2008).
7. Beginning Teacher Support and Assessment (BTSA): State-mandated program for
teachers new to the profession. Participating teachers receive coaching and support
in instruction and classroom management while engaging in an ongoing reflection
and comparison with the California Standards for the Teaching Profession (CSTP).
23
8. California Basic Educational Data System (CBEDS): “An annual collection of basic
student and staff data that; includes student enrollment, graduates, dropouts, course
enrollment, enrollment in alternative education, gifted and talented education, and
more. Statistical information about schools, teachers, and students that is collected
from each public school on a given day in October” (Ed-Data, 2008).
9. California Standards Tests: “Tests in English/language arts and mathematics in
grades 2-11, science in grades 5, 8, and 9-11, and history/social science in grades 8,
10 and 11 based on California's academic content standards. This is the core of
California's statewide Standardized Testing and Reporting Program (STAR)” (Ed-
Data, 2008).
10. Categorical Fund(s): State and federal funds allocated in addition to revenue limit
income and designated for students with special and specific needs, such as
disabilities, and for special purposes and programs.
11. Certificated/Credentialed Employees: Also referred to as “licensed” employees –
staff and faculty holding appropriate state credentials.
12. Coaching: “refers to opportunities for active learning that are often ongoing in nature
and assist teachers in active learning” (Odden et al., 2002, p. 69).
13. Comprehensive School Reform: scientifically based and effective practice
improvements implemented schoolwide and covering all aspects of a school’s
operations to provide a coherent approach to significant school improvement,
especially for low-achieving children (U.S. Department of Education, 2004).
24
14. Cost Function approach: Econometric approach to identifying costs associated with
achieving desired/required student achievement outcomes.
15. Cost Function: Cost of inputs that result in given outcomes.
16. Cost of Living Adjustment: Inflation adjustment to school revenues from the state.
Commonly referred to by its acronym: COLA.
17. Costing out: Process of determining how much it costs to provide students with an
adequate education. While costing out is not an exact science (Chambers, Levin, &
DeLancey, 2006), it provides transparency with regard to expenditures and relates
inputs to outcomes.
18. Costs: expenditures incurred to produce a certain outcome (Odden et al., 2002, p. 73)
19. Data Director: Subscription-based, online, standards-aligned student assessment data
management system.
20. Education Management Organization (EMO): Private, for-profit companies that run
educational institutions.
21. Educational Productivity: “the improvement of student outcomes with little or no
additional financial resources, or a consistent level of student performance at a lower
level of spending” (Picus, 2000, p. 5).
22. Effective Professional Development: “Professional development that produces
change in teachers’ classroom-based instructional practice, which can be linked to
improvements in student learning” (Odden et al., 2002, p. 53).
23. Encroachment: Situation in which a required or restricted program costs more to
operate than funding that is provided, thereby requiring use of unrestricted funds
25
intended for all students to support programs that benefit specific populations of
students.
24. English Learner: Most current designation for students not yet sufficiently proficient
to access and learn from the regular instructional programs offered at the school.
English language proficiency is assessed annually; synonymous with English
language learner.
25. Equalization Aid: Legislatively approved discretionary increases to increase revenue
limits for districts with limits relatively lower than other districts.
26. Equity: Equalization of funding across per pupil expenditures.
27. Evidence-Based Model: Costing out approach grounded in scientifically based
research and widely documented effective practices associated with instructional
improvement and student achievement gains leading to desired/required student
educational goals.
28. Expenditures: what is spent (Odden et al., 2002, p. 73).
29. Expert Judgment approach: Use of a panel of educators, specialists, and
administrators to determine resources necessary to achieve desired/required student
achievement outcomes.
30. Free/Reduced Price Lunch: Program to provide food to low-income students.
31. General Fund: As an accounting term, General Fund refers to all general use
expenditures not required or permitted by law to be accounted for in a separate
accounting category.
26
32. Guided Language Acquisition and Design (GLAD): Research-based instructional
protocol combining development of reading, writing, and vocabulary development
with content instruction.
33. Input-based Accountability: Emphasis on effective delivery of instruction to
determine teacher and school success.
34. Mandated Cost Reimbursements: Third largest state categorical program. Refers to
“payments to districts for costs they incur complying with various state or court
mandates” (Rose, Sonstelie, Reinhard, Heng, 2003).
35. No Child Left Behind: “The 2001 reauthorization of the federal Elementary and
Secondary Education Act (ESEA) that places comprehensive accountability
requirements on all states, with increasing sanctions for schools and districts that do
not make adequate yearly progress toward proficiency in English/language arts and
mathematics or that fail to test 95% of all students and all significant subgroups. In
California, those sanctions currently apply only to schools and districts that accept
Title I funding” (Ed-Data, 2008).
36. Output-based Accountability: Emphasis on student performance on state-mandated
exams to determine success of teacher and school.
37. Peer Assistance and Review (PAR): State-mandated program for tenured teachers
who fail to meet performance expectations or who request support. Participating
teachers receive coaching and support in instructional strategies and classroom
management from colleagues.
27
38. Production Function: Input factors that result in given outcomes.
39. Professional Learning Community: (Group of) school staff members improving
instruction by working collaboratively around a common vision for student
achievement and using data to inform shared decision-making regarding both student
and professional learning necessities (Hord, 1997).
40. Program Improvement: Federal status designation triggering a series of expected
school improvement actions for schools receiving Title I funds that fail to make
adequate yearly progress (AYP) as defined by NCLB.
41. Proposition (Prop) 13: An initiative amendment to the California Constitution passed
in June 1978 restricting property taxes to no more than 1% of "full cash value."
“Proposition 13 also defines assessed value and requires a two-thirds vote to change
existing or levy new special purpose taxes” (Ed-Data, 2008).
42. Redesignated Fluent English Proficient (RFEP): Students who meet language
proficiency and academic achievement criteria to change their English language
proficiency classification from English Learner (EL) to English Proficient (RFEP).
43. Revenue Limits: Expressed in dollars per student attending school within a district,
and determined by the prior year’s limit plus an inflation adjustment (i.e., cost-of-
living adjustment or COLA).
44. Socioeconomically Disadvantaged: Students who participate in the free/reduced
price lunch program serves as a proxy for students living in poverty.
28
45. Special Education: Identification processes, individual education plans (IEPs) and
programs to meet educational needs of emotionally, physically, and learning disabled
students – required by federal law through age 21.
46. State-of-the-Art approach: See “Evidence-Based” model.
47. Successful Schools approach: Identification of programs, strategies, and resources
used in “successful schools” with the intention of transferring those programs, etc. to
other schools to achieve desired/required student achievement outcomes.
48. Title I schoolwide program: Schools that receive Title I compensatory education
funds that opt to use funds for all of the school’s students. Schools are permitted to
consolidate federal, state, and local funds to implement a school reform program.
Schools commit to Federal NCLB adequate yearly progress expectations for all
students, not just Title I students.
49. Title I targeted assistance program: Schools that use Title I compensatory education
funds specifically for Title I-eligible students identified to be at greatest risk of
academic failure. Schools commit to NCLB adequate yearly progress expectations
for demographic subgroups to which participating students belong.
50. Title I: Federal funds for “educationally disadvantaged children” (Ed-Data, 2008)
from the Educational Consolidation and Improvement Act.
51. Training: “refers to teachers sitting and getting training of any length, from one-day
workshops to three-week summer institutes” (Odden, Archibald, Fermanich, &
Gallagher, 2002, p. 69).
29
CHAPTER TWO
Literature Review
Fairness as it pertains to opportunities for all children to succeed academically has
been redefined in a shift from efforts to promote equity through equal resource allocation
to those promoting proficiency at state and federal levels through allocation of resources
that will result in state and nationally determined achievement levels. Promotion of
adequacy through provision of appropriate resources has replaced promotion of equity
through distribution of equal shares of the education pie. Endeavors to provide students
with an adequate education are complicated by the magnitude of change suggested by
litigation and research. The political nature of legislative decisions regarding allocation
of funds reveals the tension between doing what is best for student learning and what
voters will support. The move toward educational adequacy has been characterized by
litigation focused on appropriate resources to meet students’ needs, descriptions of
resource use at the school level, and identification of resource use associated with student
achievement. The research focus has moved from a macro- to a micro-perspective.
Instead of looking at education in terms of systems, current researchers emphasize
schools as the critical hubs where instruction takes place and teachers as the actors with
the greatest power to influence student achievement.
This review of literature takes place as California public schools contend with
pressing issues of fiscal stress and increasing No Child Left Behind proficiency targets.
It is during times such as this that the K-12 education community relies on solid
information regarding essential elements of an adequate education. Multiple tensions
30
influence urgent issues of resource prioritization: tension across levels and branches of
government, research outlining effective school mechanisms and inertia regarding
significant change, and competition for scarce resources. While this study has an
essentially pragmatic purpose, it reflects the complexity of school finance and
improvement efforts today.
This literature review consists of nine sections. The first, “Educational Resource
Use” describes the growth in education spending over time, the consistency in the way
the education dollar is divided, and the need to understand education spending details –
especially within the “instruction” expenditure function. Section two, “Resource Use for
Instructional Improvement,” provides a look into use of instructional resources to
improve student learning – resources funded from the instruction expenditure function.
Section three, “Educational Adequacy,” tells the story of the shift from equity to
adequacy catalyzed by the standards and accountability movements. The fourth,
“Costing Out Adequacy,” details the research and development of an expenditure
tracking model based on the resource allocation strategies of continuously successful and
improving schools. This model makes it possible to attach dollars to specific
instructional choices. This section additionally includes descriptions of four “costing
out” models used to determine adequate funding levels for all students. The Evidence-
Based Model receives particular attention as it is a focal point of this study.
The fifth section, “Reform – Research and Recommendations,” points to the
political, social, economic, and personal urgency of making swift and dramatic changes
to ensure that all children learn what they need to become successful adults. Including
31
research-based macro- and micro-policy recommendations, this section provides
additional support to the Evidence-Based Model described in section four.
Section six, “Professional Development,” highlights a component of school
improvement considered critical to developing the necessary capacities within and among
educators to change what and how they teach to meet 21
st
century student needs.
Literature grounded in change leadership and educator professionalism undergirds
section seven, “An Environment Conducive to Adequacy.” Section eight, “Impact of
Litigation, Legislation, and Policy on Educational Adequacy in California,” provides the
context for this study and analysis of findings. It highlights the difficulty inherent in
addressing critical social issues such as educational adequacy.
The ninth section, “Essentials for School Improvement,” is based on successful
schools and districts in Washington State. Case study analysis revealed unilateral themes
that characterized instructional improvement visions and innovations in successful
schools. These core elements reflect a singular disposition toward student learning and
inform post-interview analysis of this research study. These nine topics reflect the current
scope of school finance research and inform this study of school-level resource use.
Educational Resource Use
Resource Allocation Patterns and Changes – A Macro View
As outlined below, researchers have found remarkable consistency across the
United States and over time with regard to expenditures across spending categories
defined by the National Center for Educational Statistics. Recognition that expenditures
for education have outpaced inflation without resulting in dramatic increases in
32
achievement has led researchers to examine what is taking place within specific spending
categories.
Growth and entrenchment. Hanushek and Rivkin (1997) and Hanushek (1998)
describe real dollar expenditure increases averaging 3.5% annually over the one hundred
year period spanning 1890 – 1990. Class sizes shrank; teacher salaries grew; and
expenditures beyond the classroom blossomed. Picus and Fazal (1996) examined school
expenditure data across the U.S. between the 1960s and 1990s. Similarly, they found that
between 1960 and 1992, real per pupil spending increased by 207%. However, they also
noted a wide range from a 100% increase in Utah to a 411% increase in New Jersey.
Much of the post-1980 increases may be attributed to inclusion of special-needs and
handicapped students in the full academic program of public schools.
At the same time that real expenditures have grown, several resource use patterns
remain entrenched. Odden and Picus (2008) identify six universal K-12 education
resource expenditure functions: (a) instruction, (b) instructional support,
(c) administration, (d) student support, (e) operations and maintenance, and
(f) transportation and food services.
Nationally, expenditure patterns indicate that approximately 60% of a district’s
budget will be directed toward the instruction function each year (Odden & Picus, 2008;
Picus & Fazal, 1996). Expenditures in this area include those for teacher, aide, and
instructional coach salaries and benefits, instructional materials, and purchased services.
Five percent of the budget is allocated to instructional support including professional
development, curriculum development, libraries, and media support. Another 10% of the
33
budget supports site (5%) and district level (5%) administrators as well as their support
staff and supplies. Programs and staff members dedicated to the well being of students
including guidance counselors, social workers, health and attendance workers, family
liaisons, and speech pathology services comprise 5% of the district budget and are
included in the student support function. Ten percent of the budget supports staff,
materials, supplies, and services necessary to clean, maintain, cool, and heat school
facilities. The final 10% funds salaries, services, physical and other resources requisite to
transporting students to and from school, and nourishing them during the school day.
School finance research in the 1990s repeatedly concluded with findings that
roughly 60% of school revenues are used to support the instruction function (Monk,
Roellke, & Brent, 1996; Nakib, 1995; Picus, Tetreault & Murphy, 1996). Using national
expenditure data, Picus (1994) and Odden, Monk, Nakib, and Picus (1995) additionally
noted the consistency across these broad categories. The persistence of this resource
allocation pattern even as real expenditures increased over time led a number of
researchers to consider what might be taking place within each expenditure category.
What’s happening within the instruction function? Significant changes appear to
have occurred within the instruction function. Rothstein and Miles (1995) conducted an
analysis of expenditures in nine U.S. school districts to determine where increased
revenues went and to consider why schools did not appear to be achieving great academic
gains in spite of continued infusion of additional funds. They found that “the share of
expenditures going to regular education dropped from 80% to 59% between 1967 and
1991, while the share going to special education climbed from 4% to 17%” (p. 1). Their
34
analysis indicated that per pupil expenditure growth for the regular education program
amounted to a modest 1% annually during the time period studied. Lankford and
Wyckoff (1995) corroborated Rothstein and Miles’ findings, indicating that between
1980 and 1992, the percentage of funding directed toward regular education teachers
declined while that for special education teachers increased. Lankford and Wyckoff
additionally noted that as New York entered a phase of declining enrollment and fiscal
pressure, special education mandates as per Public Law 94-142 appeared to place
additional pressure on the regular education budget.
According to research conducted by Reynolds and Wolfe (1995), 4.7 million U.S.
school children received special education services in 1993 at a cost of $32 billion. In
their longitudinal study of Chicago schools, Reynolds and Wolfe concluded that special
education placement resulted in lower student achievement (as did grade retention). In
particular, the findings suggested, “special education placements might be best targeted
to the younger grades, especially Grades 3 and 4, and on those with disabilities other than
learning disabilities” (p. 258). This interpretation contradicts findings reflecting common
practice of placing increased numbers of students in special education classes as they
progress through the elementary grades. Odden and Picus (2008) share a similar
conclusion: Special education services as presently enacted produce at best minimal and
short-lived gains in student achievement.
Prior to recent increases in special education and specialist teachers, mid-1960s
changes included the appearance of elective teachers who provided instruction in the arts,
physical education, and other non-core courses. These faculty provided core teachers
35
with release time during the school day for planning and preparation. Elective course
offerings have grown to the point that at the secondary level, non-core courses occupy
approximately half of the course offerings (Odden & Picus, 2008).
How the education dollar is spent. Odden and Picus (2008) divide the
expenditure dollar into three portions. Teachers and administrators are described as the
first portion of the expenditure dollar, with teachers essential to delivery of instruction.
This first portion represents the quantity of instructional resources determined to be
necessary to educate children. Everything intended to improve instruction and student
achievement, from professional development to counseling to special needs and electives
classes encompasses the middle portion of the expenditure dollar. Most of this middle
portion coexists with the first portion in the instruction expenditure function. The middle
portion reflects a focus on quality of teachers and services as measured by student
outcomes. The final portion consists of nuts and bolts expenditures to operate and
maintain school facilities and to feed and transport students.
An assumption that resources make a difference in student achievement drove the
Serrano v. Priest decisions in the early 1970s. This premise has resulted in questions and
research regarding effective and appropriate resource allocation.
Resource Use at the School Level – A Micro View
National data paint a picture with broad brush strokes hinting at the importance of
finding out what might be taking place at the school level. Since the 1966 Coleman
Report, education researchers have toiled to prove that schools indeed have the potential
to overcome poverty and family/community circumstances that seem to overwhelm
36
equitable distribution of funds and efforts by educators to develop necessary human
capital within their students (Hanushek, 1998; Kahlenberg, 2001). Meta-analysis of 35
years of school-level research by Marzano (2003) offers confirmation that schools do
affect student achievement and can overcome student background factors historically
related to lower levels of academic success. A number of researchers (e.g., Odden,
Archibald, Fermanich, & Gross, 2003; Mangan, 2007) have targeted school-level
expenditures in attempts to understand how schools distribute instructional resources,
determine the effectiveness of resource use, and influence state and national policy
regarding precise coding of school-level expenditures.
Resource Use for Instructional Improvement
Instructional improvement literature repeatedly includes calls for reorganization
of existing resources to achieve significant student achievement gains. Miles (2000)
identified research in a number of areas focused on educational resources, concluding
that school resources need to be reorganized to support academic achievement. High
performing schools reorganized resources in six areas supported by research: (a) teacher
collaboration time, (b) effective professional development, (c) student academic course
time, (d) salaries to attract and retain effective teachers, (e) autonomy in use and
organization of resources, and (f) securing new funding sources. Class size reduction as a
reorganization strategy and optimal school size also feature in the literature.
About time for teachers and students. Bodilly and Berends (1999), cited in Miles
(2000) support common collaboration time for teachers. Their findings suggest a
minimum of three hours weekly of co-laboring on significant professional development
37
and curriculum-related tasks. The National Education Commission on Time and Learning
prepared a seminal document, Prisoners of Time in 1994. Just over one decade later,
following enactment of No Child Left Behind (NCLB) and implementation of standards
across the country, the Education Commission of the States reprinted the document
(Prisoners of Time, reprinted 2005). The authors highlight the significance of addressing
old-fashioned assumptions and attitudes that limit time to fixed variable status when the
current absolute is student achievement.
In their introduction to the reprinted document, Milton Goldberg, former
executive director of the National Commission on Excellence in Education and former
director of the Department of Education Office of Research, and Christopher Cross,
former Assistant Secretary for Education Research and Improvement in the U.S.
Department of Education and former Republican staff director of the House Education
and Labor Committee, remind readers that the school day and year have remained
unchanged for over 100 years. Yet, students live in a global, digital world unlike anything
in the past. Some key findings pointing to the influence of the clock and calendar include
(a) salary increases based on longevity, (b) nine-month, 180-day school year, beginning
in late summer and ending in late spring, (c) secondary school graduation units tied to
seat time referred to as “Carnegie Units,” (d) a six-period secondary school day averaging
5.6 hours, with approximately 51 minutes per class at the despite level of difficulty or
academic significance, and (e) 41% of secondary school time spent on core academic
subjects.
38
Not only does the report recommend increasing time spent on academic tasks as
this results in improved achievement, but it also recommends increasing flexibility with
time by utilizing technology to extend the learning experience.
Haycock (2001) and Lake, Hill, O’Toole, and Celio (1999) indicated that more
time (as much as two and three times the regular amount) spent on literacy and
mathematics resulted in dramatic student achievement gains in these areas. Class
periods, a common feature of secondary schools, create arbitrary divisions in the school
day stifling opportunities for project-based learning and opportunities for engagement
intended for making meaning.
Class size, does it matter? Research into the effects of smaller class size continues
as the notion that smaller is better fails to be unequivocally determined. Findings from
the Tennessee STAR study of class size reduction in the lower grades indicate that class
sizes of 13-17 students in grades K-3 have both immediate and lasting effects on student
achievement, especially among African American students (Mosteller, 1995). In his
examination of the findings, Hanushek (1999) determined that the greatest influence
appeared to take place in grades K-1. His analysis of special education and other factors
led to a conclusion that it is difficult to determine what the actual benefits, if any, of class
size reduction for students beyond the first grade might be. Both Mosteller and
Hanushek indicated that instructional practice must change to leverage smaller class size
for improved student performance.
More recent research by Rivkin, Hanushek, and Kain (2005) suggests that fourth
and fifth grade students perform at statistically significant higher levels in mathematics
39
and reading when in smaller classes but that middle-grades students do not. In their cost-
benefit analysis of smaller class sizes for upper elementary and middle grades students,
Rivkin et al. explained that “a very costly ten student reduction in class size such as that
undertaken in some U.S. states produces smaller benefits than a one standard deviation
improvement in teacher quality” (p. 419). Their recommendations emphasized teacher
quality over class size. However, Hanushek (1998) is quoted by the American Federation
of Teachers, stating, “’There are likely to be situations—defined in terms of specific
teachers, specific groups of students, and specific subject matters—when small classes
could be very beneficial for student achievement’" (Benefits of Small Class Size, ¶ 4,
n.d.).
Odden et al. (2008) reviewed multiple school resource use indicators within
schools implementing comprehensive school reform programs. They noted that class
sizes ranged from 17 students to 26 students with the slightly higher numbers in grades
four and five. The instructional improvement strategy (e.g., Accelerated Schools Project,
Success for All) did not appear to be associated with class size. Additional class size
research in Connecticut by Hoxby (2000) recommends that teachers receive training and
incentives to best utilize smaller class sizes for student achievement gains. Among
naturally small classes (e.g., during years when enrollment within a given grade was
naturally lower resulting in smaller class sizes) Hoxby found no statistically significant
performance gains for any student demographic groups.
Smaller class sizes intuitively appear valuable because teachers and students can
develop stronger relationships and have fewer distractions, thus have fewer discipline
40
problems and more personalized attention (American Federation of Teachers, n.d.).
Research findings suggest potential benefits from class sizes when the circumstances are
right.
The right teacher and the right instruction. Aligning allocation of teachers with
academic achievement expectations, ensuring that students receive instruction from
certified teachers, and emphasizing teacher quality summarize research findings, which
recognize that teachers indeed influence student achievement.
Miles and Darling-Hammond (1998) and Miles (2000) indicate that nearly 50% of
school resources are allocated outside the regular education classroom. This is supported
by the previously mentioned report, Prisoners of Time, indicating that in 42 states
studied, 41% of a secondary student’s day includes instruction in core academic courses
(Education Commission of the States, 2005). In both cases, researchers recommend a
focus on academic instruction. In particular, Miles and Darling-Hammond recommend
reducing specialists to increase regular classroom teachers. Both Miles (2000) and the
Education Commission of the States (2005) advocate increasing the number of academic
teachers and increasing the time spent on academic instruction. For schools that have
chosen to augment teachers with instructional aides, Mosteller’s (1995) analysis of the
Tennessee STAR study indicates that a teacher paired with an aide in a regular-size class
was related to student achievement gains about one-third the effect size of those
associated with teachers alone in smaller classes. Adding instructional aides does not
result in the magnitude of achievement gains associated with smaller class sizes.
41
Rivkin, Hanushek, and Kain (2005) connect teacher quality to student
achievement. Their summary indicates, “a low income student with five successive years
of high-quality teaching scores at the same level as an average-income student with five
successive years of average-quality teaching” (in Rose, Sonstelie, Reinhard, & Heng,
2003, p. 91). Rivkin, Hanushek, and Kain’s Texas research included findings that “a one
standard deviation increase in average teacher quality for a grade raises average student
achievement in the grade by at least 0.11 standard deviations…in mathematics and
0.095…in reading” (p. 434). These findings concur with Krueger (2003) whose research
concludes that “resources are systematically related to student achievement” (F34) in
terms of standardized test results and adult earning capacity.
Although teacher salary schedules reward longevity and incentivize accumulation
of easy- and cheap-to-acquire continuing education units, Rosenholtz and Smylie (1984)
and Hanushek (1997) have found that experience alone after the first five years does not
appear to improve one’s abilities as a teacher. Even obtaining an advanced degree does
not translate to increased effectiveness as a teacher, yet yields a higher salary. Hanushek
encourages use of incentives including “compensation, career advancement, job status, or
general recognition” (p. 305) for teachers who produce large student achievement gains
as opposed to rewarding such unproven “inputs” as degrees obtained. Elmore (2002)
concurs and elaborates, adding that carefully considered incentives that recognize both
group and individual success in improving student academic achievement must be
balanced with adequate support.
42
Odden et al. (2008) state four reasons why resource use has moved to the top of
the school finance agenda. Each reason and findings related to it point to the importance
of high-quality teachers in every classroom: (a) “modest improvements in student results
despite significant increases in real dollars over the past decades” (p. 382), (b) national
reform expectations for dramatically increased student achievement gains without
matching revenue increases, (c) changes in the definition of fairness from equity to
adequacy along with “adequacy methodologies identifying programs and strategies that
can boost student performance and claiming that their total costs would constitute
adequate funding” (p. 382), and (d) increasingly detailed data sources accompanied by
sophisticated analysis of the “links between education dollar uses and student
performance gains” (p. 382). They additionally identified instructional resource use in
order of expenditure rank. From highest to lowest expenditures, the list includes core
teachers, extra-help strategies (e.g., teacher tutors), specialist teachers, and professional
development. In their examination of schools implementing nationally endorsed school
reform programs (e.g., Success for All) built upon research-based connections between
resource use and student achievement, one resource use area seems to have gained special
traction: professional development to increase teacher effectiveness.
Summary. Class size, instructional time and focus, and high-quality teachers all
factor into an education that may be defined as adequate. Committing to these
instructional improvement strategies entails changes in a number of current practices
ranging from length of the school year to the way that teachers move up the salary
schedule. When the focus shifts to what students need in order to be academically
43
successful, resource allocations shift as well. The section that follows provides a
definition of the educational adequacy educators work to achieve.
Educational Adequacy
School finance adequacy is defined as “the provision of adequate resources to
enable all children to meet a state’s proficiency standards” (Picus, 2004, p. 1). Current
provisions of the No Child Left Behind Act additionally state that “all students in every
school district meet ‘academically rigorous content standards and performance standards
in all major subject areas’ by the 2013-2014 school year” (Chambers, Levin, &
DeLancey, 2006, p. 15). Throughout the past 100 years states developed funding
programs with the intent of guaranteeing a minimum level of funding that they could
claim supported a basic education. This notion of a basic education has changed in the era
of standards and school and district accountability for student achievement toward
performance targets. Leading scholars in this field point to the logic connecting
adequacy and accountability. If schools and districts are to be held accountable for
student performance outcomes, they must receive resources commensurate with
accountability expectations (Elmore, 2002; Schrag, 2005).
This section of the literature review outlines several features of educational
adequacy in the accountability-based reform environment: (a) the transition from equity
to adequacy, (b) models for determining adequate funding to support achievement
expectations, (c) school improvement research informing how resources might be most
effectively utilized and distributed, and (d) challenges to connecting resources to student
achievement gains.
44
From Equity to Adequacy
Equity. The recent history of school finance reform reflects a belief that resources
make a difference in students’ educational experiences. California led the nation in the
early 1970s with litigation to equalize funding across school districts in the state. The
inequity of the school funding system was illustrated by comparing the property tax
revenues generated by a one mil tax rate in two communities in the greater Los Angeles
area. This tax rate applied to Baldwin Hills generated $176 per student whereas the same
tax rate applied to Beverly Hills generated $1,340 per student (Merrow, 2003). The
California Supreme Court declared the school financing system unconstitutional on equal
protection grounds and the legislature responded with AB 65, increasing state support for
low property wealth districts and placing a cap on revenues for high property wealth
districts. Policy Analysis for California Education co-director, Michael Kirst, described
the consequences of the Serrano decision as a rejection of adequacy and an acceptance of
mediocrity under the guise of equal and limited distribution of funds (Merrow, 2003).
A Nation at Risk, published in 1983, painted a dismal portrait of American
education. Secretary of Education T.H. Bell inaugurated the National Commission on
Excellence in Education in August 1981. The commission’s task was to report on the
quality of education in the United States and reported findings in four key areas: (a)
content, (b) expectations, (c) time, and (d) teaching. Each area was found sorely lacking.
Regarding content, the commission concluded, “curricula have been homogenized,
diluted, and diffused to the point that they no longer have a central purpose” (U.S.
Department of Education, 1983). The commission’s “Findings Regarding Expectations,”
45
reported the low selectivity levels of colleges as well as the wealth of non-academic
courses applicable to high school graduation requirements. In its “Findings Regarding
Time,” the commission indicted American education for its short school day and school
year when compared with other industrialized countries, poor classroom management
resulting in reduced instructional time, and ineffective use of class time resulting in
students graduating from high school with minimal independent study skills. In
enumerating their “Findings Regarding Teaching,” commission members found that
academically able college students were not attracted to teaching, education coursework
minimized subject-matter preparation, shortages of qualified teachers existed across
multiple content and special-needs areas, and an unacceptable professional working life
confronted teachers. Thompson and Crampton (2002) suggest that this treatise catalyzed
the shift from equity to adequacy.
The standards and measurement movement. Six years after A Nation at Risk, the
standards movement gained momentum with the National Education Summit in 1989.
Governors from all 50 states crafted six national goals for education in response to the
findings and recommendations of A Nation at Risk and committed their states to the
development of measurable state academic achievement standards (Phelan, n.d.). By
1999, California passed the Public Schools Accountability Act, which mandated academic
standards and associated proficiency exams for each core content area. Passage of the No
Child Left Behind Act in 2001 clearly articulated national expectations for 21st century
proficiency. With the exception of Iowa, each state today boasts standards and associated
minimum proficiency levels. Even the lone holdout is considering adopting standards
46
due to concerns about adequate preparation of students in terms of academics and job-
related skills (Manzo, 2008). Standards and measurable proficiency levels now provide
detailed definitions of academic adequacy.
New tests of state constitutions. While the constitutionality of states’ school
finance plans had been tested repeatedly since the 1970s, the focus of cases shifted from
equity to adequacy in the late 1980s. Picus (2004) describes the 1989 Supreme Court of
Kentucky decisions in Rose v. Council for Better Education. What began as an equity
suit to redistribute resources more equitably was influenced by concerns regarding
educational deficiency. The state supreme court declared the school funding system
unconstitutional and “ruled that a child’s right to an adequate education was fundamental
under the state constitution” (p. 3). Since the 1989 Supreme Court of Kentucky decision,
26 additional finance adequacy cases have reached the courts with a total of 20 (including
Kentucky) being decided in the favor of the plaintiffs.
Adequacy matures. Rebell (2008), co-lead counsel for the plaintiffs in the
Campaign for Fiscal Equity challenge to the constitutionality of New York’s school
funding plan in 1993, described the adequacy movement’s increasing traction over the
course of the past 20 years. He identified significant features of adequacy as a funding
and educational goal. First, Rebell highlighted the silver lining behind the June 2007
U.S. Supreme Court decision that race may not be used as a factor in affirmative action
plans to undo the effects of concentrated poverty. This opens a window of opportunity to
shift the focus from race to adequacy. Adequacy emphasizes providing every child with
what he or she needs in order to meet performance expectations wherever that child
47
attends school (Rebell, 2008; Schrag, 2005). Second, the standards movement provides
plaintiffs with “documented content, set by the states themselves, as a judicially
manageable standard” (Rebell, p. 435). This opens the door to credibly estimating the
costs to achieve these educational goals. Third, courts have created a consensus
regarding adequacy in education. Rebell explains that an adequate education enables
students to function as citizens and engage the economy. It prepares students with
reading, writing, and speaking skills in English, along with fundamental understandings
of mathematics, science, geography, history, economics, and political systems, as well as
sufficient social and communication skills. The courts recognize the connection between
resources and the skills, dispositions, and understandings related to their concept of
adequacy.
Connecting achievement to resources. Defining adequate outcomes initiated
conversation and research, which has focused on how those outcomes might be attained
most effectively. Efforts in this direction began after publication of the Coleman Report
in 1966. Input-output analysis described by Hanushek (1979) served as the method for
measuring effects of educational inputs on student achievement outcomes. Hanushek is
reluctant to refer to a production function in education as this term describes optimal
efficiency of firms in terms “the maximum output feasible with different sets of inputs”
(p. 353). Educational imperatives specify the output targets and Hanushek argues that
much could be done in the form of incentives to promote greater efficiency in attaining
those achievement outputs through more effective instructional inputs. Regardless of the
selected term, “production function” and input-output analysis research reflect ongoing
48
examination of systematic connections between resources as inputs and student
performance as outputs.
Resource-achievement connections continue to be determined and resource costs
are now more precisely coded and easily identifiable. Linking costs to achievement has
become an increasingly viable research, policy, and planning topic (Odden et al., 2008).
Costing Out Adequacy
The need and demand for “costing out” studies reflects the evolution of the
standards and accountability movement along with its critical components: school
improvement and productivity, resource allocation, and appropriate funding levels.
Literature in this area reflects linkages between resources and student achievement, and
describes development of a model to quantify school-level resource use. This model
formed the foundation for subsequent quantification of elements and costs associated
with professional development. It reflects the efforts to develop protocols for
determining the costs associated with an adequate education for all but the most severely
handicapped children whose conditions preclude them from general student performance
expectations.
Resources Matter
Resource quality, quantity, and use. Pan, Rudo, Schneider, and Smith-Hansen
(2003) examined resource use in education and concluded that both quantity and quality
of resources influence student achievement. Their recommendations for state and district
leaders and policymakers included alignment of resource allocation with the
improvement process. Darling-Hammond and Friedlaender (2008) identified a number
49
of instructional improvement design features and made federal and state policy
recommendations to provide adequate resources to meet the needs of students.
Recommendations include professional development based on standards outlined
by the National Board for Professional Teaching Standards: seven to fifteen days of
shared professional learning time per year plus several hours weekly for collaborative
planning, analysis, evaluation of student work, and curriculum design, smaller classes
with an average of 25 students that foster “continuous, long-term relationships between
adults and students” (p. 16), rigorous and relevant instruction, and a focus on
personalization as opposed to the 20
th
century factory model of schooling, which
emphasizes processing students. Elmore (2002) highlights the significance of tying
professional development to collaboration around school-specific improvement efforts
and to measuring the effects of the professional development.
A strong and growing school improvement research base with recommendations
for small classes, teacher collaboration and professional development time, and high
quality teachers (e.g., Hanushek, 1998; Miles, 2000) implies the possibility of developing
a complete list of essential resources and ultimately their costs. Additionally, when
considering educational productivity, researchers (Odden, Monk, Nakib, & Picus, 1995)
explored system restructuring possibilities to maximize the use of education dollars when
applied to student achievement goals.
Accurately tracking expenditures. In 1995 it was difficult to accurately compare
categories of expenditures due to idiosyncratic definitions of expenditure functions
(Odden, et al.). Odden, Archibald, Fermanich, and Gross (2003) subsequently developed
50
a school resource-use tracking model that permits analysis of expenditures within and
across schools, and of elements within the previously mysterious instruction function.
Elements of this model reflect the growth in the late 1990s and early 2000s of
comprehensive school reform designs and research supporting smaller classes, smaller
schools, learning communities, professional development, and instructional support. The
expenditure tracking model developed by Odden et al. supports comparison of actual
school resource allocation across schools as well as scientifically based expenditure
possibilities. Detailed below, this general expenditure model clarifies current school-
level resource allocation patterns.
The Expenditure Tracking Model
The need to track expenditures. Beginning in the mid-1990s the need to align
school finance with instructional improvement efforts at the school level prompted efforts
to develop a valid and reliable expenditure tracking protocol. In explaining the genesis of
their expenditure-tracking model, Odden et al. (2003) highlighted a number of rationales
(e.g., qualitative and quantitative differences in resource use noted in schools
implementing comprehensive school reform programs). They also pointed to a rich
research base in areas including class size, school size, professional development, length
of core class periods, use of instructional aides, and learning communities associated with
improvement in student achievement. Goertz and Stiefel’s (1998) findings indicated that
such a model would facilitate equity analyses across schools. Lee and Smith (1997) found
unique and successful strategies in smaller high schools. As a result, these schools “used
education dollars quite differently than the ‘typical school’” (Odden et al., p. 326).
51
Additionally, whole school designs featured more regular classroom teachers, more
teacher tutors, greater professional development expenditures, and lower expenditures on
resource rooms and instructional aides. It became increasingly clear that analysts needed
“a reporting system that indicated how schools used dollars, so they could be used better
in the future” (p. 325).
Elements of the model. Odden et al. set about creating a school-level fiscal
reporting structure that would report school-level resource use that “simultaneously
indicates as much as possible about the educational strategy those resource-use practices
reflect” (2003, p. 327). The resulting structure identifies school-level expenditures and
“categorizes expenditures by expenditure elements that reflect current thinking about
effective instructional strategies and resource deployment” (p. 327). The expenditure
framework components were culled from high-performing schools research as well as
such typically referenced school finance expenditure elements as instruction, staffing,
administration, and operations. The framework additionally includes empirically based
organizational features.
Specific school features along with their research bases support the expenditure
framework: (a) school size, based on research findings by Andrews, Duncombe, and
Yinger (2002), Lee & Smith (1997), and Raywid (1997/1998), (b) class size, based on
findings from Grissmer (1999), (c) use of aides, based on evidence from Gerber, Finn,
Achilles, and Boyd-Zaharias (2001), and (d) learning communities, based on research
results from Slavin and Madden (2001), and Stringfield, Ross, and Smith (1996), all
associated with high-performing and improving schools. Odden et al. also include
52
professional development within the expenditure framework, citing research conducted
by Garet, Porter, Desimone, Birman, and Yoon (2001) as well as that of Cohen and Hill
(2001) confirming the influence of effective professional development as an instructional
improvement strategy.
The resulting framework consists of nine expenditure elements “that reflect the
core components of nearly all schoolwide educational strategies” (Odden et al., 2003, p.
331). The first seven elements emphasize instruction with the final two addressing
administration and school operations. The nine elements are outlined in Table 2.1.
As outlined, the expenditure model developed by Odden et al. serves as the
research-based and practical underpinning for subsequent models developed to guide
resource allocation decisions at the state and local levels. Its indicators “provide a mix of
quantitative and qualitative data that offer context and insight into a school’s instructional
strategies” unavailable through purely financial data (2003, p. 336). This type of focus
on school resources supported subsequent development of models that attach costs to
resources with the goal of providing an authentic rationale for adequate school funding.
53
Table 2.1. Key School-wide Expenditure Elements Comprising the School-Level
Expenditure Structure
Expenditure Element Brief Description
Instruction
1. Core Academic Teachers
Regular classroom teachers at the elementary level. Core academic
course teachers (i.e., English, mathematics, science, history-social
science) at the secondary level.
2. Specialist and Elective
Teachers
Provide core academic teachers with planning time: arts,
physical/vocational/drivers education teachers, as well as librarians
and media specialists.
3. Extra Help Credentialed teachers who assist students when struggling with
academic course material and/or students with special needs.
Purpose: assist students with learning the school’s regular curriculum.
Includes tutors, extra-help labs, resource rooms, inclusion teachers,
English language development teachers, and self-contained special
education teachers (who teach a modified version of the regular
program).
4. Professional Development Includes teacher time, trainers, coaches, administration, materials,
facilities, equipment, travel and transportation, tuition and conference
fees.
5. Other Non-Classroom
Instructional Staff
Staff members who support regular classroom teachers: program
coordinators (e.g., technology coordinators), substitute teachers, and
instructional aides (excluding self-contained special education
classrooms).
6. Instructional Materials and
Equipment
Includes books, technology, and other materials necessary to
implement the instructional program.
7. Student Support Support student well being. Includes school counselors, social
workers, nurses, and parent outreach liaisons.
Non-Instruction
8. Administration Site administrators (i.e., principal, vice-principal, assistant principals),
clerical staff, office supplies, equipment and technology, and school
reserve funds.
9. Operations and Maintenance Staff members, materials, equipment, and services to clean, heat, cool,
repair, and maintain school facilities.
School Finance Adequacy Models
Hanushek (2006), Picus and Blair (2004), and Odden and Picus (2008) describe
four protocols for determining the level of funding necessary to support resource
recommendations. Each model derives its recommendations from a different set of
54
inputs. Adequacy analysts recommend using more than one when conducting a state or
district cost study. While each model results in a different bottom line, each almost
universally results in a recommendation for increased spending. The four leading models
are described below.
Professional Judgment Panel. The Professional Judgment Panel (PJP) approach
to costing out educational adequacy was developed by Chambers and Parrish of the
American Institutes for Research (AIR) in studies conducted in Illinois, 1982 and Alaska,
1984 (AIR, n.d.), and refined by Guthrie et al. (1997) in Wyoming. The American
Institutes for Research describes the PJP approach as a comprehensive study design
consisting of six phases resulting in a description of prototype schools with a report of
associated costs. The six design phases include: (a) public engagement in which
interested parties provide input regarding adequacy goals, (b) design of instructional
programs to meet designated outcomes and learning standards by four to eight panels
representing clusters of districts with unique needs and consisting of experienced
educators and leaders, (c) analysis and synthesis of results, (d) costing out and initial
report preparation in which costs are attached to identified resources and results are
compared with those using a different costing out approach for the same constituents, (e)
meeting of the stakeholder panel including PJP members, noneducators, business leaders,
school board members, parents, policy makers, and government officials with the purpose
of reviewing the report from a noneducation perspective, and (f) final report production
including a cost estimate for adequate education in each district in the state (AIR, n.d.).
55
The 2003-2004 New York Adequacy Study, a joint undertaking between the
American Institutes for Research and Management Analysis and Planning reflects the
most comprehensive recent PJP study. Implemented in at least 14 states, The PJP
approach has been the most-frequently used costing out protocol in recent years (National
Access Network, 2008). This approach utilizes experienced educators. However, critics
such as Hanushek (2006) point to potential educator bias and lack of experience among
panelists with specific programs and populations, calling the PJP approach the
“educators’ wish list model” (p. 291).
Successful School Districts. Beginning with Ohio, at least eight other states have
utilized the Successful Schools approach to costing out an adequate education. This
method entails identifying schools that meet specified criteria, then determining the per-
pupil funding for those schools. In describing this method Odden (2003) commented that
atypical districts are not included in the analysis, resulting in selection of suburban
“districts of average size and relatively homogeneous demographic characteristics, which
generally spend below the state average” (p. 5). Results from the Successful School
Districts approach do not easily transfer to large urban districts serving high percentages
of English learners and students living in poverty. Calling it the “successful students
model” (2006, p. 292), Hanushek explained that this model does nothing to identify what
leads to success. The National Access Network (2008) additionally commented on the
difficulty of agreeing upon a definition for success. Costs vary as do qualifying schools,
making success a continually moving target.
56
Cost Function Studies. Perhaps the most technically complex costing out
approach is the Cost Function Study. This method relies upon regression analysis, using
per pupil expenditure as the dependent variable. Student characteristics, desired
performance levels, and district characteristics are the independent variables (Odden,
2003). The Cost Function approach requires great quantities of highly detailed data,
often making it difficult to accurately implement. Hanushek (2006) criticized the
approach for its inability to trace minimum costs for adequate student achievement or to
identify inefficiencies. Therefore, he suggested a new name for the approach, preferring
to call it the “expenditure function” for schools (p. 292).
Evidence-Based Model. The Evidence-Based Model (EBM) developed by Odden
and Picus “summarizes research and best practices evidence on the major dimensions of
schools that have cost implications – school size, class size, core instruction, specialist
instruction, extra help for struggling students, professional development, administration,
etc. – and identifies for each school in a state a level of ‘adequate’ resources” (Odden,
2004, p. 6). District-level functions included in the previously mentioned school-level
expenditure structure (Odden et al., 2003) combine with evidence-based best practices to
create prototype schools. Expenditure recommendations are then based on local costs to
adequately resource the prototype schools.
In critiquing the EBM, Hanushek (2006) points to the possibility of skewed
results from analysts’ understandings regarding acceptable evidence. In his opinion the
“research base is simply too thin to have any consensus view about what an ‘evidence-
based’ school would look like” (p. 291). Hence, Hanushek recommends a new name,
57
“the consultants’ choice model” (p. 291). However, student achievement gains in states
that have adopted EBM recommendations (e.g., Wisconsin & Washington) reflect the
model’s objective of improving school productivity to the point of doubling student
learning as per state assessments (Odden et al., 2007; Odden, Picus, Goetz, Mangan, &
Fermanich, 2006).
Elements of the EBM, outlined in greater detail in Chapter Three (Methodology)
may be envisioned as a series of nested circles of resources supporting the learning of
students at the center. Figure 2.1 summarizes the model’s staffing elements when
implemented at the middle school level, the focus of this study.
58
Figure 2.1. The Evidence-Based Model of school finance adequacy – middle school
staffing elements.
Professional Development
• 10 Extra Days of
Professional
Development
• Instructional Coaches
• Trainers
Specialized Education
• Gifted
• Career & Technical
• Special Education
Extended Support
• Summer School
• Extended School
Day
• English Learner
Support (1 teacher
per 100 ELs)
• Tutors and Pupil
Support (1 per 100
“at risk” students)
Specialists
• Middle School:
20% of Faculty
Core
• Middle School:
25:1 teacher-
student ratio
59
Reform – Research and Recommendations
Multiple researchers, educators, and political leaders speak to the urgency,
possibilities, and, at present, rarity of education reform (e.g., Hanushek, 2006; Picus,
2000). While new money is a “necessary condition for districts to disproportionally
allocate funds into instruction” (Hannaway, McKay, & Nakib, 2002, p. 68), it alone does
not guarantee what researchers agree upon, that reorganization of instruction is the center
of educational reform (Hannaway & Nakib, 2002; Miles, 2000). A point of agreement
among many experts is that the business and work of education must be conducted
differently if dramatic achievement gains are to be realized (e.g., Elmore, 2002;
Hanushek, 1997; Miles, 2000; Odden, 2000; Odden et al., 2003; Waits et al., 2006).
Research and analysis highlight the need and possible mechanisms for education reform.
Whether the approach is from the federal policy or school-level implementation
perspective, the literature reflects the significance and breadth perceived need for
substantial, coherent change.
An Urgent Call for Action
National problem, national response needed. In spring 2008, the Forum for
Education and Democracy (Forum), a think tank dedicated to considering how every
child might be provided with a world-class education, convened a panel of nationally
recognized education researchers, school system and public officials, and an educational
commentator to respond to findings that describe the state of education in the United
States 25 years after publication of A Nation at Risk. Comparisons of U.S. students to
those in Organization of Economic Co-operation and Development countries reveal that
60
the U.S. is worse off educationally than it was in 1983. The Forum called for an initiative
at the scale of the National Defense Education Act that followed the launch of Sputnik in
1957. At that time the federal government comprehensively guided and supported the
effort.
The Forum report states that the nation’s growing population of children living in
poverty, speaking a primary language other than English, and needing 21
st
century skills
and knowledge calls for an investment of $29 billion beyond current commitments. This
investment will ensure “meeting the federal funding obligation to high-needs students,
providing pre-school and early learning opportunities; developing a world-class teaching
and school leadership force; investing in research, development and innovation; and
supporting community engagement with schools” (p. 47).
A cost-benefit analysis of public school investment includes such items as per-
student costs for each child that graduates from high school following a K-12 education
that prepares the graduate for productive membership in 21
st
century society. The
benefits side of the equation includes additional tax revenue as a result of graduates’
capacities to engage the economy and savings in public health, welfare, and crime-related
expenditures. Estimates from Levin, Belfield, Muennig, and Rouse (2007) indicate that
the cumulative costs of educational excellence amount to $82,000 per student. For every
$1 spent on an adequate education, one that meets the needs of 21
st
century citizens, the
country gains $2.50. Figure 2.2 depicts the Forum’s cost-benefit analysis.
61
$225,000
$200,000
$175,000
$150,000
$125,000
$100,000
$75,000
$50,000
$25,000
$0
Benefits Costs
Figure 2.2. Costs and benefits of an excellent education per high school graduate.
Source: The Forum for Education and Democracy, 2008, Democracy at Risk.
In Democracy at Risk, the Forum for Education and Democracy recommended a
stronger federal support role and identified Finland as a possible model for effectively
balancing centralization and decentralization. Forum panelists concurred that the
undertaking is too immense and too significant to leave states and localities to their own
devices.
$209,000
Additional Tax Revenue:
$139,000
Public Health Savings:
$40,500
Reduced Crime Savings:
$26,500
Welfare Cost Savings:
$3000
$82,000
Benefits are
2.5 times
greater than
the costs.
62
Key Elements of Successful School Reform Throughout the System
In addition to the call for federal action above, other researchers emphasized
actions that could take place at multiple levels within the system. Examples of successful
reform efforts ranging from school to federal levels dot the education landscape.
School-level findings and recommendations. Murnane and Levy (1996) explained
that schools are falling behind because students are not learning the hard and soft skills
associated with personal success in the current economy: math, reading, working in
groups, making presentations, and using personal computers. They recommended school-
level instruction and assessment in these areas as well as ongoing professional
development to support teachers in meeting current educational needs of students.
Reflecting their commitment to prove that demographics does not equal destiny,
Waits et al. (2006) set out to identify features shared by consistently successful and
continually improving schools that serve high populations of poor and English learner
students in Arizona. Using Jim Collins’ Good to Great and the Social Sectors (2005) as
well as his advice, the authors identified 12 beat-the-odds schools as well as 11
comparison schools. In comparison, the beat-the-odds schools stood out in terms of three
success-related categories: (a) disciplined thought, (b) disciplined people, and (c)
disciplined action. Disciplined thought is characterized by a clear focus on the bottom
line: achievement of every child in every classroom. Disciplined people are reflected in
strong and steady principals and effective work teams that take responsibility for student
success. Disciplined action includes commitment to a program with flexibility to be
modified if needed, a build-to-suit mentality characteristic of Starbucks that contrasts
63
with the assembly-line mentality associated with Ford Motors, and an ongoing cycle of
instruction-assessment-intervention.
As a scaffold for school improvement reform, the Arizona findings reflect a
number of elements embedded within the Evidence-Based Model. Findings support the
importance of professional development, the notion of school as a set of interlocking and
synergistic practices and policies, valuing of new evidence regarding best instructional
practices, meaningful collaboration, and effective use of technology.
Chambers, Levin, and DeLancey (2006) applied a professional judgment costing
out approach with the intention of informing development of a future “system of resource
distribution that ensures all districts can provide adequate educational services to their
children” (p. 5). In costing out an adequate education for California students, Chambers
et al. sought transparency with regard to expenditures through use of a systematic process
that permits “an understanding of what combinations of inputs would be necessary to
produce any given set of outcomes” (p. 2). Their resulting recommendations address four
key areas: (a) class size reduction, (b) extension of the school day and year for all
students, (c) specialists to work with small groups of students, and (d) increased
opportunities for professional development. This reform represents an increase of 53% to
71% over existing spending levels in order to conduct the business of education in a
qualitatively different manner.
Comprehensive school reform, described by Odden (2000) includes high-quality
curriculum, effective student groupings and scheduling, teacher planning time, student
support and home outreach, professional development, technology, and implementation
64
of instructional strategies so all students learn. Design components reflecting an
amalgam of locally designed and pre-developed models include administration,
instructional facilitators, teachers, regular education specialists, support for struggling
students, professional development, teacher aides, pupil support/family outreach, and
technology. Enumeration of these design elements guides efforts toward comprehensive,
cohesive instructional programs designed for significant student achievement
improvement and reflects research and calls for reorganization of schools.
Alignment across levels. Darling-Hammond and Friedlaender (2008) emphasize
the significance of school structures and instructional resources among successful high-
poverty urban schools in California. In addition to noting features shared by these
schools, Darling-Hammond and Friedlaender offer state and federal policy
recommendations. First, they recommend a focus on human capital development of
teachers by underwriting high-quality preservice preparation in return for commitments
to teach in high-needs schools, ongoing funding for professional development of
inservice teachers, and funding principal development programs based on change
management, professional development planning, and school redesign. Organizing
schools as smaller learning communities and effectively disseminating information
regarding best practices in school organization and student instruction reflect the tenor of
the second recommendation. Third, the authors recommend a focus on development of
challenging curriculum grounded in “modern conceptions of learning and curriculum,
including interdisciplinary and applied learning that incorporates new technologies” (p.
20). Fourth, in the area of funding, Darling-Hammond and Friedlaender recommend
65
weighted funding formulas with dollars that follow students as well as a streamlined
school funding programs that facilitate comprehensive and cohesive instructional
programs. Professional development, smaller learning units, dissemination of best
practices information, relevant curricula, and responsive funding have the potential to
change the look, feel, coherence, and quality of the public school experience.
How do different levels of government and community interact when reform
efforts are successfully implemented? Results of Dee’s (2002) study of finance reforms
in three states (Massachusetts, Georgia, & Tennessee) indicate that a number of factors
together influence the viability of efforts to bolster instruction for students in the poorest
districts. Highly visible educational initiatives when combined with local control and
parental involvement strengthen reform efforts. Court-ordered reforms are more likely to
result in instruction-related expenditures including teacher training and smaller class
sizes. Lottery funds, bearing the stigma of being considered a regressive tax of the poor,
tend to be directed toward areas outside instruction: capital investments, equipment, and
scholarships for students. Successful reforms stem from aligned legislative, court, and
local efforts toward a high-profile initiative.
In summary, researchers agree that funding commensurate with resources
necessary to achieve student academic performance outcomes is a necessary but
insufficient component of school improvement. How the funds are used becomes the
keystone to both effectiveness and efficiency in providing high-leverage resources
aligned toward a common purpose. While funding increases traditionally have not
altered the balance between support for instruction and that for other expenditure
66
functions, effective resource allocation supports re-visioning how instructional funds
might used with a focus on instructional support that includes capacity building of
educators through professional development to meet the needs of diverse 21
st
century
students
Professional Development
Professional development repeatedly emerges in the literature as a key component
of instructional improvement. Current school improvement research points to the
essential role played by classroom teachers (Cuban, 1993) as well as the challenges of
providing adequate instructional services to 21
st
century students. High-quality
professional development is the means for developing capacities of teachers so they may
be responsive to continually changing student needs. The Evidence-Based Model itself,
the “straw man” for analyzing current school improvement efforts in this study,
recognizes and includes professional development as a critical expenditure area.
Professional development research grounds this section of the literature review.
Findings
Effective professional development – purposes and components. The National
Staff Development Council (2008) cites a singular purpose for professional development
for educators: Improve the learning of all students. NCLB provides criteria for
professional development that results in student achievement gains. Professional
development shall (a) be sustained, intensive, and content focused, (b) align and directly
relate to state academic content standards and assessments, (c) improve and increase
teacher knowledge of subject matter, (d) improve and increase teacher knowledge of
67
scientifically based pedagogy, and (e) be regularly evaluated for influence on teacher
effectiveness and student achievement (Yoon, Duncan, Lee, Scarloss, & Shapley, 2007).
Professional development research findings highlight specific elements associated with
teacher learning that results in better teaching that further results in increased student
learning and achievement (Odden, Archibald, Fermanich, & Gallagher, 2002; U.S.
Department of Education, 1995; Yoon et al., 2007).
A key feature of effective professional development is time. Research findings
indicate that continuous, long-term professional development with substantial hours
annually (Odden et al., 2002) is connected to student achievement gains. Yoon et al.
(2007) found that 14 hours annually of professional development targeting a particular
area resulted in consistent 21-percentile student achievement growth across studies. In
comparing public schools with the success of the Edison Schools model, Hannaway and
Sharkey (2004) noted Edison’s emphasis on professional development. Edison teachers
received 20 days of professional development targeted toward Edison-style classroom
management and the reading and mathematics programs used in Edison Schools: Success
for All and Everyday Math. Garet, Porter, Desimone, Birman, and Yoon’s (2001)
findings indicated that both contact hours and time span influenced professional
development effectiveness. Longer activities included more active learning (e.g.,
opportunity to plan for classroom use), observation of and by participants, coherence with
school goals and standards, skill and pedagogy development, review of student work,
demonstrations, and content knowledge focus. Overall, Garet et al. determined that
68
“professional development is likely to be of higher quality if it is both sustained over time
and involves a substantial number of hours” (p. 933).
Researchers also identified the form of and participants in professional
development as significant features. Hannaway and Sharkey’s (2004) findings that
Edison schools invest in on-site content coaches to augment intensive institutes supported
Odden et al. (2002), Garet et al. (2001), and Elmore’s (2002) conclusions that job-
embedded and school-based professional learning is more likely to result in desired
changes in teacher actions in the classroom. Collective participation by teachers from the
same site, grade level, and subject enhance implementation coherence and support
continued professional communication regarding new learning (Garet et al.).
In addition to time, form, and participants, the focus of professional development
is linked to effectiveness as measured by student achievement. Specifically, focusing on
both content knowledge and how students best learn that content is more likely to result
in changed classroom instruction (Garet et al., 2001; Odden et al., 2002). Bonstingl
(2004) suggests five essential skill areas for teachers to master through professional
development so that they can increase student academic achievement and life success.
These areas include (a) personal management, (b) effective studying, (c) thinking,
reasoning, and problem solving, (d) graphic organization, and (e) information processing.
Additionally, Marzano, Pickering, and Pollock (2001) identify nine instructional
strategies connected to student achievement gains. Mid-Continent Research for
Education and Learning (2008) has developed targeted professional development
modules to prepare teachers to implement these strategies including use of metaphor,
69
graphic organizers, non-linguistic representation, and summarizing to support student
learning. Partnerships between K-12 schools and universities provide a commonly
traveled avenue for developing teacher content knowledge. Content knowledge alone
correlates with reduced change in instructional practice. Garet et al. (2001) report that
after enhancing both content and pedagogical understandings, classroom changes
followed.
Expenditures on professional development. Recognition that professional
development is critical in changing what and how teachers teach to attain student
achievement outcomes has led researchers to explore how and how much is spent on
professional development (Miles, Odden, Fermanich, & Archibald, 2004; Odden et al.,
2002). Current school and district accounting systems do not support accurate
calculations of professional development expenditures. As a result, Odden et al. utilized
the research base regarding effective professional development and developed an
expenditure framework to clarify how money in this area is spent.
Odden et al. (2002) identified six professional development expenditure
categories: (a) teacher time, (b) training and coaching, (c) administration of professional
development, (d) materials and equipment, (e) travel and transportation, and (f) tuition
and conference registration. Use of these categories assists in overcoming accounting
practices that lead to crude estimates, defining professional development, including site
and district expenses in calculating professional development totals, decreasing existing
fragmentation of professional development across departments and topics (Miles &
70
Hornbeck (2000), and offering a framework against which different professional
development programs may be compared.
Because each category is represented by a dollar figure, schools and districts now
can attach professional development features to dollars. Features of effective
professional development (e.g., time, form, etc.) can be connected to dollars, as can the
measured effects of a given professional development program on student achievement.
At present, schools and districts tend to be uncoordinated in their professional
development offerings (Miles et al., 2004). Additionally, almost one-half of professional
development is externally funded with the implication that it may not be connected to
specific school improvement efforts (Miles et al.). Finally, expenditures tend to be up to
80% more than what is actively managed, again with implications regarding
fragmentation and redundancy in professional development offerings.
The possibility for this research-based framework to link dollars to professional
development to student achievement warranted its inclusion in the previously described
Evidence-Based Model developed by Odden and Picus (2008).
An Environment Conducive to Adequacy
Creating the capacities and momentum for systemwide instructional improvement
rely on new models of leadership and new understandings of professional development.
Three overarching questions guide quests toward school improvement: Who are we (and
to what are we committed)? Why are we doing this (i.e., what’s the rationale behind our
choices)? And why are we doing this, this way (i.e., why do we use the strategies we
use)? (Garmston & Wellman, 1999). As educators and leaders become more
71
organizationally aware, they develop the capacity to effectively utilize available
resources. Innovative resource allocation and implementation result in the actions that
move schools toward desired states in terms of student achievement. Leadership and
ongoing professional development serve critical functions in preparing educators to
provide students with an adequate education. Literature in these areas informs both the
research questions for this study and the Evidence-Based Model in which it is grounded.
Leadership
Business leadership models have gained traction in the arena of instructional
leadership as measurable student achievement outcomes have become identified as a
school’s product or output. Leadership distribution, responsibilities, and types described
below reflect the magnitude of difference between managing incremental change and
guiding metamorphosis.
What, why, and how of leadership. Spillane, Halverson, and Diamond (2001)
developed a distributed theory of leadership grounded in distributed cognition and action
theory. They explained that the people, materials, and structures of schools are
intertwined and influence what people do and how they do it. Developing an inquisitive
lens for looking at actions (i.e., what is being done and how it is being done) makes the
espoused and enacted vision, values, and beliefs (i.e., the reasons why) transparent. By
revealing the extent to which the espoused vision is achieved through current actions,
leaders and educators can consider necessary structural and material changes required as
well as professional support needed by teachers to make the vision a reality.
72
Marzano, Waters, and McNulty (2005) offer a succinct rationale for including
leadership in any discussion of instructional improvement in stating, “whether a school
operates effectively or not increases or decreases a student’s chances of academic
success” (p. 3). As they considered the significance of leadership in student achievement,
Marzano et al. reviewed numerous leadership theories and employed meta-analysis to
determine that leadership makes a difference in student achievement. They identified
two levels of change and delineated 21 responsibilities of school leadership applicable to
differing degrees to each type of change. As with Spillane et al. above, Marzano et al.
emphasize distributed leadership. Focusing on practical leadership applications, they list
five steps to provide leadership that matches the order of change: (a) identifying a team
that collectively comprises the 21 responsibilities of leadership, (b) distributing
responsibilities among the team members, (c) selecting high-leverage work,
(d) identifying the magnitude (i.e., order) of change, and (e) matching leadership to the
magnitude of the change.
The two orders of change these authors reference come from the work of Argyris
and Schon in describing single- and double-loop learning (Argyris, 1990). First order
change, associated with single-loop learning is characterized as incremental, fine-tuning,
an obvious next step, and employing strategies that have worked in the past. The change
is not considered dramatic. Second-order change is described as deep change that “alters
the system in fundamental ways, offering a dramatic shift in direction and requiring new
ways of thinking and acting” (Marzano et al., 2005, p. 66). When a problem cannot be
resolved by the strategies in the system’s repertoire, the problem itself requires
73
reconceptualization or new strategies must be created. Second-order change leaders and
leadership are critical when schools move into new territory in terms of student
performance expectations and affiliated mechanisms.
Marzano et al. (2005) identified features of effective schools and explained that
these features indicate a significant departure from previous approaches to educating
students, and therefore reflect second-order change. These features focus on
implementation of a research-based comprehensive school redesign program supported
by high-quality professional development, proven methods for student learning, teaching,
and school management, and meaningful involvement of parents and community. The
headline here is that deep change requires a specific set of leadership components that are
distributed among a leadership team that carries the vision.
Level five leaders. In his blockbuster study of companies that moved from
survival to preeminence, Collins (2001) sought to identify leadership factors associated
with achieving enduring success. He created labels for different leadership levels,
defining the “Level Five” leader as one who creates enduring success and cultivates
leadership among others. Level Five leaders embody personal humility combined with
intense drive to see the corporation succeed. These are highly disciplined individuals
who actively encourage application of creative and intellectual skills along with
accountability for results. Level Five leaders realize that good people make the
difference. In elaborating on the notion of “getting the right people on the bus,” Collins
explained the significance of bringing on board and cultivating individuals with the
commitment, talents, and flexibility to take a journey to new and possibly unexpected
74
frontiers. Level Five leaders seek highly disciplined employees willing to apply their
creative and intellectual skills along with high-stakes accountability for results. They
emphasize disciplined thought and action with all energy focused on specific goals (i.e.,
the Hedgehog concept). Additional features of highly successful corporations and their
leaders include active use of a high council of advisors to energetically debate critical
issues, a view of competition as an asset, and a willingness to acknowledge even the most
dire situation in order to build toward a positive future.
In short, Collins (2001) found that leadership requires sharp foci on people,
systems, accountability, honesty, persistence, creativity, and experimentation tightly
aligned with a belief and vision of being the best in a specific field. Waits et al. (2006)
discovered Collins’ success criteria being used in achieving and improving schools in
Arizona, showing that these success-related practices from the business world indeed
transfer to and inform instructional improvement efforts.
Change leadership. Like the Level Five corporate leader, instructional leaders
move beyond maintaining high standards and catalyze commitment to a compelling
vision. They recognize the increasing need for adaptability necessary to create solutions
for (second order) problems whose solutions require creativity, innovation, and change
(Fullan, 2005). While instructional improvement occurs in the classroom, leadership that
reflects understanding of the change process and the importance of employee capacity
development establishes an environment in which adequacy may be achieved.
Echoing Collins (2001), Fullan (2002, 2005) identified essential leadership
elements necessary for continual and sustained school improvement. The effective leader
75
has a moral purpose and seeks systemic improvement – wanting to change all schools for
the better, not just his or her own. This leader understands change and recognizes the
significance of changing what individuals value to create collective meaning and
commitment to goals. The effective leader focuses on improving relationships and builds
relationships and teams. The effective leader models knowledge sharing and inquiry.
This leader taps into professional learning communities and relationships as essential
social processes that permit information to become useful knowledge. Finally, the
effective leader strives to make coherence from complex initiatives and constraints. This
leader remains focused on student learning while balancing tensions inherent in
remaining focused and being innovative to achieve goals.
Fullan’s (2002) advice to leaders includes placing resources closest to need,
examining real problems that exist in one’s system, cultivating leaders at all levels of the
organization, and enhancing the conditions of teacher work to attract quality teachers. He
further emphasizes a trusting and demanding culture, internal and external networking,
productive conflict, and capacity to engage in complex work (2005) as essential elements
of sustained change.
Summary
Leadership literature highlights the need to distribute leadership throughout the
organization. It emphasizes a focus on common values and goals, and recognizes the
complexity and turmoil inherent in second-order change. This body of literature
underscores the importance of relationships, developing capacity, and accountability for
demanding goals. Providing students with an adequate education requires innovative
76
instructional approaches that may not yet have been considered. It requires continual
professional inquiry and evaluation of programs, curricula, and practice based on student
achievement results. Instructional leadership is essential to maintaining the direction and
morale of the “troops” in educational environments characterized by revolutionary
change and ambiguity.
Impact of Litigation, Legislation and Policy
on Educational Adequacy in California
State-level school finance equity litigation in the 1970s began a chain reaction of
legal, legislative, and policy decisions with tremendous implications for California school
funding. Fiscal federalism provides a theoretical view as well as both political and
economic perspectives for considering the tension inherent in making school resource
allocation decisions. How litigation, legislation, and policy decisions have influenced
school economics during the past 30 years is outlined in the next section.
Reform Litigation
Researchers examining the effects of litigation on California school finance and
education reform have drawn a number of salient conclusions regarding the de facto role
of courts, the nature of the effects of litigation, and litigation-related challenges to
attainment of educational adequacy.
Impact. School finance reform litigation has been found by researchers to be
detrimental to the education system, resulting in state control, centralization, bureaucracy,
curricular rigidity, and reduced voter control (Thompson & Crampton, 2002). In
particular, the Serrano v. Priest II decision of 1976 catalyzed events that moved
77
California from the “cutting edge of the American dream” (Kevin Starr in Merrow, 2004,
p. 4) to schools faced with reductions in force and cutbacks in course offerings,
programs, and school materials.
Following passage of Proposition 13, California’s school funding base decreased
from 5.5% (Merrow, 2004) to 3.3% of personal income earmarked for public education
(Carroll, Krop, Arkes, Morrison, & Flanagan, 2005). Picus (1991) explained that after
passage of Proposition 13, school spending in California fell below the state’s capacity to
support education and trailed behind that of the nation throughout the 1980s with
California’s real growth at 13% contrasted with 30% real growth in spending for public
schools across the United States.
School funding in Proposition 13-era California depends upon the economic
health of the state as opposed to the more stable property tax base (Picus, 1991). Not only
does education continually compete with other state-funded programs for resources, but it
also loses in times of fiscal crisis. Theobald and Picus (1991) explain the influence of
reduced competition for quality services and funding from 1,000 school districts to one
state agent in that it lowered public interest and subsequent funding.
Politicians, courts, and adequacy. Two key tensions emerge in research
conclusions regarding the impact of litigation on school finance reform (Thompson &
Crampton, 2002): (a) tension between liberty and equality, and (b) tension between
representative government (with no motivation to behave morally) and the judicial
system (with the potential for judicial imperialism). Thompson and Crampton further
determined that by default the courts are not neutral. As integral elements of the political
78
system, courts experience pressure from politicians, media, and the public. Legal
decisions result in legislative consequences (Lindseth, 2006; Thompson & Crampton,
2002; Stern, 2006).
Schrag (2005) describes the ongoing dispute between the legislature and the court.
Legislators contest judicial decisions as overstepping the bounds of the court and
infringing on political decisions. Courts are reluctant to mandate adequacy elements or
take action that may make the legislature liable for that which the people of the state
can/will not support. The very fact of school finance litigation reflects a reality that
“fairness in school funding is not securely rooted either in legal precept or legislative
practice” (p. 136).
Fiscal Federalsim and Locus of Decision-making
Where should important decisions be made? Multi-governmental allocation of
educational resources reflects fiscal federalism in action. Based on intergovernmental
grant theory, the central government establishes large-scale goals for the public good and
interacts with more local levels of government to implement actions toward these goals.
NCLB reflects national interest in education as well as implementation decisions
devolved to state and local levels. California’s high level of state control over local
decisions and program implementation is reflected in its categorical programs and their
requisite constraints. These result in relatively rigid mechanisms for achieving school
improvement goals (L. Picus, personal communication, June 21, 2008).
Pointing out the delicacy and inefficiency of politically derived funding schemes,
some policy analysts have suggested that localities may be more effective than central
79
governments in meeting local needs (Oates, 1999). Researchers have noted localities as
being more trustworthy (Weissert & Schram, 1996) and as adding value by contributing
more than their required share to programs initiated at higher levels of government
(Posner & Wrightson, 1996). Additional research indicates that good federal intentions
may be good for politics, but lead to ineffective and economically unsound results. For
example, Crane and Boaz (2005) cited a statistical analysis conducted in 2002 by Nora
Gordon of the University of California, San Diego. Results indicated that federal funds
directed toward poor school districts ultimately displaced the districts’ own funding. As
a result, “poor schools ended up being no further ahead than they had been without the
federal program” (p. 133).
Intersections of state and local responsibility. The American Civil Liberties
Union filed class action litigation in 2000 to remedy situations in which schools failed to
provide students with adequate instructional materials (e.g., textbooks), decent facilities,
and qualified teachers. The resulting 2004 Williams Settlement included a state allocation
of $138 million for instructional materials for schools with California’s lowest
achievement levels and $800 million for critical repairs (California Department of
Education, 2008b). Oakes (n.d.) contributed to and synthesized expert reports informing
legislators with regard to what California students need in terms of adequate materials,
resources, time, and instruction to attain state and federal educational objectives.
Multiple litigation-spurred state statutes reflect the experts’ findings.
Although the case originated with approximately 100 San Francisco County
students, the State and the California Department of Education were named in the suit for
80
their failure to provide students with the most basic of resources. Litigation resulted in
awareness, public interest, and legislative action. Williams resulted in a settlement
concession maintaining limited state responsibility for the details of instruction (Schrag,
2005) and tighter controls on school and district spending. By implication it appears that
devolution of spending decisions to more local levels may not always lead to efficient,
effective, and adequate management of resources. Yet, the state sidestepped the critical
issue of the case, provision of highly qualified teachers, thereby avoiding a giant step
toward state-level adequacy commitments. As a result, possibly inefficient, ineffective,
and inadequate local resource management continues within a context of increased
compliance measures passed down from Sacramento.
As a backdrop to this study, fiscal federalism provides a rationale for understanding
the nature of goal setting, action, implementation, and accountability for school
improvement and student learning in California. Comparison of school resource
allocation with the Evidence-Based Model offers the opportunity to gain understanding
of current influences on resource allocation decisions and the levels of school site
capacity and autonomy in instructional improvement decision-making.
Barriers to and Recommendations for School Improvement Redesign
Policy barriers. While schools must provide basic resources, a variety of legal and
policy barriers block efforts at redesigning California high schools for significant
improvement. Four barriers to school improvement identified by Price (2006) include (a)
lack of autonomy regarding key areas of school control, (b) credentialing requirements
that limit what teachers can teach, (c) tension between standardized assessments and
81
schools’ desires to use performance-based assessments and exams aligned with college
readiness, and (d) oversized, antiquated facilities.
Price (2006) recommends use of charters and weighted student funding to increase
autonomy, experimentation with performance-based assessment, and examination of
newly created pilot small schools facilities in California. Advice to the state includes
flexibility in offering interdisciplinary credentials. Loeb, Bryk, and Hanushek (2007)
described the condition of California school governance as “a hodgepodge of restrictive
rules and regulations that often hinder, rather than promote, student achievement” (p. 2).
They note a lack of alignment between educational objectives and accountability as well
as constraints that limit innovation as significant policy barriers to school improvement.
Without deep change at the state level, including school-level flexibility, teacher and
administrator capacity building, and funding that reflects actual costs to meet needs, site-
level efforts alone will not result in educational adequacy.
Summary
Litigation may increase legislative and public sensitivity (Thompson & Crampton,
2002) and highlight growing needs for increasingly more intense resource applications
(e.g., smaller classes & qualified teachers) (Carrol et al., 2005). However, sensitivity
alone does not lead to finance reform. Researchers indicate that awareness,
embarrassment, pressure, and a sound economy all contribute to financial reforms that
stick. Increased state control over education funding has not improved student learning.
Yet, local failure to provide such basic educational tools as textbooks has resulted in even
more state control. Thompson and Crampton concluded that the state’s budgetary and
82
political entities might be incapable of committing to both equality and excellence at the
same time for all children. That said, coming to agreement on what constitutes adequate
funding permits an informed debate over educational adequacy to move forward.
Essentials for School Improvement
Common Features of Successful Schools
The stories of schools and districts deemed successful based on a set of common
academic achievement indicators and on-time graduation rates provide insight into how
improving schools prioritize and allocate resources. Successful systemic reform resulted
in implementation of research-based best practices of professional development, extra
help for struggling students, smaller class sizes, teacher collaboration, and extended time
for learning (Fermanich, Mangan, Odden, Picus, Gross, & Rudo, 2006). Case studies
detailing the significant shifts in resource allocation associated with the doubling and
even tripling of student academic achievement results also revealed core themes of
successful school system reform.
Based on Washington Learns findings (Fermanich, et al., 2006), successful school
reform depends on (a) a focus on, sense of responsibility for, and vision of teaching all
students, (b) use of disaggregated student data to make goal-oriented decisions regarding
instruction, resource use, professional development, and curriculum, (c) adoption of a
rigorous, engaging curriculum that aligns to state standards, (d) professional development
that supports student learning-goal attainment by developing educator expertise with
curriculum, strategies, and assessment, (e) restructuring the learning environment to meet
student needs (e.g., time, groupings, class size, & collaborative teacher planning time),
83
(f) extended learning opportunities for struggling students (e.g., pre-kindergarten,
extended day, & extended school year programs), (g) leadership that focuses on
instruction, and (h) professional collaborative communities in which teachers are
respected as experts who share with and learn from each other.
Each successful school and district acted upon the eight core elements in its own
way, innovating to meet student-learning needs. These themes emerged organically as
positive descriptors of successful districts and schools. They inform the post-interview
analysis of the Program Improvement middle school cases examined in this study.
Summary
The purpose of this study is to describe how currently struggling schools are
allocating resources in their efforts to provide students with an adequate education. This
review of literature pertaining to educational adequacy and school reform provides the
historical and research-based background that informs the research questions and case
study analysis.
A rich and growing research base suggests best practices for instructional
improvement. This research base undergirds the Evidence-Based Model, outlining
elements of a prototype school against which real schools may be compared and
analyzed.
California’s litigation, legislation, and policy history illustrates the tightly
controlled environment in which the schools in this study operate while quality leadership
emphasizes adaptivity and innovation. Schools dedicated to teaching all students to state-
and federally determined academic performance objectives create ecosystems of trust,
84
accountability, creativity, collaboration, and discipline. Successful schools reflect a set
of common themes that appear to go hand-in-hand with their instructional reform efforts.
As detailed in this review, the notion of educational adequacy has grown to
maturity. Standards, assessment, and NCLB set a stage for high expectations for all
children. These butt against budget and compliance constraints and educators are
expected to make dramatic changes in their approach to instruction. The methodology
used to respond to these issues and resulting research questions is described in Chapter
Three.
85
CHAPTER THREE
Methodology
Overview
Review of Purposes and Research Questions
This research study was intended to provide a glimpse into resource use within a
purposeful sample of California Title I middle schools. The goal of this new knowledge,
applied research study was to increase understanding of how the right quantity and
quality of resources may be allocated consistently and effectively to students to attain
adequate educational outcomes. Upon conclusion of data collection and analysis, the
author determined (a) what exists in terms of actual site-level resource use, (b) how that
resource allocation and use compares with the Evidence-Based Model (Odden & Picus,
2008), (c) how those resources are used to improve instruction and student performance,
and (d) factors that influence resource allocation. Comparison of sample schools’
resource-use data to the Evidence-based Model offered a critical third point of analysis
indicating the extent to which school-level resource allocation aligns with those
educational practices understood to improve student academic performance.
The question of how schools that receive federal poverty-mitigation funds allocate
their instructional resources is particularly significant in today’s standards-based reform
atmosphere, which includes high-stakes accountability for student academic achievement
and a federal imperative that “underperforming” schools utilize scientifically based
improvement strategies to meet significant performance improvement goals each year
(United States Department of Education, 2003). Additionally, at the time this research
86
was undertaken, California schools were more than six months into a period of budget
shortfall. In early 2008, a steady stream of news releases from the state superintendent of
public instruction bemoaned the “evisceration” of the public education system, just as
efforts toward student achievement gains appeared to be taking hold and as federal
expectations were about to increase substantially (California Department of Education,
2008c; 2008d). At the point of data analysis, state legislators faced a second deadlock
regarding additional cuts and taxes for the 2008-2009 fiscal year.
A strong and growing base of school-improvement research, a commitment to
student achievement as determined by performance on state academic content standards
tests, and recognition that effective education requires effective use of available resources
established the purpose of this study. Hartman, Bolton, and Monk (2001) explain that
exploration of how individual Title I schools use the resources available to them has the
potential to inform the arenas of research, policy, education, and advocacy as they
consider resource allocation in schools. They further explain the chasm separating
resource accounting from expenditure accounting. Data regarding allocation of human
capital, the most costly element of instruction, currently is disconnected from data
regarding allocation of funds. Through this study the investigator sought to understand
the allocation of resources as a first step toward making the connection between
instructional resources and necessary expenditures to create budgets that support schools’
instructional visions.
Additionally, this study was grounded in research pointing to the importance of
high quality teachers (Archibald, 2006) in increasing student academic performance.
87
Quantity and quality of resources count, pointing to the significance of such elements
such as professional development indicated in the Evidence-Based Model.
Four research questions undergirded this study:
1. What are the current instructional improvement strategies at the school level?
2. How are resources used to implement the school’s instructional improvement
strategies? (instructional vision, plan, goals, etc.)
3. How are the actual resource patterns at the school sites aligned with or
different from the resource use strategies that are suggested by the Evidence-
Based Model?
4. How does the availability of resources affect the development and
implementation of the instructional improvement plan?
Overview of Methodology
The author conducted this study under the direction of Professor Lawrence O.
Picus in partial fulfillment of the requirements for the Ed.D. at the University of Southern
California’s Rossier School of Education (RSOE). This study took place concurrently
with seven other school finance adequacy studies conducted as part of a thematic
dissertation project at the RSOE. Each of the parallel dissertations addressed the same
research questions with purposeful samples reflecting various populations of schools
(e.g., elementary, middle, high, high-achieving, Title I, and improving). The author
contributed to modifications to the Evidence-Based Model data collection protocol in
order to maximize collection of data relevant in this state. All researchers participated in
a two-day training in data collection based on the Evidence-Based Model Codebook,
88
Web-based data entry and retrieval, interview techniques, and calibration for inter-rater
reliability among all researchers. Prior to the training, the author, along with three of the
other researchers assisted the lead researcher in responding to items required for
Institutional Review Board (IRB) approval.
In late spring, 2008, the author contacted principals of schools meeting sampling
criteria to solicit their voluntary participation in the study and obtain written statements
of acceptance to participate on their school letterhead as per IRB expectations. Following
IRB approval, the author finalized interview dates and subsequently conducted case study
interviews with eight Title I middle school principals between mid-September and late-
October 2008. Qualitative and quantitative data entry followed each interview and
directly supported case study preparation. Upon completion of all interviews and data
entry, the researcher analyzed the data and prepared the results and findings. Data were
entered into a password-protected, Web-supported database, affording all researchers the
opportunity to identify patterns and themes across schools and studies.
Description of the Research Design
Sample and Population
This study included data collected from eight southern California Title I
Schoolwide Program middle schools. Eligibility for Schoolwide Program funds requires
a concentration of 40% students in poverty at the time of designation (California
Department of Education, 2008f). Title I funds may be used to provide compensatory
educational support to all students attending “schoolwide” Title I schools. This contrasts
with targeted assistance schools in which only economically disadvantaged students may
89
receive additional support with Title I funds. The investigator sought schools with
similarities in terms of accountability pressures (e.g., Title I Program Improvement
sanctions) and socio-economic status of students. Selection of schools from different
districts made it possible to identify similar resource use patterns across school districts
as well as varying management approaches. This selection method supported the
possibility of revealing uses of educational resources unique to specific school districts.
Criteria for determining the sample included school type (i.e., middle school), location,
economic status (with Title I status as a proxy for poverty), and failure to meet NCLB
expectations as indicated by Program Improvement status.
Middle schools provided a point of special interest for exploration as eighth grade
marks the first point at which the current state accountability system measures student
academic performance in all of the four core content areas of reading/language arts,
mathematics, science, and history-social science. A number of Title I middle schools
notably have reduced, or eliminated entirely, science and/or history-social science in
attempts to dodge or exit Program Improvement. Title I middle schools uniquely struggle
with balancing California’s academic growth goals in all tested content areas alongside
federal NCLB proficiencies, which focus more exclusively on reading/language arts and
mathematics. Middle schools are appropriate for examination of how schools under
pressure respond to multiple tensions between state and federal accountability, and short-
and long-term academic needs of students through use of resources.
Determination of the sample began with the state’s population of middle schools.
The Education Data Partnership (2008) reported that 1,289 schools served middle grades
90
students during the 2006-2007 fiscal year. From this universe of California middle
schools, the focus turned to Title I schoolwide program schools, commonly understood to
serve students in poverty. These schools receive funding intended to improve educational
services for children educationally at risk due to living in poverty. NCLB fiscal
provisions permit them to consolidate and blend numerous funding sources to develop
focused schoolwide plans for instructional improvement. Title I schoolwide program
schools, with their relative wealth in funds, make it possible to see how California
schools intentionally connect research-based instructional improvement strategies to
resources to dollars at the levels recommended by the Evidence-Based Model.
Additional information from EdSource (2008b) guided selection of the sample
toward Title I Program Improvement middle schools. During the 2006-2007 school year,
867 of California’s middle schools (67%) received federal Title I program funds. Of
those middle schools, 555 acquired or retained Program Improvement status during the
2006-2007 school year due to repeated failure to meet NCLB performance expectations.
NCLB expectations rose during the 2007-2008 school year; the number of Title I middle
schools decreased to 860, and the number of Program Improvement middle schools rose
to 560.
Probing specifically into resource use at schools with high incentives and
mandated steps to align resources to improve student academic performance facilitated
analysis of how Program Improvement requirements as implemented at various schools
jibe with the Evidence-Based Model for school improvement. In selecting this sample,
the author additionally sought an increased understanding of how resource use reflects a
91
school’s attempts to define and provide an adequate education to its students in the face
of conflicting messages about what constitutes adequacy.
California’s diversity has led to natural regionalization. Examples of the state’s
diversity include the extremes of its largest and smallest school districts. According to
the Fact Book (California Department of Education, 2008e), California’s largest school
district during the 2006-07 school year was Los Angeles Unified with 707,626 students
representing over 11% of the state’s students (p. 25). At the opposite end was Panoche
Elementary School District in San Benito County with four students. One consistent
feature across schools and districts is the county office that serves them. Each county
office is charged with providing oversight to protect the financial solvency of each
district within its purview. The possible trickle-down effect of some form of overlying
consistency across districts suggested limiting the sample to schools within one county.
While this resulted in limiting the generalization of any findings, this geographical
narrowing of the sample provided a counterpoint to samples identified by other
researchers exploring the same topic.
As a member of a group of eight researchers intending to initiate development of
a California school resource database, the author’s sample provided focus in terms of
school type (i.e., middle schools), achievement status (i.e. Program Improvement), and
location (i.e. Orange County, California). The population of schools from which the
sample were drawn is designated by a number of terms: junior high schools, middle
schools, and intermediate schools -- based on grade levels served and conceptual
approaches to educating students at the first level of their secondary education
92
experience. Junior high schools traditionally have served students in grades seven
through nine and have offered a junior version of the high school experience. A move
toward middle schools began approximately 30 years ago (Mizell, 1999) with an
emphasis on student-centered instruction and the social-emotional-academic development
needs of young teens. Today, myriad configurations of schools for middle grades students
include schools with terms such as “intermediate,” “middle,” and “junior” in the name
and various grade configurations for grades five through nine. For convenience, the term
“middle school” reflects homogeneity associated with a sample of schools configured for
grades six through eight and for grades seven and eight. Conceptual approach (e.g.,
developmental approach associated with the middle schools movement) to education was
not included in the selection process.
This study utilized a purposeful, homogeneous sample of Title I Program
Improvement middle schools drawn from a similar schools list in one county in southern
California. The author solicited volunteer participants from a list of Title I Program
Improvement schools in Orange County, California.
Of the 73 Title I schools in Orange County with middle grades students, six were
eliminated as they were either K-8 schools or were schools of choice within their
districts. Of the remaining 67 Title I middle schools, 20 held Program Improvement
status. These schools, distributed across 10 (of 28) school districts within Orange County
represented the pool from which the sample was drawn.
93
Preparation
The author and fellow researchers attended training on June 20 and 21, 2008,
conducted by Michelle Mangan, Ph.D., study manager of legislated statewide school
resource use studies conducted in Arkansas (Odden, Mangan, & Picus, 2006) and
Wyoming (Odden, Picus, Mangan, Goetz, & Aportela, 2008). The training agenda
included interview protocols, familiarization with the Evidence-Based Model Codebook,
data entry and analysis techniques, case study preparation, and inter-researcher
calibration information. Prior to and following the training, the author, along with co-
investigators addressing the same research questions with different samples, initiated
submission of all required information to the University of Southern California
Institutional Review Board (IRB) for its approval. IRB determined that the study did not
require typically mandated forms including a Site Permission Letter signed by the
principal of each participating school or a Study Information and Voluntary Participation
Form. However, each principal received an Invitation to Participate including a
description of the study (Appendix B).
Principals of eligible schools that earlier indicated interest in participating in the
study received an Invitation to Participate (Appendix B). The invitation to participate
explained the purpose of the study: to examine resource use in Program Improvement
schools and compare it with the Evidence-Based Model. It outlined the process including
a two-to-four hour interview and document examination with the principal and/or other
personnel determined by the principal to be knowledgeable about 2008-2009 resource
use. The invitation additionally included a list of documents for the principal to provide
94
on the day of the interview. In addition to the principal, potential interviewees included
Title I coordinators, assistant principals, department leaders, and others as appropriate to
identifying resource allocation at that school.
The interview included examination of public-access resource-use documents
such as master schedule, school budget, school bell schedule and calendars, School
Accountability Report Card, and attendance and enrollment data. A complete list of
possible documents and artifacts to be examined for this study may be reviewed in
Appendix D. No confidential documents or student records were examined. Participants
were informed that no schools would be identified in the study. Student demographic
data, similar schools ranking and achievement report information, as well as general
district data were collected, primarily from the California Department of Education Web
site, in order to provide context while the school served as the unit of analysis.
The letter of invitation included the researcher’s telephone and email contact
information, as well as a response date. The letter indicated that a follow-up call would
be placed to the interested principal approximately 10 days following receipt of the letter
to schedule an interview date in the event that they had not already responded. When
scheduling interview dates, the researcher offered to contact superintendents as requested
by principals if they needed approval in order to participate. No principal requested this
support. The researcher communicated via telephone and email with the participating
principals and their administrative assistants to confirm interview dates. Interviews took
place between mid-September and late-October 2008. Following each interview, the
95
principal received a personal thank you note. With the principal’s verbal permission, the
superintendent also received a copy of a formal letter of appreciation sent in March 2009.
Instrumentation
The Evidence-Based Model as developed in 2003 by Odden and Picus (2008)
identifies school improvement strategies based on successful school reform models.
Using this model, researchers and practitioners may determine the level of expenditures
for each implemented and desired program; aggregate them; and determine a budget in
which the school’s vision for student achievement is manifested in its resource use. The
model is grounded in the heavily researched ideas that high quality teachers and school
leaders are essential to student achievement. Examples of such research include meta-
analyses of school leadership research by Marzano, Waters, and McNulty (2005) and
similar studies of effective educational practices by Marzano (2003). Key foundations to
the Evidence-Based Model include the “How People Learn” framework (Bransford,
Brown, & Cocking, 2000) and current understandings of the skills and knowledge
teachers need to be effective in facilitating student linguistic and cognitive development
and supporting academic achievement of students in the twenty-first century (Darling-
Hammond & Bransford, 2005). The model suggests instructional resources based on
student characteristics including educational vulnerability due to poverty, diagnosed
learning disorder, and English learner status.
Envisioned as a series of concentric circles with classroom instruction of students
at the center, the Evidence-Based Model proposes specific use of resources to maximize
student learning as depicted in Figure 3.1.
96
Figure 3.1. Evidence-Based Model for adequate resource allocation at the middle school
level.
Professional Development
• 10 Extra Days of
Professional Development
• Instructional Coaches
• Trainers
Specialized Education
• Gifted
• Career & Technical
• Special Education
Extended Support
• Summer School
• Extended School Day
• English Learner Support
• (1 teacher per 100 ELs)
• Tutors and Pupil Support
• (1 per 100 “at risk students)
Specialists
• Middle School: 20% of Faculty
Core
• Middle School:
25:1 teacher-
student ratio
97
Components of the Evidence-Based Model for adequate resource allocation are
outlined in the tables below along with the research rationales supporting their use.
Table 3.1. Evidence-Based Adequate Staffing and Resource Components for K-12
Schools – Core Staffing and Specialist Teachers
Resource Ratio Research Rationale
CORE STAFFING
Core Teachers
15:1 K - 3
25:1 4 - 12
Tennessee STAR (K – 3)
Education reformers and national average (4
– 12)
SPECIALISTS
Specialist/
elective/
noncore teachers
Add 20% to teaching staff K-
middle school.
Add 33% to teaching staff at high
school.
Block scheduling for high school
Imbedded professional development time for
core teachers
Source: Odden, A.R. and Picus, L.O. (2008). School Finance: A Policy Perspective (4th
ed.).
Table 3.2. Evidence-Based Adequate Staffing and Resource Components for K-12
Schools – Extended Support
Resource Ratio Research Rationale
EXTENDED
SUPPORT
Tutoring
20 minutes 1:1 for
20th %ile & below
45 minutes 1:3 or
1:5 21st %ile –
“basic”
1:100 Students Living in Poverty
(services available
to all)
Mathes and Fuchs, 1994
Shanahan, 1998
Shanahan and Barr, 1995
Wasik and Slavin, 1993
Cohen, Kulik, and Kulik, 1988
English Language
Development
1:100 English learners Gandara, Rumberger, Maxwell-Jolly,
Callahan, 2003
Extended Day 1:15 eligible students (50% of free
and reduced price lunch count)
Kleiner, Nolin, and Chapman, 2004
Cosden, Morrison, Albanese, and Macias,
2001
Fashola, 1998
98
Table 3.2 continued
Resource Ratio Research Rationale
Summer School (full
day, 8-week, 6 hour
days, 15 students per
class, 4 hours of
academics, 2 hours
of nonacademics)
1:15 (funding initially determined
by 50% of adjusted free and
reduced price lunch recipients,
subject to monitoring for actual
student needs)
Vandell, Pierce, and Dadisman, 2005
Borman, 2001
National Education Commission on Time
and Learning, 1994
Cooper, Nye, Charlton, Lindsay, and
Greathouse, 1996
Alexander and Entwisle, 1996
Roberts, 2000
Cooper, 2000
Borman and Boulay, 2004
Special Education
Emphasis on early
intervention for
reading and autism,
clustering of
severely challenged
students for cost
control, and census
approach for high-
incidence, lower-
cost students
(providing services
at equal rate across
schools)
Elementary School (432 students):
three teachers
Middle School (450 students):
three teachers
High School (600 students): four
teachers
States reimburse 100% of severely
disabled students’ costs
Census funding decisions in multiple states:
Alabama, Arkansas, California,
Massachusetts, Montana, North Dakota,
Pennsylvania, and Vermont
Landry, 1999
Tilly, 2006
Career and
Technical Education
Weight CTE students at additional
0.3 for reduced class size and
$7000 teacher salary increase for
designing applied CTE core
courses
American College Testing Company, 2004
Phelps, 2006
Odden, Picus, Goetz, et al., 2005
Gifted, Talented,
Able, and Ambitious
Students
Add $10 – 25 per student for
additional services
Many researchers in the following areas:
“Discovering hidden talents in low-income
and/or culturally diverse high-ability
learners” (Odden and Picus, 2008)
“Access to curriculum”
“Access to trained teachers”
“Program and policy implications”
National GATE leaders: Elissa Brown,
Joseph Renzulli, and Ann Robinson
Sources: Odden, A.R. and Picus, L.O. (2008). School Finance: A Policy Perspective (4th
ed.) and Tilly (2006).
99
Table 3.3. Evidence-Based Adequate Staffing and Resource Components for K-12
Schools – Intensive Professional Development
Resource Ratio Research Rationale
INTENSIVE
PROFESSIONAL
DEVELOPMENT
Summer and School Year
PD (training, coaching, &
collaboration)
Add 10 days to teaching contract.
Cost for training: $100 x # of
pupils in school.
Embedded coaches/ instructional
facilitators: 2.3 per elementary and
middle school (432 and 450
students, respectively), 3.0 per
high school (600 students) –
described under “Additional
Resources” below)
Rowan, Correnti, and Miller, 2002
Sanders and Horn, 1994
Sanders and Rivers, 1996
Webster, Mendro, Orsak, and
Weerasinghe, 1998
Cohen, Raudenbush, and Ball,
2002
Elmore, 2002
Joyce and Showers, 2002
Miles, Odden, Fermanich, and
Archibald, 2004
Fullan, 2001
Loucks-Horsley, Love, Stiles,
Mundry, and Hewsen, 2003
Supovitz and Turner, 2000
Garet, Birman, Porter, Desimone,
and Herman, 1999
Bransford, Brown, and Cocking,
1999
Source: Odden, A.R. and Picus, L.O. (2008). School Finance: A Policy Perspective (4th
ed.).
Table 3.4. Evidence-Based Adequate Staffing and Resource Components for K-12
Schools – Additional Resources
Resource Ratio Research Rationale
ADDITIONAL
RESOURCES
Substitute Teachers 10 days per teacher Currently listed in adequacy funding
formulas in states that have adopted the
model
Student
Support/Family
Outreach
1 licensed professional per 100
low-income students
And, Middle and High School: 1
guidance counselor per 250
students
Stringfield, Ross, and Smith, 1996
Brabeck, Walsh, and Latta, 2003
Steinberg, 1997
Comer School Development Program
Roots and Wings/Success for all
100
Table 3.4 continued
Resource Ratio Research Rationale
Aides
2:400-500 students – playground
and school supervision
Allocate at-risk resources for
instructional aides trained as
reading tutors.
Achilles, 1999
Gerber, Finn, Achilles, and Boyd-Zaharias,
2001
Miller, 2003
Librarians 1 per school
High School: add library media
technician
Principal and
Instructional Coaches
Elementary (432 students) and
Middle Schools (450 students): 1
principal and 2.3 instructional
coaches
High Schools (600 students): 1
principal and 3 instructional
coaches
Hallinger and Heck, 1996, 2002, 2003
Murphy, 1994
Murphy, Beck, Crawford, Hodges, and
McGaughy, 2001
Heck, Larsen, and Marcoulides, 1990
Halverson, 2003
Louis and Marks, 1998
Newmann and Wehlage, 1995
Louis, Kruse, and Marks, 1996
Louis and Marks, 1998
Louis, Marks, and Kruse, 1996
Spillane, Halverson, and Diamond, 2001
Secretarial Staff Two per Elementary and Middle
School
Three per High School
Clerical and administrative assistance,
telephone, reception, paperwork, etc.
Source: Odden, A.R. and Picus, L.O. (2008). School Finance: A Policy Perspective (4th
ed.).
Table 3.5. Evidence-Based Adequate Staffing and Resource Components for K-12
Schools – Technology, Equipment, and Instructional Materials
Resource Ratio Research Rationale
TECHNOLOGY
AND EQUIPMENT
One computer per three students --
becomes 1:4 ratio because numbers
include administrators, teachers, etc.
$250 per pupil per year
Earle, 2002
Becker, 2000
Lewis, 2002
Valdez et al., 2000
Archer, 2000
Kulik, 1994, 2003
INSTRUCTIONAL
MATERIALS
Textbooks Elementary and Middle School: $120
per student annually
High School: $150 per student
annually
Ravitch, 2004
101
Table 3.5 continued
INSTRUCTIONAL
MATERIALS
Formative
Assessments
$25 per student
Library Resources Elementary and Middle Schools: $20
per student
High Schools: $25 per student
Michie and Holton, 2005
Tenopir, 2003
Source: Odden, A.R. and Picus, L.O. (2008). School Finance: A Policy Perspective (4th
ed.
A Codebook containing interview questions designed to ascertain school-level
resource use based on the Evidence-Based Model provided a foundation for consistency
across interviews. The interview protocol led directly to responses to the research
questions in this study.
Although conducting individual studies, each member of the thematic dissertation
group attended the June 2008 training in the data collection and analysis protocols,
adhered to the same model, was responsible for checking data for entry errors, and
contributed to a collective database of California findings. The collection and analysis
protocols had been used previously in states across the United States, including use in a
randomized study of schools throughout Arkansas in 2006 (Mangan, 2007).
The data collection protocol included both open- and close-ended questions to be
used during the interview with principals and other key staff members. Definitions for
Codebook terms were drawn from Odden, Archibald, Fermanich, and Gallagher’s (2002)
article describing features of effective professional development. Other key sources for
definitions included the Education Data Partnership’s glossary (2008) and the North
102
Central Region Education Laboratory (2008). Schools were defined as institutions
providing educational services to one or more grade groups PK-12 with one or more
teachers, located in one or more buildings or sites, with an assigned administrator,
receiving public funds as a primary support for operations, and operated by an education
agency (Young, 1999).
Section I of the data collection protocol included contact information for schools
and for any staff designated to provide information. This information was not included in
the on-line database to protect participant anonymity. Section II delineated school and
resource indicators including but not limited to the number of students per grade level,
grade span of school, school size, number of students eligible for free-and-reduced-price
lunch, number of students with Individual Education Plans (IEPs), number of students
with disabilities in self-contained classrooms, number of English language learners,
length of instructional day, length of core class periods, number of Full Time Equivalent
(FTE) employees per subject and grade, and organization and allocation of FTEs as per
the Evidence-Based Model.
The Model was revised to meet the California context by including California
Department of Education terms for student ethnicity, special needs, English language
status, and economic status. Recognizing California’s large number of English language
learners and multiple designations based on level of language acquisition, California
English Language Development Test (CELDT) status as well as Reclassification Fluent
English Proficient (RFEP) data assisted in establishing the contextual backdrop for
collecting and analyzing school-level resource use data. Additional contextual elements
103
included state Academic Performance Index (API) and similar schools ranking data for
each participating school.
Data Collection
Framework and Process
This study was grounded in a practical framework. Its purpose was to increase
the understanding of resource use at school sites as a first step toward connecting high-
yield resources with dollars to reveal what it costs to provide all students with an
adequate education as determined by state and federal expectations. The researcher
conducted interviews with principals and other site-level personnel knowledgeable of
allocation of human and material instructional resources at the school. Interviews were
conducted between mid-September and late-October 2008. Immediately prior to each
interview, the researcher provided interviewees with a copy of the Study Information and
Voluntary Participation form. Interviews were held following participants’ review of the
form, response to any questions, and their verbal approval. Interviews took place at the
schools being researched, typically in the principal’s office or conference room, at times
convenient to individual principals. A list of school interview dates appears as
Appendix C.
Data were entered into the Evidence-Based Data Collection Codebook as
interviews were conducted. The researcher took notes on paper to capture principals’
responses to interview questions. Schools and interviewees were coded to protect their
anonymity. Close-ended standardized interview questions for the middle school sample
fell into 13 question categories: (a) School Profile, (b) School Contacts, (c) District
104
Profile, (d) District Contacts, (e) School Resource Indicators, (f) Core Academic
Teachers, (g) Specialist and Elective Teachers, (h) Library Staff, (i) Extra Help Staff, (j)
Other Instructional Staff, (k) Professional Development Staff and Costs, (l) Student
Services Staff, and (m) Administration. Examples of close-ended standardized interview
questions included:
1. Number of Students Eligible for Free- or Reduced-Price Lunch (FRL): Number of
enrolled students who are eligible for the federal free- and reduced-price lunch
program,
2. Number of ELL/Bilingual Students: As of the day of the site visit, the number of
students eligible for services as an English language learner (ELL) as defined by
the federal No Child Left Behind Act of 2001,
3. English/Reading/LA, History/Social Studies, Mathematics, Science, and Foreign
Language: Number of Full Time Equivalent (FTE) licensed subject-specific
teachers who teach the core subjects,
4. Drama/Technology/Health: Number of FTE teachers who provide instruction in a
subject area that represents a special academic focus,
5. Certified Teacher Tutors: Number of FTE tutors who are licensed teachers and
provide help to students one-on-one or in small groups of 2-5,
6. Number of Professional Development Days in the Teacher Contract: Number of
days the teacher contract specifies for professional development, and
7. Secretary: Number of FTE Secretaries.
This study also included a series of standardized open-ended interview questions
based on school finance adequacy research conducted across the United States, then
105
modified to meet the California context. The wording and sequence of these questions
was determined during the Codebook Protocol training and all interviewees were asked
the same fundamental questions in the same order to reduce interviewer bias and to
increase comparability across responses (Patton, 2002). While this format minimized
flexibility in “relating the interview to particular individuals and circumstances” (Patton,
2002, p. 349), it permitted multiple researchers within the dissertation group to collect
similar data for possible subsequent comparison. Examples below illustrate the nature of
the open-ended questions:
1. What has been the content focus of your improvement process?
(e.g., Reading, Math, etc.)
2. What is the instructional vision for your improvement effort?
(e.g., California standards, the Danielson Framework, etc.)
This study did not include visits to classrooms or interaction with students.
However, one principal insisted upon visiting a meeting of the school’s Discipline and
Attendance Review Team, in which no students were present. The purpose of this study
was to identify resources and the perceptions of school leaders regarding their best use.
While a number of data points may be extracted from an independent review of school
documents, the interview protocol offered the opportunity to acquire specific quantitative
resource-use data alongside qualitative data regarding efforts to improve student
achievement. It additionally supported triangulation and clarification of data located
within the documents. The entire set of standardized close- and open-ended interview
questions may be found in Appendices E and F titled Data Collection Codebook
106
and Open-Ended Data Collection Protocol-School Sites respectively. Site and district
profile information was collected as illustrated in Appendix G titled Data Collection
Protocol – School Sites.
Data Analysis
Approach and Processes
The purpose of this study was to capture 2008-2009 resource use in public Title I
Program Improvement middle schools in one county in southern California. Those data
were then compared with the Evidence-Based Model. Procedures included delineation of
resources used within each school, comparison with other schools within the study,
comparison of each school with the model, and analysis of open-ended interview
responses for patterns, themes, and understandings relating to effective allocation of
resources.
Analysis of interview data included quantitative data entry and clean up,
identification of type and quantity of instructional improvement resources utilized at each
school (e.g., number of English teachers), and comparison of resource use at each school
with the Evidence-Based Model. Close-ended data also provided context for themes and
patterns as they emerged in the open-ended data analysis. The investigator anticipated
that principals’ responses to open-ended questions in which they described their
perceptions of the alignment of resource use with instructional improvement would
reveal themes and trends regarding resource allocation challenges.
107
Following each interview, the researcher entered data into a Web-supported,
password protected Excel database. Student and school information included number,
socio-economic, English learner, and special education status of students. Resource
indicators subjected to analysis by averaging included instructional time as measured by
instructional minutes and length of class periods; class size in the core subject areas of
mathematics (special focus on Algebra I, a gatekeeper course), English language arts,
history, and science; class size in non-core and non-academic classes such as art and
music, physical education, and foreign language; class size in special education resource
and self-contained classrooms; and class size in English language development classes.
Analysis also included development of staffing ratios by comparing the number of each
category of teacher (e.g, core, non-core, teacher tutor, special education) to the total
number of teachers at each school.
Chapter Four includes a description of each participating school. Complete case
study write-ups for the participating schools are available in Appendices I through P.
Each case study contains a narrative description, analysis of findings, and illustrative
graphs. Descriptions of the processes used to determine averages for each school are
available in Appendix H to this study, explicitly outlining the analysis processes.
Patton (1987) explains the value of qualitative research. This form of research
reflects the exploratory nature of research and adds depth, detail, and meaning to
findings. Qualitative findings illuminate the diversity among, idiosyncrasies of, and
unique qualities exhibited by individual cases. In this mixed methods study, qualitative
findings served an important role in illuminating the complex processes associated with
108
resource allocation in public schools. Participating schools met Krathwohl’s (1993)
definition of “model cases.” Reflective of this definition, they met specific criteria (e.g.,
middle school, Program Improvement status, & principals who were the school’s chief
administrators during the 2007-2008 school year). They served as clear-cut and
representative samples of Program Improvement middle schools in Orange County,
California.
Qualitative data analysis focused on principals’ understandings of their school
visions, instructional improvement plans, and goals, and how instructional resources are
allocated to achieve those objectives. Analysis also addressed principals’ perceptions
regarding the influence of resource availability on strategic plans for instructional
improvement. Creswell’s (2003) six-step general protocol provided guidance for analysis
of the qualitative data collected during the interviews.
The six-step qualitative data analysis protocol includes (a) transcribing data
followed by (b) a reading of the data for a general sense of their meaning. It continues
with (c) chunking the data into categories, assigning each category a meaningful name
and identifying connections between and across categories. Bogdan and Biklen (as cited
in Creswell, 2003, p. 193) offer a number of possible category types including categories
for participants’ perspectives, activities, strategies, and setting/context. During the fourth
step, the author (d) generated a description of the setting, instructional strategies, resource
allocation, and challenges for each middle school case study. The author also identified a
short list of themes and patterns representing multiple perspectives and reflected in
evidence and quotations that emerge during the categorizing and describing processes.
109
The protocol then includes (e) preparing a narrative of qualitative findings
including themes, evidence, and quotations that connect and differentiate each school
with regard to the same set of interview questions. Creswell’s protocol is further
informed by Stake’s (2002) emphasis on comparing and contrasting the cases studied.
Finally, the researcher (f) attached meaning to the data by explaining the findings,
relationships, and patterns. Interpretation also included identification of how the data
relate to educational adequacy research and consideration of possible new questions that
could enhance the data collection protocol. Analysis of these qualitative data were
vulnerable to researcher subjectivity, but offered potential insight into the direction from
which Title I Program Improvement middle schools approach efforts toward educational
adequacy.
Case study narratives, findings, analysis, and discussion of the findings follow in
Chapter Four.
110
CHAPTER FOUR
Results of the Study
This study was conducted with the intent of identifying current types and levels of
resource allocation across a number of schools meeting specific criteria. Analysis of data
gathered during the course of interviews with Program Improvement middle school
principals led to detailed portraits of each school as well as a comparison of resource
allocation across the eight participating schools. The following will unfold in this
chapter, which emphasizes results of the study: (a) restatement of the research questions,
(b) description of each of the case study schools, (c) presentation of quantitative analysis
of resource allocation and interview-based rationales for resource allocation, and (d) a
summary of the findings framed by the research questions that guided this study.
Restatement of Research Questions
The purpose of this study was to capture 2008-2009 resource use in public Title I
Program Improvement middle schools in one county in southern California. Those data
were then compared with the Evidence-Based Model. Procedures included delineation of
resources used within each school, comparison with other schools within the study,
comparison of each school with the model, and analysis of open-ended interview
responses for patterns, themes, and understandings relating to effective allocation of
resources. The interview protocol and post-interview data analysis supported answering
four research questions:
111
1. What are the current instructional improvement strategies at the school-level?
2. How are resources used to implement the school’s instructional improvement
strategies? (e.g., instructional vision, plan, goals, et cetera)
3. How are the actual resource patterns at the school sites aligned with or different
from the resource use strategies that are used in the Evidence-Based Model?
4. How does the availability of resources affect the development and
implementation of the instructional improvement plan?
Following a brief description of each participating school, this chapter will
continue with a presentation of findings and a subsequent summary.
The Sample Schools
Eight Title I Program Improvement middle schools in Orange County, California
participated in this study of school-level resource allocation. Participating schools
represented six school districts. The principal at each school had served as the school’s
chief instructional leader for at least one year prior to the 2008-2009 school year and is
deeply involved with the school’s improvement efforts. An alias for each school and its
district obscures their identities. Descriptive data include California Standards Test
(CST) results, Academic Performance Index (API) and other assessment and
demographic data available on the California Department of Education Web site (Data &
Statistics page). Participating principals provided additional descriptive data during case-
study interviews conducted between mid-September and late-October 2008. Case study
descriptions are presented in alphabetical order as per the alias assigned to each school.
They provide the context for the study.
112
Case Study School Description: Botticelli Junior High School
Botticelli Junior High School is a grade seven and eight school with 927 students
in an urban district in Orange County, California. The Guggenheim School District
serves approximately 10,200 students in grades kindergarten through eight. Botticelli
Junior High School is one of three junior high schools in the district. This Title I-assisted
school has been identified as a Year Four Program Improvement school for the 2008-
2009 school year.
Demographic characteristics of Botticelli JHS students diverge from those of the
district overall. Roughly 86% of the students at Botticelli Junior High School are
classified as “minority” as compared with 73% with minority classifications across the
district. Seventy-six percent of the school’s students are identified as Hispanic/Latino, a
rate that is 30% higher than that of the district. At 14%, Botticelli’s White population is
13% lower than that seen district-wide.
Approximately 74% of Botticelli Junior High School students are eligible for the
FRPL program, as compared with 41% of students district-wide. Roughly 15% of
Botticelli’s students have been identified to receive special education services as
contrasted with 11% across the district. While 21% of the district’s overall population is
Asian, this group represents just four percent of students at Botticelli. Students
identified as English learners comprise 39% of Botticelli Junior High School’s student
population, a rate that is seven percent higher than the district’s overall English learner
proportions.
113
Botticelli’s Academic Performance Index (API) has increased 100 points from
591 in 1999 to 691 in 2008 with its greatest growth (29 points) between 2005 and 2006.
Disaggregated student achievement data indicate differences in the achievement
results of various student demographic groups. English learners and special education
students attain lower CST results than those of their schoolmates. Similarly, a
performance gap exists for every student group at Botticelli when compared with their
peers throughout the district.
API data assist in creating a detailed portrait of student achievement at Botticelli
over time. Since the inception of the Academic Performance Index in 1999, Botticelli
has been identified with the lowest academic scores among the three middle schools in
the district. Student achievement as measured by the API at Botticelli Junior High School
has remained relatively flat during the six years between 2003 and 2008. In comparing
Botticelli’s student achievement with demographically similar schools in the area, the
principal has noted that with its 100 API point growth between 1999 and 2008, the school
continually far underscores those in its similar-schools comparison group. The
elementary schools that feed into Botticelli all have attained APIs over 700 while
Botticelli remains in the 600s.
Botticelli historically has faced inconsistent mathematics achievement results.
Fluctuations in proficiency levels in Algebra reflect those in other mathematics measures
(e.g., general mathematics) at the school. In spite of Algebra being offered to students
who meet course entry criteria (e.g., successful completion of pre-Algebra), student
achievement has seesawed from 29% proficient and advanced one year to 70% another.
114
Since attaining Program Improvement status in 2005, Botticelli Junior High
School has concentrated school improvement efforts on language arts instruction,
instructional strategies to maximize student learning, specific mathematics
understandings and skills (e.g., multiplication), specialized interventions to improve
reading and mathematics skills, increased time for struggling learners in language arts
and mathematics, development of pacing guides in each content area, teaching at grade
level, positive school environment, professional collaboration and use of data, distributed
leadership, student use of technology, and moving first to well managed then to highly
effective classrooms. Details regarding Botticelli’s improvement efforts are available in
the case study write up in Appendix I.
Case Study School Description: Chagall Junior High School
Chagall Junior High School is a grade seven and eight school with 1,360 students
in an urban district in Orange County, California. The Getty Union High School District
serves approximately 26,000 students in grades seven through twelve. Chagall Junior
High School is one of eight middle schools in the district. After failing to meet federal
Adequate Yearly Progress proficiency targets, this Title I-assisted school has been
identified as a Year Three Program Improvement school for the 2008-2009 school year.
Student demographic characteristics at Chagall JHS differ somewhat from those
across the district. Roughly 89% of the students at Chagall Junior High School are
classified as “minority” compared with 82% district-wide. Comprising 71% of Chagall’s
students, the predominant student subgroup consists of Hispanic/Latino students.
Chagall’s Hispanic/Latino population is 10% higher than that of the district while its
115
White population is seven percent lower than seen across the Getty Union High School
District.
Approximately 67% of Chagall Junior High School students are eligible for the
Free and Reduced Price Lunch (FRPL) program, as compared with 56% of students
district-wide. In line with the district, nine percent of Chagall’s students receive special
education services. The school serves as a central service point for visually and
orthopedically impaired students. Students identified as English learners comprise 62%
of Chagall Junior High School’s student population, a rate that is 15% higher than the
district’s overall English learner proportions.
The story of Chagall’s API growth includes gradual growth as student
demographics have changed. Chagall’s API has risen 144 points, from 571 in 1999 to
715 in 2008. Its greatest API increase occurred in 2008 when this score increased by 36
points. Significant gaps persist across the school’s federally significant student
subgroups: White, economically disadvantaged, special education, English learner, and
Hispanic/Latino. Yet, Hispanic/Latino students have experienced greater API increases
during the past 10 years than their White counterparts.
Between 1999 and 2008, White students experienced 119-point API growth while
Hispanic/Latino scores increased by 216 points. Economically disadvantaged student
scores increased by 147 points between 2001 and 2008. Between 2006 and 2008, English
Learner API scores increased by 22 points and special education student scores have
grown by 19 points.
116
Because the state added various student groups to its API reporting system in
different years, finding an average (mean) rate of increase helps in painting a picture of
student academic achievement gains. At 7.3% and 6.3% respectively, English learner
and special education students have experienced the slowest average growth in API
scores. These student groups also are identified with the lowest API results at the school.
Hispanic/Latino students have experienced the greatest growth (21.6%) with
economically disadvantaged students gaining 18.4%. In the center of the ranking, White
students have averaged an 11.9% API growth rate.
Eighth-grade Algebra instruction and assessment illustrate the achievement gap
and efforts to bridge it at Chagall. During the past six years the percentage of
Hispanic/Latino eighth-graders tested in Algebra has increased from 1.8 in 2003 to 21.1
in 2008. The number of economically disadvantaged students enrolled in Algebra has
increased from four percent of the population to 21.5%. English learner participation has
increased from one percent (7 students) to 4.9% (36 students) of the eighth-grade
population. The single special education student that took Algebra at Chagall JHS was
the first to take the Algebra exam since 2003. White students take the Algebra exam at a
higher rate than their Hispanic/Latino, economically disadvantaged, English learner, and
special education peers.
It is of interest to note that seven percent of the non-economically disadvantaged
Hispanic/Latino students enrolled in Algebra in 2007-2008. At 14.2%, just over twice as
many economically disadvantaged Hispanic/Latino students participated in this subject.
Furthermore, 53% of the economically disadvantaged Hispanic/Latino group’s students
117
attained proficient or advanced Algebra results compared with 37% of its non-
economically disadvantaged Hispanic/Latino members.
As mentioned above, Chagall JHS has experienced an increase in its annual API
results. Incongruously, annual CST results (upon which the API is calculated) indicate a
dramatic drop in language arts scores between 2006 and 2007 as well as a decline
between 2005 and 2006 in Algebra scores. However, between 2003 and 2008, the school
increased the number of students taking the Algebra exam from 43 to 274. Between 2005
and 2006, the number of Algebra students more than doubled, increasing from 61 to 147.
At the same time, the percentage of proficient and advanced students declined from 92 to
61%. By 2008, the proficiency percentage had nudged above its all time low in 2007 of
56% while providing Algebra instruction to the largest percentage of students during the
period measured. More students than ever participated in Algebra in 2007-2008, yet
participation rates are not distributed equally across all student groups. Further
examination of CST scores indicates that White student scores exceed those of all other
student groups in all content areas. Science scores have increased rapidly since inclusion
of this exam in 2006. The percentage of students attaining proficient and advanced social
studies standards test scores has never exceeded the mid-twenties.
Since attaining Program Improvement status in 2006, Chagall Junior High School
has concentrated school improvement efforts on student engagement, mentoring, use of
benchmark assessments and resulting data, and effective academic and social
interventions. The detailed case study in Appendix J describes the school’s improvement
efforts at length.
118
Case Study School Description: Degas Intermediate School
Degas Intermediate School is a grades six-through-eight school with 1,173
students in an urban district in Orange County, California. It is a school within the
Whitney Unified School District, which serves approximately 57,000 K-12 students.
Degas Intermediate School is one of nine middle schools in the district. Three of the nine
are labeled “fundamental” schools and provide no services to English learners or special
needs students. Degas Intermediate School did not attain the minimum percentages of
students with scores at or above “proficient” in language arts (35.25) and mathematics
(37%) to meet federal Adequate Yearly Progress goals in 2008. This Title I-assisted
school has been identified as a Year Five Program Improvement school for the 2008-
2009 school year. Additionally, the Whitney USD has been identified as a Year Three
Program Improvement district as well as being identified by the state for District
Assistance and Intervention Team (DAIT) services.
Almost 100% of the students at Degas are classified as “minority.”
Hispanic/Latino students comprise the primary ethnic subgroup and represent over 99%
of the student population at Degas. This ethnicity proportion is slightly higher than the
district’s overall enrollment of 94.4% Hispanic/Latino students.
A demographic comparison between Degas and Whitney USD reveals greater
representation of historically underserved students at Degas than across the district
overall. Approximately 96% of Degas Intermediate School students are eligible for the
FRPL program, as compared with 88% of students district-wide. Approximately 11% of
Degas students receive special education services compared with nine percent throughout
119
the district. Students identified as English learners comprise 87% of Degas’ student
population, diverging from the district’s English learner percentage (44%) at the middle
school level.
Of the nine intermediate schools in the district, six are in Year Five of Program
Improvement and one of the three fundamental intermediate middle schools is in Year
Two. The two intermediate schools not identified as “failing” by federal standards are
two of the three fundamental schools.
Degas student achievement test results trail those of the other five “traditional”
intermediate schools. Along with four other intermediate schools, Degas has not broken
the 700 API ceiling. During the four years that the state has calculated school-, district-,
and state- level Academic Performance Indexes, Degas consistently has attained lower
numbers annually than those of the district overall. In 2008, Degas missed the state
middle school average API by 122 points. The school trailed the Whitney USD API
(grades 2-11) by 86 points. Not only does the district fall behind the state API average,
Degas also trails the district for each of the school’s significant student groups.
During the 10 year period that APIs have been calculated by the California
Department of Education, Degas’ API has grown 162 points: from 441 in 1999 to 603 in
2008. The school to date has been unable to eclipse its highest API, a score of 621
attained in 2005 at the end of the current principal’s first year at the school.
Based on CST results, mathematics achievement is the high point in calculating
Degas’ school success. Twenty-four percent of Degas’ eighth-grade students enrolled in
Algebra during the 2007-2008 school year. Roughly 43% of those Algebra students
120
attained proficient or advanced scores on the CST in this subject. The majority of eighth-
grade students (74%) were enrolled in a pre-Algebra course. They were tested on sixth-
and seventh-grade California mathematics standards. Twenty-one percent (74 students)
attained proficient or advanced scores on this assessment. Degas teachers continue to
consider the learning needs of special education and English learner students in
mathematics.
A multi-year, multi-subject view from 2003-2008 (years that the state has
implemented criterion-referenced, standards-based assessments annually) affords a
longer-term perspective on student achievement efforts. Scores in science appear to be
moving consistently and steadily upward while the trajectory of language arts and social
studies appears similar and relatively flat. Achievement scores in Algebra dipped in 2004
and 2005 when the school included as much as 88% of the eighth-grade student
population in the course. Scaling back the number of participating students, then
gradually increasing it has coincided with achievement score increases. Degas’ API has
moved from 441 in 1990 to 603 in 2008 at a lurching pace including a 31 point decline in
2006 from which the school is recovering.
During the past five years, Degas Intermediate School has concentrated school
improvement efforts on language arts, mathematics, science, and student behavior. The
school has worked to organize its faculty and schedule to support consistent collaboration
and increased instructional time in mathematics and language arts. Appendix K contains
the case study, which includes detailed descriptions of the school and its efforts.
121
Case Study School Description: Picasso Middle School
Picasso Middle School serves 963 students in grades six through eight in the
6,400-student K-8 Hermitage School District Orange County, California. Picasso Middle
School is one of two middle schools in the district. After failing to meet NCLB
proficiency targets with special education and English learner students in language arts
and mathematics, both middle schools have been identified as Year Three Program
Improvement schools for the 2008-2009 school year. Low achievement rates at the two
middle schools resulted in the district attaining Program Improvement status as well.
Student demographics at Picasso are nearly identical with those of the district
overall. Roughly 82% of Picasso Middle School’s students are classified as minority,
with 77% listed as Hispanic/Latino. The next largest minority group consists of Asian
students and comprises 2.6% of the school’s enrollment. Sixty-two percent of Picasso’s
students are considered economically disadvantaged and 53% are identified as English
learners. Eighteen percent of Picasso’s students are identified as White. Special
education students represent 11% of both Picasso’s and the district’s enrollments. Student
groups numerically significant for federal achievement data reporting purposes include
those identified as Hispanic/Latino, White, English learner, special education, and
economically disadvantaged.
Picasso’s student body has undergone demographic changes between 2003 and
2008. The percentage of White students has declined from 35% of the student population
to 17.7%. A corresponding increase in Hispanic/Latino students is reflected in the
increase in this group’s representation from 61% to 76.2%. The percentage of English
122
learners has not increased in such a notable fashion, moving upward from 33% in 2003 to
35.5% in 2008. The percentage of students identified as economically disadvantaged has
increased from 50% of the student population to 61.2%. While the school has
experienced recent demographic changes, perennially low achievement test results in two
stable student groups, English learner and special education, led to Picasso initially
attaining Program Improvement status in 2007.
A glance at student language arts achievement rates during the six years that the
California Standards Tests have measured student academic achievement in a criterion-
referenced format highlights several trends. Even with a recent decline, White students
outscore other subgroups by as much as 40%. Special education students have a history
of low achievement levels. In both 2005 and 2007, no eighth-grade special education
student attained a proficient or higher language arts test result. Of the 24 special
education eighth-grade students tested in language arts in 2008, two attained scores
deemed proficient. Eight of Picasso’s 128 eighth-grade English learners attained
proficient or higher scores in language arts.
While improvement rates for special education and English learner students
appear relatively flat and that of White students seems to be declining, improvement for
students identified as economically disadvantaged or Hispanic/Latino has roughly
doubled between 2003 and 2008. Picasso strives to overcome a persistent performance
gap between White and all other student groups combined with lower-than-federally-
acceptable achievement levels among all non-White student groups.
123
In addition to improving language arts performance, Picasso Middle School also
must improve performance in mathematics. Eighth-grade test results in both general
mathematics and Algebra reflect challenges in meeting NCLB achievement goals.
Eighth-grade student achievement in Algebra has fluctuated wildly both during the six-
year period between 2003 and 2008 and across student groups within a school year.
Notable are 2005 when no White students attained proficient results and 2006 when the
same held true for Picasso’s economically disadvantaged students. In the years for which
data are provided, no students from either the special education or the English learner
group have attained proficient or advanced results in Algebra. These details illustrate the
complexity of Picasso’s school improvement challenge.
Looking at one year across disciplines reveals current areas of strength as well as
what is happening within one cohort of students. Federal proficiency targets in 2008 for
language arts and mathematics were 35.2% and 37% respectively. The White student
group was the only one to attain an adequate percentage of students at and above
proficient to meet NCLB requirements. Each of these targets increases by approximately
11% in 2009.
During the past two years since acquiring Program Improvement status, Picasso
Middle School has focused its improvement efforts on effective instruction and support
for struggling students. This has entailed changes to the student and staff schedules,
professional development, and additions to the professional staff roster. A detailed case
study write-up, Appendix L, tells the story of Picasso Middle School’s efforts to improve
instruction to increase student learning and performance.
124
Case Study School Description: Renoir Intermediate School
Renoir Intermediate School, a grades six-through-eight school with 1,738
students, is one of nine middle schools in the 57,000-student Whitney Unified School
District in Orange County, California. Three of the district’s nine middle schools are
labeled “fundamental” schools and provide no services to English learners or special
needs students. Renoir Intermediate School did not attain the minimum percentages
(35.2%) of students with scores at or above “proficient” in language arts to meet federal
Adequate Yearly Progress goals in 2008. As a result, Renoir has been identified as a
Title I Year Five Program Improvement school for the 2008-2009 school year. One of the
fundamental middle schools has attained Year Two Program Improvement status and all
of the traditional middle schools are in Year Five of Program Improvement. The Whitney
USD has the distinction of being identified as a federal Program Improvement and a state
School and District Assistance and Intervention Teams-supported district.
Renoir’s student population demographics diverge somewhat from those of the
district overall. Roughly 99% of the students at Renoir are classified as “minority.”
Hispanic/Latino students comprise the school’s largest ethnic subgroup and represent
98% of the school’s population. This ethnicity proportion is slightly higher than the
district’s 94.4% Hispanic/Latino population.
Approximately 96% of Renoir Intermediate School students are eligible for free
and reduced price lunch, as compared with 88% of students district-wide. The school is
applying for 100% free-and-reduced-price lunch status for the 2008-2009 school year.
Close to 10% of Renoir students receive special education services compared with nine
125
percent throughout the district. Students identified as English learners comprise 53% of
Renoir’s student population, diverging from the district’s English learner percentage
(44%) at the middle school level overall.
The Academic Performance Index (API), calculated from CST results in the core
content areas of mathematics, language arts, science, and history-social science, provides
a means of comparing schools with each other. Based on this measure, Renoir student
achievement results outpace those of the other five traditional intermediate schools. With
a difference of 34 API points between itself and the nearest-performing fundamental
intermediate school and 50 points between itself and the nearest-performing traditional
intermediate school, Renoir student achievement approaches that of the district’s
selective-acceptance schools.
During the four years that the state has calculated school-, district-, and state-
level Academic Performance Indexes, Renoir consistently has attained slightly higher
numbers annually than those of the district overall. In 2008, Renoir approached the state
average middle school API, trailing it by 15 points. The school exceeded the Whitney
USD API (grades 2-11) by 31 points. The school, district, state middle school, and state
overall APIs grew at roughly the same rate between 2007 and 2008 (i.e., between 1.02
and 1.04%). As federal proficiency targets increase by approximately 11% annually
between 2009 and 2014, Renoir’s growth must accelerate in order to exit Program
Improvement status. The school recognizes a persistent achievement gap between
English learner and special education students with the remainder of the school
population.
126
Renoir’s approach to improving student achievement in mathematics includes
gradually increasing the number of students enrolled in Algebra. High student
achievement scores in this area are tempered by participation rates. Nearly half of
Renoir’s students took Algebra while 11% of its English Learners participated in this
course. Just one in eight special education students took Algebra. Resolving this
difficult and complex challenge is a longstanding piece of Renoir Intermediate School’s
improvement puzzle.
A multi-year, multi-subject view illustrates trends in student achievement. Since
its inception in 2006, scores in the annual science test appear to be moving consistently
and steadily upward while the trajectory of language arts and social studies appear similar
and relatively flat. Achievement scores in Algebra have fluctuated as the school has
increased its number of students participating in this gatekeeper course. Renoir’s API has
moved from 440 in 1999 to 720 in 2008 at a relatively consistent rate of roughly 28
points each year.
During the past four years, Renoir Intermediate School has directed school
improvement energy toward expository writing and mathematics. The school has worked
to organize its faculty and schedule to support consistent collaboration and time in
mathematics and language arts. Appendix M provides a complete description of Renoir
Intermediate School and its school improvement efforts.
Case Study School Description: Rodin Intermediate School
Rodin Intermediate School serves students in grades seven and eight in the
D’Orsay Unified School District, which serves approximately 21,400 Pre-K through
127
grade 12 students in Orange County, California. Elimination of sixth grade in 2008
resulted in an overnight decline in enrollment from approximately 1,000 students to the
652 students enrolled this year.
Unique among the case study schools, Rodin met federal Adequate Yearly
Progress goals in 2008. As a result, this Program Improvement school has retained its
Year Four status and has the potential to exit Program Improvement altogether if all
student groups meet next year’s increased proficiency targets.
Roughly 78% of the students at Rodin Intermediate School are classified as
minority. Comprising 73.6% of Rodin’s students, the predominant ethnic student
subgroup considered numerically significant for state and federal reporting purposes
consists of Hispanic/Latino students. Student race and ethnicity proportions at Rodin
diverge from those of the district overall. Notably, Rodin’s Hispanic/Latino population is
32% higher than that of the district while at 21.5%, its White population is nearly 30%
lower than that seen across the district.
Approximately 70% of Rodin Intermediate School students are eligible for the
FRPL program, as compared with 40% of students district-wide. Approximately 10.2%
of Rodin’s students receive special education services as compared with 9.7% across the
district as a whole. Students identified as English learners comprise 38.4% of Rodin
Intermediate School’s student population, a rate that is 12% higher than the district’s
overall English learner proportions. In addition to diverging from the district overall,
Rodin Intermediate School’s student demographics are markedly different from the
district’s other intermediate schools.
128
Rodin’s Academic Performance Index has risen from 584 in 1999 to 756 in 2008
with 38 points growth in 2008 and 43 points in 2007. During the past six years, the
school has experimented with increasing the percentage of students participating in
Algebra. Rodin began in 2003 with 21% of its eighth-grade students taking the Algebra
exam to a peak in 2007 when 95% of Rodin’s eighth graders took the test. In 2008,
36.8% of Rodin’s eighth-grade students took the Algebra exam with a high proficiency
rate. Participation rates for Hispanic/Latino and economically disadvantaged students
have been nearly identical during the six-year period from 2003 to 2008. Both student
groups have trailed the overall and White student group participation rates. The number
of special education students participating in Algebra has been low, with no students
from this group taking the Algebra exam in 2008. English learners participate at a lower
rate than other student groups as well.
A summary of 2008 proficiency results is incomplete without highlighting the
school’s high percentage of students attaining proficient and advanced scores in Algebra.
White students attained the highest proficiency rates, followed by economically
disadvantaged students, and Hispanic/Latino students. Special education and English
learner students alternated in terms of proficiency rates, with special education students
outscoring English learners in general mathematics. For their part, the English learner
group attained high levels of proficiency in Algebra, contrasted with special education
students who did not participate in this test. English learners additionally experienced
higher achievement levels than their special education peers in science and social studies.
129
Special education-English learner performance data point to the challenges of serving
diverse student learning needs.
In addition to the single-year perspective above, a multi-year view reveals trends
in student achievement. Examination of student achievement data from 2003 to 2008 in
the core content areas of mathematics, language arts, science (added in 2006), and social
studies indicates the greatest gains in student performance in mathematics (Algebra &
general mathematics combined), accompanied by a doubling of students proficient in
language arts. Algebra scores dipped in 2007 when the school offered Algebra as the
only math option to all eighth-grade students, and rose dramatically in 2008 when it used
placement criteria. Additionally, when general mathematics was reinstated during the
2007-2008 school year, the percentage of proficient and advanced students (35%)
reflected 32% growth from 2006. History-social science scores indicate a relatively flat
growth rate while scores in science have more than doubled since implementation of this
exam in 2006.
Since the current principal joined Rodin in 2006, the school has set its sites on
increasing expectations for student learning and becoming a California Distinguished
School within four years. School improvement efforts include a focus on student
learning. Frequent assessments, checking for understanding, higher-level questions,
teaching to multiple learning modalities, pre-teaching, using instructional strategies that
keep students engaged and active, frequent analysis of student formative assessment data,
and not permitting students to fail reflect the focus on learning as opposed to teaching.
130
Teacher collegiality and collaboration form a cornerstone of school improvement
that is supported by concerted efforts to develop a community of teachers who are
comfortable with being observed by peers and taking risks. The complete case study
detailing Rodin Intermediate School’s improvement story is available in Appendix N.
Case Study School Description: Van Gogh Middle School
Van Gogh Middle School serves 822 students in grades six through eight in an
urban district in Orange County, California. The Louvre School District serves
approximately 10,000 students in kindergarten through eighth grade. Van Gogh Middle
School is one of three middle schools in the district. The school did not attain the
minimum percentages of students with scores at or above “proficient” in language arts
(35.2%) and mathematics (37%) to meet federal Adequate Yearly Progress goals in 2008.
As a result, this Title I-assisted school has been identified as a Year Five Program
Improvement school for the 2008-2009 school year.
Roughly 83% of the students at Van Gogh Middle School are classified as
minority. Student subgroups considered numerically significant at Van Gogh primarily
include Latino/Hispanic (51%) and Asian (32%) students. During the past five years the
White subgroup has declined in enrollment to the point that the subgroup is no longer
considered significant for No Child Left Behind reporting purposes. Proportions of
Hispanic/Latino and White students diverge from those across the district. The district
overall enrolls approximately seven percent more White students and 13% fewer
Hispanic/Latino students than Van Gogh.
131
Approximately 70% of Van Gogh Middle School students are eligible for free and
reduced price lunch, as compared with 63% of students district-wide. Ten percent
receive special education services in comparison with 9.3% throughout the Louvre
School District. Students identified as English learners comprise 44% of Van Gogh
Middle School’s student population, aligning with the district’s K-8 English learner
proportions overall. Student misbehavior during the 2006-2007 school year resulted in a
suspension rate 2.5 times that of the district and an expulsion rate five times greater than
the district’s.
The most recent chapter in Van Gogh’s school improvement story includes
increases in students achieving at state-determined adequacy levels and participating in
increasingly rigorous courses. Between 1999 and 2008, Van Gogh’s API has increased
125 points, with 41 points growth in 2008 – at the same time that the school’s enrollment
of White students decreased to fewer than 100 of the school’s students. In 2008, the state
average middle school API was 735. Van Gogh Middle School exceeded this by 15
points.
Elaboration of Van Gogh’s improvement story includes gradual success along
with a commitment to provide all students with an academic foundation to serve them in
high school and beyond. The percentage of students attaining CST results indicating
proficiency in language arts has increased from 34% to 44% from 2003 to 2008.
Examination of eighth grade Algebra CST results shows an increase from 12% to as high
as 91.6% in the number of students taking the exam between 2003 and 2008, with a
decrease from 91% to as low as 24% in the percentage of student test results at proficient
132
and advanced levels. The most recent results indicate high levels of student participation
with 35% of tested students attaining proficient and advanced results.
Even with recent school-wide API growth, student test scores reflect differences
in achievement across student groups. Student scores in language arts, Algebra, history-
social science, and science at the eighth grade reveal gaps in performance in all areas
between special education, English learner, and Hispanic/Latino students when compared
with their White and Asian schoolmates. In 2008, the API for economically
disadvantaged students increased at a higher rate than that of other student groups.
Each spring, eighth-grade students take a battery of standards-based exams in
language arts, mathematics, science, and social studies. A look at results on these tests
over time reveals student achievement trends. Scores in science appear to be moving
consistently and steadily upward. Language arts scores gradually moved upward until a
slight decline in 2007. History-social science scores peaked in 2005 followed by recent
flattening. Achievement scores in Algebra declined as the school increased its number of
students participating in this gatekeeper course. Those scores increased by 11% in 2008.
As the number of students taking the general mathematics exam reached its lowest point,
the percentage of students deemed proficient and advanced eclipsed that of any previous
year. Van Gogh’s API has moved in fits and starts from 627 in 1999 to 750 in 2008.
During the past two years Van Gogh Middle School has concentrated school
improvement efforts on professional collaboration among teachers, increased time in
language arts and mathematics, “Gradual Release of Responsibility” (a teaching protocol
emphasizing modeling, purposeful student groupings, and support of increasing academic
133
independence), student engagement, effective instructional strategies, and Response to
Intervention (supporting students as needed to minimize inappropriate identification for
special education services). The school has utilized multiple grants focused on
improvement of mathematics instruction and teacher collaboration to provide
professional development for teachers. Appendix O provides a detailed description of
Van Gogh Middle School and its improvement efforts.
Case Study School Description: Warhol Intermediate School
Warhol Junior High School serves 1,722 seventh- and eighth-grade students in the
26,000-student, grades seven-through-twelve Getty Union High School District in
Orange County, California. Warhol Junior High School’s performance levels are
identified as the lowest of the eight middle schools in the district. Warhol) did not attain
the minimum percentages of students with scores at or above “proficient” in language
arts (35.2%) and mathematics (37%) to meet federal Adequate Yearly Progress goals in
2008. In particular, the special education student group’s CST results fell below federal
proficiency targets. This Title I-assisted school has been identified as a Year Five
Program Improvement school for the 2008-2009 school year.
Demographic groups at Warhol diverge from those represented across the Getty
Union High School District. Roughly 97% of the students at Warhol Junior High School
are classified as “minority.” Comprising 94% of Warhol’s students, the predominant
ethnic student subgroup considered numerically significant for state and federal reporting
purposes consists of Hispanic/Latino students. Warhol’s Hispanic/Latino population is
134
32% higher than that of the district while at three percent, its White population is 15%
lower than that seen across the district.
Approximately 88% of Warhol Junior High School students are eligible for free
and reduced price lunch, as compared with 56% of students district-wide. In line with the
district, 9.6% of Warhol’s students receive special education services. Students identified
as English learners comprise 82% of Warhol Junior High School’s student population, a
rate that is 35% higher than the district’s overall English learner proportions.
Warhol’s Academic Performance Index has risen from 479 in 1999 to 665 in 2008
with its greatest growth (41 points) in 2008. During the past six years the percentage of
students tested in Algebra has increased from 12% to 40%. English learner and
economically disadvantaged students’ scores reflect similar increases. The percentage of
special education students taking the Algebra test never has risen above one percent.
Student achievement varies across subject areas. History-social science and
English/language arts performance is roughly similar for all student groups. Since 2003,
history and language arts have followed a nearly identical, gradual growth trajectory.
Student test results in science indicate greater achievement in this subject area than in the
other core disciplines. Since its inclusion in the CST battery in 2006, overall proficiency
rates in science have grown dramatically each year, with roughly 40 points growth
between 2006 and 2008. Algebra proficiency rates moved quickly into the 60% and
higher range until Warhol doubled its participation rate in 2007. Following a one-year
dip and yet another participation increase, proficiency rates have climbed almost to the
50% mark. Proficiency rates in general mathematics have never exceeded 21% and have
135
declined to seven percent in 2008. In terms of achievement results, Algebra performance
in 2008 took second place behind science.
In addition to identifying subject areas with higher results, a look across student
groups highlights the performance gap between the school’s English learners and special
education students with their schoolmates. At 48% and 9.2% respectively, English
learners and special education students comprise over half the Warhol student population.
The school recognizes the paramount significance of meeting their learning needs.
Since attaining Program Improvement status in 2004, Warhol Junior High School
has concentrated school improvement efforts on language arts and mathematics
instruction, teaching at grade level, increasing the number of students taking Algebra,
college readiness, class-size reduction, professional collaboration and use of data,
distributed leadership, parent education, and partnerships with external agencies,
universities, and other schools in the district. A detailed description of the school and its
improvement efforts may be found in Appendix P.
Description Summary
The eight participating schools contain a number of commonalities as well as
differences. Similar to other Program Improvement schools in Orange County, each
participating school has demographics reflecting high numbers of Hispanic/Latino,
English learner, and economically disadvantaged students relative to schools across the
county and the state. These three groups were considered numerically significant at each
school in 2007-2008 for federal reporting purposes (i.e., at least 100 students). Table 4.1
highlights the population characteristics at each of the case study schools.
136
Table 4.1. Student Demographic Groups Comparison
Botti-
celli
Chagall Degas Picasso Renoir Rodin Van
Gogh
Warhol
Asian 3.8 6.8 0.3 2.6 0.5 1.9 32.0 1.0
Hispanic/
Latino
76.0 70.7 99.0 77.0 98.0 73.6 51.0 86.0
White 13.7 11.3 0.6 18.0 0.4 21.5 11.0 5.0
Econ.
Disadv.
74.0 67.8 96.0 62.0 96.0 40.0 70.0 82.0
EL 39.0 62.0 87.0 53.0 82.0 38.4 44.0 75.0
SpEd 14.5 9.0 11.0 11.0 10.0 10.2 10.0 9.0
Three schools, Degas, Renoir, and Rodin each enrolled fewer than 100 special
education students during the 2007-2008 school year. This contributed to Rodin
Intermediate School meeting its federal AYP goals last year. However, according to the
data provided during each interview, the number of special education students at each of
the eight schools is at least 100 students in 2008-2009, warranting inclusion in upcoming
state and federal accountability calculations.
Four schools, Degas, Renoir, Van Gogh, and Warhol enrolled fewer than 100
White students during the 2007-2008 school year. Rodin’s student population contained
the greatest number of White students. Yet, at 21.5%, this group’s representation trails
California’s overall White student enrollment percentage (28.5%) as well as that within
Orange County (33.8%). At each of these four schools, White students’ APIs were as
much as 149 points higher than those of the second-highest scoring group. A similar
performance gap exists at Warhol between the school’s Asian student group and its next-
highest scoring group. Table 4.2 summarizes student achievement at each school for
each significant group as measured by the API.
137
Table 4.2. 2008 API by Student Group
Botti-
celli
Chagall Degas Picasso Renoir Rodin Van
Gogh
Warhol
Asian * * * * * * * 862
Econ.
Disadv.
654 698 598 658 718 719 653 738
EL 571 673 576 631 688 674 637 692
Hisp./
Lat.
657 686 603 671 717 711 661 672
Sp. Ed. 475 458 * 526 * * 441 529
White 764 795 * 799 * 868 * *
*Note: California does not report APIs for groups with fewer than 100 students.
In addition to the enrollment differences noted above, participating schools have
held Program Improvement status for varying numbers of years. Of the four Year Five
Program Improvement schools participating in the study, three have maintained their
Year Five status for a number of years. Botticelli JHS is the “true” Year Five school,
while Renoir is in its eleventh year as a Program Improvement school. Van Gogh is in its
ninth year and Warhol is in its sixth. Because Rodin IS met its AYP targets in 2007-
2008, the school retained its Year Four status for 2008-2009. The table below illustrates
the similarities and differences in terms of Program Improvement across the eight
schools.
Table 4.3. 2008 Program Improvement Status and Prior Years in Program Improvement*
Botti-
celli
Chagall Degas Picasso Renoir Rodin Van
Gogh
Warhol
Previous
# of
years in
PI
4 2 5 2 10 4 8 5
PI
Status
5 3 5 3 5 4 5 5
*Note: Program Improvement status does not extend beyond Year Five.
Student achievement results as measured by the API have changed since each
school attained Program Improvement (PI) status. In half of the schools, average API
138
growth declined after attaining PI designation. In the case of Degas, the school moved
from an average of 20.1 points growth annually to 1.8. In three schools, average
achievement results increased anywhere from four to 7.8 points annually. Because it has
held Program Improvement status since 1999 when the API was first implemented,
Renoir’s average “before and after” achievement growth is reported as being the same.
Table 4.4. Average (Mean) Annual API Growth Before and After Attaining Program
Improvement Designation
Botti-
celli
Chagall Degas Picasso Renoir* Rodin Van
Gogh
Warhol
Before 10.8 14.8 20.6 10.1 28.0 10.8 8.0 19.0
After 7.0 22.5 1.8 4.0 28.0 17.8 12.0 17.6
*Note: Renoir has held federal Program Improvement status since 1999 when the California API was first
implemented.
A summary of school-wide API results during the 10 years that such results have
been calculated by the state provides an opportunity to compare student achievement
gains over time. Additionally, California began to calculate a state-wide average (mean)
API in 2004-2005. This score provides an additional point of comparison.
Ten-year API growth across the eight participating schools ranges from 66 points
at Picasso to 280 at Renoir. The average (mean) growth is 152.4 points with a median of
155.5 over the 10-year period from 1999 to 2008. Rodin and Van Gogh both exceed the
state average middle school API (735 points). And Renoir, the school with the greatest
API increase trails the state by 15 points.
Over the past four years (the period during which a state average API has been
calculated), the state average API has grown by eight points annually. Renoir’s average
growth has matched that of the state while Chagall, Rodin, Van Gogh, and Warhol have
exceeded it. The average (mean) API growth across the eight schools studied is 7.6
139
points. Degas has experienced 4.5 points negative average annual API growth. Picasso’s
growth has averaged one point per year and Botticelli’s growth averages at seven points
annually. The case-study schools generally reflect growth that is similar to that across
the state.
Table 4.5. 1999 - 2008 Overall API – Case- Study Schools and California Middle
Schools Overall
School 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Botticelli 591 617 619 640 646 656 650* 679 675 678
Chagall 571 569 587 574 622 648 679 689 670* 715
Degas 441 487 488 510 544 594* 621 570 576 603
Picasso 637 629 644 646 659 686 699 718 695* 703
Renoir 440* 480 501 536 584 612 688 680 692 720
Rodin 584 621 637 641 651 649 685* 675 718 756
Van Gogh 627 643 654* 642 686 703 710 725 709 750
Warhol 479 490 530 542 574 577* 614 623 624 665
Calif** ----- ----- ----- ----- ----- ----- 703 715 720 735
*Note: This marks the API at the end of the school’s first year in Program Improvement.
**Note: California began to calculate a statewide average API in 2005.
The previous individual school descriptions and overall summaries provide a
contextual backdrop for the instructional improvement efforts currently in effect at each
case-study school. The next section addresses the analysis of data collected from the
eight schools that participated in this study.
Presentation of Findings
As addressed in the chapter on methodology, a purposeful sample of eight
Program Improvement middle school principals in Orange County, California agreed to
be interviewed for this study. The data collection process entailed a face-to-face
interview with each principal, and in some cases, additional personnel as determined by
the principal. Principals responded to a series of close-ended quantitative response
questions as well as open-ended questions regarding their school improvement efforts.
140
Quantitative data were entered onto a password-protected Web site and exported as one
cumulative Excel spreadsheet that included all participating schools. A case study write
up for each school detailed both the qualitative and quantitative findings in an effort to
tell the story of each school’s improvement journey. By comparing each school’s
resource allocation data with the Evidence-Based Model for school adequacy as
developed by Odden and Picus (2003), it is possible to examine quantifiable school
improvement elements and their usage. What follows is a presentation of quantitative
analysis of resource allocation and interview-based rationales for resource allocation.
This presentation of findings reflects the organization of resource indicators and
expenditure items as identified the Evidence-Based Model.
School Resource Indicators
School resource indicators inform the allocation of instructional resources in the
Evidence-Based Model. These indicators include school size and categories of students
(reflecting varying types of student needs), and instructional time.
School size. Schools varied in size from 652 to 1738 students. The average
(mean) middle school enrollment was 1170 and the median enrollment was 1068. The
smallest school eliminated one grade level in the 2008-2009 school year, reducing its size
from nearly 1,000 to 652 students.
School composition. The Evidence-Based approach to school adequacy notes the
number of students within the categories of English learners, economically
disadvantaged, and special education to inform resource allocation levels and sources.
141
The number of English learners varied from 248 to 1481 with an average (mean)
of 732.5 and a median of 749. The percentage of English learners varied from 35% to
87% of the student populations at the participating schools. Here, the mean was 59% and
the median was 57.5%.
The number of students participating in the FRPL Program, a proxy for being
economically disadvantaged, varied from 522 to 1669. The mean was 901 and the
median was 713. Percentages of these same students ranged from 60% of a school’s
population to 96%. The range was 36%, with a mean of 74.5% and a median of 69% of
the school’s populations identified as economically disadvantaged.
The total number of students identified to receive special education services at
each case-study school varied from 80 to 159 students. The range was 79 students with a
mean of 126.5 and a median of 131 students. The percentage of special education
students varied from 8.3% to 14.6% across the case-study schools. The percentage range
was 6.3% with a mean of 11.4%, a median of 11.8%, and a mode of 8.3%.
Time. Indicators in this category include length of school day, length of
instructional day, and time allocated to core subjects.
A school day represents the time that class begins in the morning until the time
that school ends in the afternoon. This includes passing periods, breaks, and lunch. It
includes all classes, both core and elective. Measured in minutes, school days varied
from 367 to 408 average minutes daily over the course of a typical school week. Six
schools reported either a late-start or early-release day for students each week to create
time for professional development, collaboration, and/or data analysis. The range was 41
142
minutes with a mean of 381.4 minutes, a median of 378.5 minutes, and a mode of 380
minutes in the school day. Converted to hours, the average (mean) length of the school
day was approximately six hours and 21 minutes.
An instructional day represents the time students spend in class. The typical
instructional day varied from 288 minutes to 352 minutes with a range of 64 minutes.
The median and the mean both were 317.3 minutes. Converted to hours, the average
(mean) instructional day was five hours and 18 minutes. Interestingly, Rodin
Intermediate School, the only case-study school to meet federal proficiency targets in
2008, also had both the longest school day and the longest instructional day in terms of
minutes. Eighty-seven percent of Rodin’s school day is dedicated to instruction.
The percentage of the school day committed to instruction varied from 78% to
91%. The range was 13% with a mean of 83.6% of the school day dedicated to
instruction. The median and mode were the same: 84%.
Core subjects as identified by the state include language arts, mathematics,
history-social science, and science. Time dedicated to language arts and mathematics
instruction varied from 45 minutes to 90 minutes daily with a range of 45 minutes, a
mean of 62.6 minutes, a median of 54.5 minutes, and a mode of 45 minutes of
instructional time for these courses. Language arts is and mathematics are taught daily in
each case-study school. The range and variance across mean, median, and mode reflect
the extra time allocated to language arts and mathematics at a number of case-study
schools.
143
Instructional time committed to science and to history-social science is identical
within each case-study school. The two subject areas will be addressed simultaneously.
The number of minutes dedicated to instruction in each of history and science classes
varies from 21 to 59 minutes daily. The average (mean) for each history and science
classes was 41 minutes with a median of 45 minutes. Two modes appeared: 21 minutes
and 45 minutes.
Instructional minutes in social studies and science reflect different schools’
approaches to instruction in these content areas. As designed by the state, both science
and history-social science are intended to be year-long courses. However, doubling up
instructional time in language arts and mathematics classes at some schools results in
diminished time for social studies and science. In some schools, these classes are taught
for one semester during a single class period or every other day during a double blocked
class session while language arts and mathematics receive a daily doubling of
instructional time. Some schools provide the same allocation of instructional minutes to
every core content class.
The Evidence-Based approach to school adequacy includes foreign language as a
core course. Although foreign languages are not considered core classes in California,
they are required for university admissions and are mentioned here (and again in the
section addressing elective courses). Findings with regard to instructional time dedicated
to foreign language include minutes of instruction varying from 45 to 59 minutes daily.
This results in a range of 14 minutes with an average (mean) of 46.6 minutes, a mode of
45 minutes, and a mode of 45 minutes. Five of the eight participating schools include
144
foreign language instruction in Spanish. The other three do not offer foreign language
courses. The level of instruction further varies with some students receiving high-school
level instruction and others being introduced to Spanish language and culture in a survey-
type course. Coincidentally, schools making the greatest student achievement gains in
2008 also offer foreign language classes.
School Resource Use
This section is organized into eight resource allocation/expenditure elements
associated with the Evidence-Based Model (Odden and Picus, 2008) based on research-
and evidence-based components of school resource use (Odden, Archibald, et al., 2003).
Each element contains results of actual allocations found in case-study schools compared
with corresponding resources recommended by the Evidence-Based Model for school
adequacy. Analysis of data consists primarily of averages and ratios of quantifiable data
to compare resource use across schools and with the Evidence-Based Model.
Core Teachers
Core teachers are defined as all certified classroom teachers providing instruction
in a school’s core subjects of reading/language arts, mathematics, history-social science,
and science. In the eight middle schools studied there was an aggregate of 287.9 FTE
core teachers. The resulting average student-to core-teacher ratio is 36.0 to one. Ratios
varied from a low of 27.0 students to one teacher to a high of 42.0 students to one teacher
in core classes. In 2008-2009 Degas IS and Warhol JHS respectively have student-
teacher ratios of 27.0 students to one teacher and 37.2 students to one teacher as a result
of using Quality Investment in Education Act (QEIA) funds to reduce class size.
145
The Evidence-Based Model suggests a student-teacher ratio of 25 students to one
teacher. Findings reveal that two schools (i.e., Degas & Van Gogh) nearly meet the
research-based adequacy recommendation for class size (Odden, Picus, et al., 2003). The
remaining six schools reflect a reality of substantially larger class sizes than those
associated with dramatic student achievement gains. The table below summarizes core
class size findings.
Table 4.6. Core Class Size Analysis
School Actual Core Class Size Actual-Model Comparison
Botticelli JHS 38.2 1.5 times larger than Model
Chagall JHS 42.0 1.7 times larger than Model
Degas IS 27.0 1.1 times larger than Model
Picasso MS 38.5 1.5 times larger than Model
Renoir IS 36.1 1.4 times larger than Model
Rodin IS 41.2 1.6 times larger than Model
Van Gogh MS 27.1 1.1 times larger than Model
Warhol 37.2 1.5 times larger than Model
Specialist and Elective Classes and Teachers
Specialist and elective teachers include those providing instruction in subjects
including art, drama, world languages, physical education, shop, computers, career and
technical education, AVID, and other non-core academic subjects. The Evidence-Based
Model suggests that in addition to core academic teachers, schools add another 20% of
elective and specialist teachers. This allocation suggestion is based on a six-period
school day in which each FTE teacher instructs for five class periods and uses the sixth
period for preparation, planning, and collaboration. The Model includes specialist and
elective teachers to ensure that all teachers have a free planning/preparation period during
the regular school day.
146
Participating middle schools all offer physical education, a requirement in
California. Examination of findings in the elective/specialist category reveals wide
variation in the quantity and type of elective courses available at each school. Spanish
language classes are included in this analysis of specialist/elective teachers as the courses
are not required, are not offered at all schools, and where available, are not taken by all
students.
The number of FTE elective teachers varies from 6.1 to 24.3. As mentioned
above, the Model suggests that the number of elective courses represent 20% of core
courses. Converting elective course quantities to percentages yields a range of 49% with
elective/specialist courses representing as little as 14% of a school’s core course offerings
to as much as 63%. The average (mean) percentage of elective/specialist courses in
relation to core courses was 36.9%. The median was 35.5% and the mode was 14%. The
table below summarizes the findings and analysis of elective/specialist courses as a
function of core class offerings.
Table 4.7. Elective/Specialist Courses as a Percentage of Core Courses
Evidence-Based Model Recommendation for
Elective/Specialist Courses
Average Actual Allocation as a Percentage of
Core Courses
20% of Core Courses 36.9%
(Range: 14% to 63%)
Includes physical education, Spanish language,
AVID, and intervention courses as well as shop,
arts, consumer and home economics, technology,
multicultural awareness, leadership, and other
courses.
In addition to physical education and scattered Spanish language offerings,
electives included wood shop, instrumental music, art, home economics, leadership,
journalism, dance, and AVID. Seven of the eight case-study schools include AVID
147
classes. All eight schools included language arts and/or mathematics intervention
courses. At some schools, these were part of the elective course menu. At others, they
were offered in lieu of an elective course.
Degas, the school with the fewest and lowest percentage of elective courses offers
AVID to students who do not need language arts intervention. At the other end of the
spectrum, Warhol and Chagall, at 16.3 and 17.4 FTE specialist/elective teachers
respectively, include Spanish language, music, drama, technology, career-technical
education, AVID, leadership, journalism, consumer education, wood shop, speech,
conflict resolution, and multicultural awareness courses. Warhol’s intervention program
includes zero-period classes to ensure that students’ schedules include time to participate
in elective classes during the school day.
Findings in the area of elective/specialist course offerings reveal differing
approaches to meeting students’ academic and developmental needs. Elective time has
been pre-empted in some cases by school-mandated intervention classes to support
struggling learners. This creates a challenge in accurately counting FTEs in this area as
some schools count interventions, even if mandated, as electives, while others use
intervention courses in lieu of core courses in language arts and/or mathematics.
However, the paradigm of students attending more classes than teachers teach continues
to prevail at seven of the eight schools included in this study (one school has
implemented a common before-school prep period for all teachers).
148
Librarians
Librarians are certified teachers who attain an additional credential and
demonstrate expertise as specialists in both serving as quality filters and in teaching
students to utilize the highest-quality information sources both in print and electronic
media. The prototypical school of 450 students described by the Evidence-Based Model
includes one librarian. A range of library-related positions requiring differing levels of
certification was identified during the data collection process. These positions include
librarians, library media specialists, media/library aides, and technology support.
The total number of librarians from the eight schools was reported at 2.2 FTEs.
The average (mean) number of FTE certified librarians was 0.3 per school. The median
was 0.15 and the mode was zero. Four schools have no librarian.
Schools employed more media specialists (4.6 FTEs), a classified staff position
requiring no credential, than librarians. Here, the mean number of FTE media specialists
was 0.6 with a median of 0.8 and a mode of zero. Three schools have no media
specialist.
Library aides, typically hired to shelve and code books, were found at two
schools. The mean number of FTE library aides was 0.1 and the median was zero.
None of the case study schools employs a technology coordinator; however, six
schools employ technology assistants to troubleshoot computer problems and provide
basic maintenance. The master plan for technology is part of each school’s plan for
school improvement and is a product developed by site (and sometimes district) leaders
along with school site council members. Botticelli, previously a one-to-one laptop
149
school, has taken a step back to pilot instructional use of laptop computers with a smaller
group of students and teachers. The principal discovered that the school will require a
full-time technology coordinator in the future to support the full staff in using computer
technology with the intention of attaining student achievement gains. Table 4.8
summarizes the analysis of resource allocation toward library and technology staffing.
Table 4.8. Library Staffing Analysis
Staffing Category Evidence-Based Model
Recommendation
Actual Staffing Level in FTEs
Librarian 1 per prototypical middle
school of 450 students
Aggregate recommendation
for case-study schools: 20.8
Aggregate across 8 schools: 2.2
Average (Mean)= 0.3
(Range: 0 – 1)
Present in 50% of schools.
Part-time librarians paired with full-
time media specialists in two schools.
Media Specialist Recommend a certified
librarian
Aggregate across 8 schools: 4.6
Average (Mean) = 0.6
(Range: 0 – 1)
Present in 63% of schools.
Paired with a library aide in one school.
Paired with part-time librarians in two
schools.
Library Aide Recommend a certified
librarian
Aggregate across 8 schools: 0.9
Average (Mean) = 0.1
(Range: 0 - 0.5)
Present in 25% of schools.
The only library support in one school.
Technology Recommend a certified
librarian
Aggregate across 8 schools: 1.5 FTE
classified tech. troubleshooters
Present in 75% of schools.
Chagall JHS attributes some of its 45 point API gain in 2008 to a school-wide
pleasure reading program requiring frequent use of the school library. Both Chagall and
Renoir boast having the highest circulation rates in their districts. Their librarian-to-
student ratios are one librarian to 4,530 students and no librarian to 1,738 students
respectively. However, Chagall employs a media specialist, creating a library staff-to-
150
student ratio of one staff member to 1,045 students. Renoir’s 0.9 FTE media specialist-
to-student ratio is one library staff member to 1,931 students. Rodin, the smallest case-
study school, is the only school with a full-time librarian. Even when combining all
library staff, no school approaches the library staff allocation recommended by the
Evidence-Based Model. Ratios of librarians to students and of all library staff combined
to students reflect the findings in this area and are summarized in Table 4.9.
Table 4.9. Librarian- and Library Staff- to-Student Ratios
Evidence-Based Model
Librarian-to-Student Ratio
Average Librarian–to–Student
Ratio
Average Combined Library
Staff–to–Student Ratio
1 Librarian to 450 Students 1 Librarian to 4252.7 Students
Four schools have zero librarians.
Three schools have part-time
librarians.
One school has a full-time librarian
with a ratio of one librarian to 652
students.
1 Library Staff Member to
1359.5 Students
Staffing variation:
0.5 Library Assistants to 1.4
combined part-time Librarian
(0.4) and Media Specialist (1).
The data indicate that on average, case-study school libraries are staffed with
librarians at roughly 9.5% of the Evidence-Based Model recommendation. When
factoring in classified media specialists and library assistants, the staffing allocation is
one-third that recommended by the Model. At Van Gogh MS, teachers agreed to forego
half their prep time in part to fund the school’s media specialist and library aide positions.
The principal at Picasso MS indicated that parent volunteers assist the school’s media
specialist to keep the library adequately staffed. This area is resourced at a substantially
lower level than that recommended by the Evidence-Based Model even in schools that
rely heavily on the school library to instill good reading habits as part of their school
improvement plan.
151
Extra Help for Students
Extra help consists of teacher tutors, special education teachers, extended day and
summer school programs, and English language development teachers. The Evidence-
Based Model identifies English learners and FRPL participants as proxies for students
likely to be in need of extra assistance in quantifying adequate resource allocations. The
Model is informed by research indicating that one-to-one tutoring from a teacher is the
most effective mechanism for helping struggling students (Odden, Picus, et al., 2003).
Best practice with regard to assisting English learners in attaining English language and
subject-area proficiency includes additional teachers to support their unique language and
academic needs. Additionally, the model suggests three special education teacher
positions for a middle school of 450 students for students with mild and moderate
learning disabilities. The model places responsibility for allocation of staff for services
for severely disabled students into the hands of state and federal government. This
section is sub-divided into specific Model recommendations for extra help: tutors, EL
support, special education, alternative education programs, and extended day and summer
school.
Tutors. Tutors provide immediate one-on-one or small-group support to students
when they struggle in a core content class. Furthermore, tutors are defined as certified
teachers without student rosters who are available to students in need. By allocating
teacher tutors, schools commit to proactive measures that prevent students from falling so
far behind their classmates that they cannot catch up. A meta-analysis by Wasik and
Slavin (1993) reported an effect size of approximately 0.75 for one-to-one tutoring
152
programs, and similar analyses by Mathes and Fuchs (1994), Shanahan (1998), and
others resulted in identifying effect sizes ranging from 0.4 to 2.5. These effect sizes
highlight tutoring as a powerful research-based strategy that supports student
achievement.
The Evidence-Based Model suggests allocating one certified tutor for every 100
FRPL program participants (economically disadvantaged) in a school. The eight sample
schools serve 7,342 students who participate in the FRPL program. If allocated based on
the population of economically disadvantaged students at each of the case-study schools
and in accordance with the Model, an aggregate of 73.3 FTE teacher tutors would work
with students individually and in small groups up to five students across the eight case
study schools. Renoir’s principal indicated that the Title I coordinator and the bilingual
coordinator, both certified teachers, each spend approximately five percent of the day
providing tutoring to students on a pull-out basis. No other school allocates resources
toward teacher tutors. However, four of the case-study schools employ college students
as part-time (10 hours per week) AVID program tutors. In this case, college tutors work
only with AVID students on homework, projects, and studying for tests. Roughly 2.3
FTE college tutors work at four schools within the study and earn between $10 and $12
per hour. The Model makes no recommendation for non-credentialed tutors.
On-campus suspension teachers. This category reflects the possible use of funds
to support students who otherwise might be sent home and miss instructional time as a
result of violating school and district policies. Three case-study schools utilize (or plan to
implement) an on-campus suspension program. The Getty Union High School District
153
provides Warhol JHS with a substitute one day per week for the school’s on-campus
suspension program. The school allocates an instructional aide to support the teacher. At
the time of the interview, Chagall’s on-campus suspension program consisted of spending
the day in the principal’s office. However, the district was two or three weeks away from
opening the doors to its “ASSIST” program. This full-day program was scheduled to
include computer-based learning experiences and counseling for suspended students.
This program investment was intended to provide a needed service as well as regain
previously lost funds due to absences when students were sent home on suspension. A
third school uses two aides to supervise on-campus suspension.
To sum up, a 0.2 FTE substitute teacher and 1.1 instructional aides are allocated
across two of the case-study schools to provide an instructional experience for students
who have misbehaved. Another school was awaiting the opening of a district program
that centralizes this function.
Title I teachers. Each of the participating schools receives Title I funds. The
Model is informed by the allocation of resources from this funding source. Ten teacher
positions across the eight case-study schools are possible due to Title I funds. These
positions included reading intervention and Step Up to Writing program and regular
classroom teachers.
Title I Aides. The Evidence-Based Model makes no recommendation for
instructional aides as research findings associate student improvement gains with teachers
and smaller class sizes as opposed to use of instructional aides. However, current
instructional practice often includes use of such personnel. In the case-study schools,
154
Title I funds were used to support 9.5 FTE aides, most of whom were special education
instructional aides. One school used one FTE aide for bilingual and office support. Use
of Title I funds to support the special education program highlights the challenge of
special education program requirements that encroach upon resource availability for other
needy students.
English language development (ELD) teachers. For every 100 English learner
students, the Evidence-Based Model funds one teacher to support language development.
Based on the number of English learners enrolled in each of the eight schools in this
study, the Model generates an allocation of 59.0 FTE ELD teachers. Actual allocation
was 9.6 teachers directly supporting language development of English learners.
Counting teachers in this area proved challenging for two reasons. One, two
schools officially do not offer ELD classes (as per district policy). Instead, they provide
instruction in reading intervention with placement determined by diagnostic assessments.
Struggling readers at a given level of proficiency, whether due to English language
fluency, specific learning disability, or other reasons, are placed in a reading intervention
class together. Two, schools across the study provide differentiated language arts classes.
Some, represented by the 9.6 FTEs identified in the study, name one form of
differentiated class “ELD.” Others use names such as “Basic English 7” and “Advanced
English 7.” Students enrolled in the Basic class typically, but not necessarily are English
learners.
All but one school employs roughly twice as many language arts teachers as
counterparts in other core disciplines. The single school that distributes core teachers
155
evenly also designates ELD courses in addition to the language arts allocation. In the
seven schools with approximately double the number of language arts teachers, English
learners enroll in multiple class periods of differentiated language arts and reading
instruction. The increase in the number of language arts course sections has been
accomplished by cutting social studies and science instruction in half. The overall
number of teachers is not influenced by the presence of English learners.
ELD aides. A total of 7.2 ELD aides is assigned to support learners in the six
schools offering ELD instruction. The Evidence-Based Model makes no
recommendation for instructional aides, recommending instead that fiscal resources be
directed toward hiring teachers to reduce class sizes.
Gifted and talented education (GATE/honors). All eight case study schools
provide advanced level courses in language arts and mathematics. Seven include social
studies and science in their advanced course options.
Examples of honors and advanced choices include the Corporate Academy
offered at Degas. Here, proficient and advanced students participate in an enriched
program including opportunities to connect school to careers through job shadowing,
learning about different careers and educational requirements to pursue them, and
focusing on basic business success skills ranging from writing a business letter and
resume to dressing for success. Both Rodin and Renoir provide humanities courses in
which language arts and social studies are taught in two-period blocks in an
interdisciplinary fashion. The other five schools provide courses labeled “honors” or
“advanced” in the four core subjects. Student placement is determined by factors
156
including CST proficiency levels, previous year’s grades, and at Rodin, by student self-
selection. Although districts attempt to identify elementary-age English learners for
GATE services, at the secondary level, CST proficiency levels have the potential to limit
participation in advanced courses.
The Evidence-Based Model suggests an expenditure of $25 per student to uncover
hidden talents and fund enriched learning experiences for low-income and/or culturally
diverse high-ability learners (Odden & Picus, 2008). Based on combined enrollment of
9,356 students across the eight schools, the Model generates a funding amount of
$233,900. This compares with zero dollars allocated to GATE services at six case-study
schools. Warhol and Picasso, both of which receive GATE funding from their districts,
received a total of $31,594 for the 2008-2009 school year. Warhol’s funding equals
$5.23 per student while Picasso’s equals $23.46 per student. Of the eight case-study
schools, one approached the support level suggested by the Model.
Special education staffing. The Evidence-Based Model distinguishes between
resource allocations for severely handicapped students in self-contained classrooms from
that for students with mild-to-moderate specific learning disabilities. With regard to
severely handicapped students, the Model recommends that the state reimburse for all
necessary services, effectively removing the burden of funding for this high-cost budget
item from school districts. The Model generates funding for mild-to-moderate special
education teachers at a rate of one teacher for every 150 students in the school.
California utilizes a census-based approach to special education funding in which
the state allocates a fixed dollar amount per student in a district using average daily
157
attendance (ADA) as the student count. These funds are channeled through Special
Education Local Plan Areas (SELPAs). The intention of Assembly Bill 602, which
resulted in the census approach in California in 1997, was to equalize special education
funding from the state source of support for the approximately 11% of students across the
state identified with conditions ranging from mild speech disabilities to serious medical
ailments requiring far more than instructional assistance (Romeo, 2007). In developing
the Evidence-Based Model, Odden and Picus (2008) followed a census approach to
special education funding, resulting in a calculation of three special education teachers
for every 450 students. This reflects the current Response to Intervention trend (Tilly,
2006) toward addressing the needs of all struggling students with an expectation that
most academic challenges can be met without special education referrals and
accompanying resource restrictions.
Actual mild-to-moderate special education teacher allocations included 11.6
inclusion teachers across the eight case-study schools and 19.5 special education resource
room teachers. The Model suggests 62.4 FTE mild-to-moderate special education
teachers across the case-study schools. At 31.1 FTE mild-to-moderate special education
teachers, actual allocations are roughly half that of the Model.
The approach to inclusion varied somewhat across case-study schools. Chagall
offers a “Reteach” period in which struggling students, regardless of special education
status, go to catch up while under the eye of a special education teacher for informal
observation and determination of possible educational disability. At Van Gogh resource
specialist teachers and their aides partner with mathematics teachers to reduce the
158
student-to-teacher ratio. Mild-to-moderately disabled special education students are
placed in mathematics classes during a block that straddles the lunch period. All students
in these classes receive direct instruction before lunch in the regular education classroom
and individual/small-group support after lunch. Van Gogh also uses special education
teachers to teach its reading intervention classes to all students meeting placement
requirements. Finally, Botticelli, Degas, Warhol, and Renoir special education teachers
partner with content-area teachers where they assist and/or co-teach.
The Evidence-Based Model suggests that services for students with severe
learning and other disabilities, the approximately 25% of special education students
whose instruction and care incurs the greatest cost, be fully reimbursed by the state.
Recognizing the unique needs of students in this category, the Model makes no allocation
recommendation.
Actual allocation of teachers for severely disabled students includes a total of 29.3
FTE teachers with a range from two to seven FTE teachers per school. The teacher-to-
student ratio in this category ranges from one teacher to 7.7 students to one teacher to
28.7 students. The average (mean) stands at one teacher for every 13.3 students, with the
median at one teacher per 11.6 students across the case-study schools.
Each school employs special education aides in self-contained, inclusion, and
resource rooms. Self-contained classrooms include 54.4 FTE aides, identified as
instructional aides, but in actuality perform support functions for medically fragile
students. Four schools, Degas, Van Gogh, Warhol, and Renoir, employ a total of 9.3
FTE special education inclusion aides. These staff members work with all students in
159
classes in which special education teachers collaborate with regular-education, subject-
specialist colleagues. Each school employs special education resource room aides
assigned to specific resource specialist teachers. Actual allocation was 7.1 FTE special
education resource room aides.
The Evidence-Based Model makes no recommendation nor generates resources
for instructional aides based on small class size research findings indicating that teachers,
not aides are more likely to influence student achievement. The Model is silent on self-
contained classroom aides as they provide necessary support for the 25% of special
education students with unique needs addressed by special education and medical
specialists. At 61.7 FTEs, special education aides reflect a category of student
instructional assistance over-resourced in relation to the Model in contrast with the
earlier-identified allocation of special education teachers at roughly half the model’s
recommendation.
Extended day. Research findings on high-quality extended day programs that
support student achievement highlight such features as “targeting of high-poverty
children and youth; services at least four days a week for most of the afternoon;
coordination with nearby schools that serve their participants; a trained staff; and
enrichment and recreational activities and academic supports” (Vandell, Pierce, &
Brown, 2002). The Evidence-Based Model generates a school funding element based on
international research findings that provision of additional learning time outside the
school day supports student proficiency in the U.S. (Darling-Hammond, Wood, Glenn, et
al., 2008) and abroad (Education Commission of the States, 2005). The Model suggests
160
that schools provide services based on a calculus including one teacher per 15 students,
with the initial number of participating students determined by halving the number of
FRPL program participants.
If resourced as per the Model, approximately 3,670 students attending the eight
schools would be the foundational number for calculating the number of teachers. Data
collected indicate that roughly 980 students participate in after-school programs offered
by seven of the eight schools. Additionally, unknown numbers participate in programs
provided at the various school sites by external providers such as Boys and Girls Clubs
and non-profit community organizations in partnership with the school districts. The
schools do not collect data on attendance at programs not provided by them, and it was
beyond the parameters of this study to obtain such data.
The average (mean) extended-day time allocation across the case-study schools is
322.5 minutes daily. Programs include both voluntary and mandatory before- and after-
school homework assistance, tutoring, and enrichment opportunities. Mechanisms for
identifying students to participate include benchmark assessment scores, failing grades at
the quarter, and failure to complete assignments/homework.
For the 3,670 students attending the eight schools, the Model suggests an overall
total of 61.2 FTE extended day teachers. Actual allocation is 11.3 FTE teachers with an
approximate one-teacher-per-21.7 students ratio. The Model makes no recommendation
regarding extended day classified support. Two schools include instructional aides to
support extended-day students: Botticelli uses a special-education aide three hours per
161
week for its mandatory after-school “Academic Empowerment Club” and Chagall
includes one aide six hours per week using Title I and EL funds.
Principals voiced varying opinions of the externally provided programs on their
campuses. Renoir’s principal indicated a sense of lost opportunity with its program
because of program management, accountability, and areas of emphasis that diverge from
school goals for student achievement. At the other end of the spectrum, principals at
Picasso and Warhol cited strong and positive partnerships with the agencies that provide
after-school academics and recreation on their campuses. Warhol’s principal shared that
the school provides the city program with school instructional materials and together
have crafted systems to ensure that the same behavioral expectations for students hold
true from the moment students step onto campus until they leave at the end of the day.
Degas depends on the city program for all its after-school academic support. Principals
collectively indicated that the number of students participating in non-school after-school
programs on campus far exceeded the number participating in school-sponsored
programs.
Summer school. Another means of increasing time in support of attaining
proficiency goals is to extend the school year. Research findings indicate that without
summer school, economically disadvantaged students begin the school year about a
month behind the point at which they left off in the spring (Borman & Boulay, 2004).
Based on evidence pertaining to effective summer school programs, the Model suggests
an eight-week program with four hours of instruction daily. This would provide students
with 9,600 minutes of summer instruction.
162
Actual summer programs vary in length of instructional day and duration of
session. The average session length across the eight case-study schools was 5,848
minutes in comparison with 9,600 minutes suggested by the Model. Duration of sessions
ranged from one school’s three-week session to two schools with six-week sessions.
Table 4.10 illustrates the range across the eight case study schools.
Table 4.10. Summer School Instructional Minutes and Session Length
Model
Recommendation
School Minutes of Instruction
per Session
Length of Session
Botticelli 4,800 4 weeks
Chagall 7,200 6 weeks
Degas 8,100 6 weeks
Picasso 4,332 19 days
Renoir 6,750 5 weeks
Rodin 4,800 4 weeks
Van Gogh 3,600 3 weeks
(2 3-week sessions)
Eight week session,
four hours of
academics daily (9,600
minutes), 1:15 teacher-
student ratio
Warhol 7,200 6 weeks
Summer school staffing recommendations mirror those for extended day
programs: overall enrollment calculated by halving the number of FRPL participants at
each school and further dividing by 15 to determine the number of teachers. To translate
the number of summer school teachers into FTEs, the calculation requires further
dividing the number of teachers by four (as the Model’s suggested summer school
allocation reflects one quarter of the school year). Based on the number of FRPL
students at each of the case-study school, the overall Model recommendation for the eight
case-study schools was 61.2 FTE teachers.
Challenges associated with collecting summer school data included the
centralized nature of summer school administration. Data regarding student attendance
and staffing allocations typically are not provided to principals. Without engaging
163
district office employees to perform data sorts not used by the district, school- and
district-level data tended to be lumped into one document, making the nuanced data
desired difficult to collect. Six schools of the eight were able to provide precise and
specific summer school data regarding both teachers and students. As a result, they serve
as a focus for examination of resource allocation in this area.
According to the Model, Botticelli, Degas, Picasso, Renoir, Rodin, and Van Gogh
would expect summer programs serving a total of 2,441 students (half of their combined
FRPL program participants). A total of 40.6 FTE teachers would teach summer school.
The actual combined teachers across these six schools equaled 17.5 FTEs and 3,063
students attended. Attendance rates were 1.3 times higher than the enrollment anticipated
by the Model’s designers. However, summer school was staffed at 43% of the suggested
rate. A total of 3.5 FTE office managers and special education aides completed the
summer school staff across the eight schools. These data represent only “regular”
summer school programs as each district operates summer programs for special-
education students separately.
Summary. Extra support services are intended to minimize unnecessary
identification of students for special education services and to provide the extra time and
assistance needed by students who struggle as a result of English fluency levels,
economic status, or other factors associated with school readiness. Case-study schools
provided extra support via school-sponsored and external-agency-sponsored before- and
after-school academic programs. Seven of the eight schools either sent students to or
provided an on-site summer-school program. One school indicated that, due to budget
164
cuts, the district had eliminated summer school for its middle-school students, as it was
not necessary for graduation. By and large, tutors were limited to part-time college
students providing tutoring to students in schools’ AVID programs. In all but one school,
AVID students had to meet minimum language-arts proficiency levels in order to free up
a class period during the school day to take the AVID class. As a result, AVID tutors
tend to work with the more proficient students across the case-study schools.
Both summer school and extended day programs were staffed at levels
substantially below those reflected by the Evidence-Based Model. Additionally, summer
school programs did not meet the eight-week length expectation noted in the Model. It is
unclear how many students are actually served by the variety of extended-day programs
available or precisely how many students attended summer school across each of the
eight schools.
All eight case-study schools incorporate a combination of inclusion, intervention,
content, and resource-room classes to support mild-to-moderately disabled special
education students. This includes severely disabled students attending content classes
with resource specialist teachers, as they are able. While the Evidence-Based Model does
not allocate resources toward instructional aides, each of the eight case-study schools
relies on these employees, especially in special-education classrooms.
The table below illustrates the extra support services provided across the eight
schools included in this study.
165
Table 4.11. Extra Help for Struggling Students
Extra Help
Category
Category Elements Model Recommendation Actual Resource Allocation Comments
Tutors Teacher Tutors 1 per 100 FRPL program participants
(73.4 FTEs)
0.1 FTE
(0.13 of Model Recommendation)
0.05 of Bilingual
Coordinator and Title
I Coordinator jobs at 1
school.
Non-Teacher Tutors No recommendation 2.3 FTE Part-time college
students tutor AVID
students at 4 schools.
English Learner
Support
Teachers 1 per 100 EL students
(59 FTEs)
9.9 FTEs
(0.17 of Model Recommendation)_
ELD classes not
offered at all case-
study schools –
replaced with reading
intervention classes.
Aides No recommendation 7.2 FTEs Funded through Title I
and Bilingual
Education funds.
Gifted & Talented
Education
GATE Funds $25 per student
($233,590)
$31,594.00
(0.14 of Model Recommendation)
2 schools receive
funds.
Warhol: $5.23 per
student;
Picasso: $23.46 per
student.
166
Table 4.11 continued
Extra Help
Category
Category
Elements
Model Recommendation Actual Resource Allocation Comments
Special Education Mild-to Moderately
Disabled Special
Education Teachers
3 teachers per 450 students
(62.4 FTEs)
31.1 FTEs:
11.6 Inclusion Classroom; 19.5
Resource Room
Average (Mean) Teacher: Student
Ratio: 1:301 (0.5 of Model
Recommendation)
Multiple inclusion
models including co-
teaching/providing
support in content
classes & teaching
study-skills and
reading intervention
classes to all students
in need.
Severely Disabled
Special Education
Teachers
As needed
Full reimbursement by the state
29.3 FTEs
Average (Mean) Teacher:Student
Ratio: 1:13.3
Substantially higher
teacher-student ratio
at one school (1:28.7).
Aides in Self-
Contained Special
Education
Classrooms
As needed
Full reimbursement by the state
54.4 FTEs Referred to as
“instructional aides,”
but also support basic
functions and needs of
medically fragile
students.
Aides in Special
Education Resource
Classrooms
No recommendation 7.1 FTEs Most are funded
through Title I funds.
Aides in Inclusion
Classrooms
No recommendation 9.3 FTEs Accompany special
education teachers to
provide support and
increase adult-to-
student ratio.
167
Table 4.11 continued
Extra Help
Category
Category
Elements
Model Recommendation Actual Resource Allocation Comments
Extended Day Students Anticipate enrollment of 3,670 students
(1/2 FRPL)
Before/after school program
enrollment: 980 Students (0.3 of
Model Recommendation)
All 8 schools also
have on-site before-
&/or after-school
academic & recreation
programs provided by
non-profit or city
agencies that serve
large, but
undetermined
numbers of students.
Teachers 1 teacher per 15 students based on
anticipated enrollment of 3,670 students
and overall time equivalent to ¼ of a
school year
(244.7 teachers; 61.2 FTEs)
11.3 FTEs (0.2 of Model
Recommendation)
7 of 8 schools provide
before &/or after
school voluntary or
mandatory
tutoring/homework
assistance/computer
lab time.
Classified Staff No recommendation 0.4 FTEs Aides at 2 schools.
Funded through Title
I, California High
School Exit Exam
support & Bilingual
Education funds.
Extended Year Summer School
Time
9,600 minutes of academic instruction
over 8 weeks
Average (mean): 5,847.8 minutes of
academic instruction (0.6 of Model
Recommendation)
Session length ranged
from 3-week sessions
to 6-week sessions.
168
Table 4.11 continued
Extra Help
Category
Category
Elements
Model Recommendation Actual Resource Allocation Comments
Extended Year Enrollment Anticipate enrollment of 3,760 students
(1/2 FRPL)
3063 students across 6 schools
participating in summer school (1.3
x higher than Model
Recommendation)
No summer school for
students at one school
due to budget cuts.
One school provided
its own summer
program for credit
recovery in addition to
district’s program for
special education
students.
6 of eight schools
participated
exclusively in district-
sponsored summer
school programs.
Teachers 1 teacher per 15 students based on
anticipated enrollment of 3,670 students
and overall time equivalent to ¼ of a
school year
(244.7 teachers; 61.2 FTEs across 8
schools)
(42.1 teachers; 10.5 FTEs for 632
students at Botticelli and Van Gogh –
representative schools elaborated upon
in the next column)
31 teachers at Van Gogh and
Botticelli for 663 students and
representing 4.5 FTEs (reflecting
length of session & # of minutes
relative to school-year contract
minutes)
17.5 FTE teachers overall at 6
schools with data: 0.43 of Model
recommendation
These data reflect
summer school
teachers at 2 of the 8
case-study schools
due to district-related
issues with accurate
data collection.
Classified No recommendation 3.5 FTEs Office managers and
special ed. aides
169
Professional Development
School-improvement research points to the importance of effective professional
development to support high-quality teachers who provide students with excellent
learning experiences. To paint a picture of professional development across the eight
case-study schools, principals provided data regarding the quantity and type of
professional development for their teachers as well as resources allocated to this school
improvement element. The Evidence-Based Model includes resource allocation
recommendations for specific quantities, types, and levels of intensity for professional
development. This section contains both presentation of data and analysis including
comparison with the suggestions generated by the Model.
Professional development time. Six of the eight case-study schools include three
days of professional development time in the teacher contract. One includes two days.
Another contract provides for four professional development days, but the school has
opted to take three in order to use the fourth for teaching. Contracted professional
development days focus on district-determined topics combined with school-level data
analysis and goal-related learning and planning time. The Model’s time allocation and
intensity recommendation is 10 days of intensive professional development offered
during the summer.
In addition to their pupil-free days, case-study principals identified places within
the regular school day for teachers to engage in professional collaboration, data analysis,
planning, and learning. As illustrated in Table 4.12, six of the eight case-study schools
include weekly, organized professional development time focused on student
170
achievement goals and the school improvement plan. This ongoing time reflects one of
the elements linked with dramatically improving schools: professional learning
communities. It also encompasses additional planning and preparation time throughout
the school year for professional development and support as needed.
Table 4.12. Plans for Regular School Day Professional Collaboration and Learning
School Professional Collaboration & Learning Scheme
Botticelli
Degas
Picasso
Renoir
Rodin
Van Gogh
Weekly early-release or late-start day for students to facilitate school-wide
professional development – approximately 1 hour per week.
Warhol Common prep periods for department members.
Chagall One department with common prep periods.
Release time provided to teachers for collaborative planning as requested.
Instructional facilitators/coaches. Diciembre (2002) and Elmore (2002) provide
research-based reminders of the significance of using expert teachers as facilitators in
developing the expertise of their colleagues. Instructional facilitators provide a number
of services including lesson and strategy modeling, classroom observations, planning
assistance, and pre- and post-lesson conferencing. Garet et al. (2001) highlight the
elements of effective professional development inherent in coaching: time, coherence
with instructional objectives, and active/applied learning for the participating teacher.
Grounded in evidence that connects improving schools with instructional facilitators, the
Evidence-Based Model suggests placement of one coach for every 100 FRPL
participants. Based on this calculation, the Model generates a total of 73.3 instructional
coaches based on numbers of economically disadvantaged students at each of the case-
171
study schools. Actual allocation of certificated teachers to this position is 2.1 FTE
teacher coaches in three schools. Van Gogh staffs a 0.5 FTE “Student Achievement
Teacher” who serves as a resident expert in the school’s language arts program. At the
time of the interview, Degas employed a 0.4 FTE math coach and was in the process of
adding a 0.8 FTE language arts coach. At Warhol, each department chair has two prep
periods, one of which is dedicated to conferring with colleagues whose students are
falling behind. Four principals cited a common desire to have the resources to employ
(additional) coaches, including Warhol’s principal desiring a personal leadership coach.
Consultants. Four of the eight case-study schools utilize outside consultants for
on-site professional development and other school improvement efforts. The Hermitage
School District provides Picasso Middle School with $6,000 to pay for work with a
professional developer/consultant representing Targeted Leadership and hired to work
throughout the district. Botticelli JHS plans to spend $96,000 to bring Action Learning
Systems and Choice of Action to provide professional development for teachers
throughout the school. Warhol JHS has hired a language arts consultant to provide
ongoing professional development for that department. Degas contracted with the county
office of education for professional development in language arts instruction, Positive
Behavioral Interventions and Supports (PBIS), a student behavior/school climate
professional development series, and principal coaching. Additionally, Degas retains a
contract ($89,000) with UCLA’s School Management Program (SMP) for training in and
support with the SMP’s classroom walkthrough protocol by which staff members engage
in observations of colleagues to improve instruction.
172
Tuition and conference-related fees. Registration and other fees ranged from no
planned expenditures to $25,000 planned for the 2008-2009 school year. Two schools
indicated that conferences and trainings have been eliminated for this year as a result of
budget constraints. Each of the seven AVID schools sent its teachers to the annual
summer training conference in San Diego. Picasso’s principal is committed to sending
each teacher to at least one conference annually. The per-student expenditure for teacher
professional development ranged from zero dollars to $14.52. The average (mean) was
$5.25 with a median of $7.75.
Travel-related funds for conference and training attendance varied from zero to
$25,000. One principal commented that a team had visited a successful school in
Vancouver, BC, as part of its benchmarking and learning process. Other travel-related
expenditures were for more local events (i.e., within the state).
Substitute teachers. Each of the eight principals indicated that teachers would be
released from classroom instruction for a variety of professional development
opportunities during the school year. Such opportunities include working with on-site
consultants and trainers as well as attending off-site conferences and training events.
Funds allocations range from $4,200 to $20,000 for the 2008-2009 school year. The
average (mean) planned expenditure for substitute teachers is $6,150 with a median and
mode of $5,000. Two schools had no dollar amount allocated for this resource. Of these
two, one principal indicated that funds would be made available for approved out-of-class
expenses (e.g., collaboration time) even though the district has frozen funds for
conferences, consultants, and trainings. Another principal indicated that substitutes
173
would be made available to release teachers for training events related to the school’s
improvement goals.
The Evidence-Based Model suggests a total of 10 days per teacher for substitutes,
a figure that captures both personal/sick days and professional days. Across each of the
eight case-study schools, the teacher contract includes 10 personal/sick days that carry
over and may apply toward retirement. These are separate from approved professional
development days for which substitute teachers are employed. As a result, all actual
school-year substitute allocations are in addition to the 10 days recommended by the
Model.
Professional development materials, equipment and facilities. When asked to
consider anticipated expenditures related to materials and equipment for on-site
professional development, principals provided a range from zero to $20,000. Most
anticipated providing photocopied materials to teachers (e.g., professional articles,
meeting schedules, data sheets). The average (mean) planned expenditure was $3,500
with a median of $1500 and a mode of $2000.
Administration of professional development. When asked to identify a level of
resource allocation toward administration of professional development, principals’
responses ranged from “nothing” to 0.2 FTE professional development administrators.
At one school the principal alone oversees and coordinates professional development,
while at six others the principal works with the leadership/curriculum team to orchestrate
this area. At the eighth school, the Title I coordinator’s duties include coordinating and
keeping track of professional development hours.
174
A general theme emerged in that principals did not associate resource allocation
with professional development administration. However, across the eight schools, some
combination of the principal/designee and/or a leadership team considers what might best
serve the staff in attaining its student-learning goals based on achievement data. Also,
two of the eight principals interviewed mentioned district personnel associated with
professional development. Each of the districts has a central office position with
professional-development oversight responsibilities. Each district runs a Beginning
Teacher Support and Assessment (BTSA) program and a Peer Assistance and Review
(PAR) program funded by the state. At least two of the districts provide a menu of after-
school training opportunities while each orchestrates mandated trainings (e.g., language
arts and mathematics materials and program trainings). Those district personnel involved
with professional development coordination are not included in the 0.05 average (mean)
FTE allocation toward professional development administration for each school, as
principals did not identify a connection between district professional development
offerings and their site-specific efforts to improve instruction.
Summary. Principals’ “wish lists” frequently included a desire to provide teachers
with more professional development and coaching. Three in particular noted the urgent
need to develop a cadre of “excellent” teachers with the capacity and flexibility to meet
their students’ needs. At the same time, professional development dollars seem to be
shrinking as districts freeze expenditures in this area. Schools across the study have
devised mechanisms for staff members to meet regularly for on-site professional
development during the school day and plan to send teachers to training events as
175
possible during the school year. Anecdotally, six of the eight principals shared that they
have reduced their use of outside experts as a result of the budget deficit. Principals did
not tend to tease out an allocation level associated with professional development
administration or identify the coordination of services provided by the district. Most
district services did not relate directly to school student achievement goals, but instead to
state mandates. Salient points regarding professional development are further
summarized and highlighted in Table 4.13.
176
Table 4.13. Allocation of Resources for Professional Development
Professional
Development
Category
Category
Elements
Model
Recommen-
dation
Actual Resource Allocation
Time Contract
Days for PD
10 Days of
Intensive PD in
Summer
Aggregate average (mean) of 2.9 days per teacher per school.
Aggregate average (mean) of 1.5 days prior to opening of instructional school year.
School-day
PD
As needed
• 6 schools: 1-hour weekly school-wide professional development time.
• 1 school: common prep periods for collaboration.
• 1 school: 3 data analysis days per year & substitute coverage for collective (e.g.,
departmental) PD as requested.
Inside
Expertise
Instructional
Facilitators
1 per 100 FRPL
program
participants
(73.4 FTEs)
2.1 FTEs
PD Providers No recommen-
dation
Teachers and administrators provide school-day and after-school training in topics including
data analysis, Step Up to Writing, & Proficient Paragraph.
177
Table 4.13 continued
Professional
Development
Category
Category
Elements
Model
Recommen-
dation
Actual Resource Allocation
Outside
Expertise
On-site PD
Providers
As needed Contracted Services at Case-Study Schools:
• Targeted Leadership
• Action Learning Systems
• Choice of Action
• O.C. Department of Education
• Language Arts experts
• Positive Behavioral Interventions & Supports
On-site
Consultants
As needed Degas: UCLA SMP
District-
provided PD
No recom-
mendation
• BTSA (mandated)
• PAR (mandated)
• Special Education – through SELPA
• Nutrition Network
• Mandated language arts & mathematics trainings
• GATE
• E-mail Etiquette
• District Budget Process
• Et cetera
Conferences/
Off-site
Training
As needed • Tuition: Average (mean): $5636.50 per school. Average (mean) per-student tuition ratio:
$5.25 spent per student on teacher professional development.
• Substitute Coverage: Average (mean): $6150 per school to pay for substitutes while teachers
attend trainings during the school year.
• Travel: Average (mean): $7437.50 per school with a range from $0.00 to $25,000.
178
Table 4.13 continued
Professional
Development
Category
Category
Elements
Model
Recommen-
dation
Actual Resource Allocation
Outside
Expertise
Other PD As needed Degas & Van Gogh: Mathematics grant – year-round PD with summer institute.
Materials/
Equipment/
Facilities
Average (mean): $3506.25 per school
Median: $1500.00
Mode: $2000.00
Administration
of PD
0.5 FTEs identified in 3 schools.
179
Student Support Staff
Counselors, social workers, nurses, community liaisons, attendance/dropout
prevention personnel, psychologists, speech/occupational/physical therapists, and
supervisory aides all are included in the category of student support. The Evidence-
Based Model generates resources to fund one guidance counselor for every 250 students
and one additional student support staff member for every 100 economically
disadvantaged (FRPL program participant) students. Summing up allocations for reach
of the eight case-study schools, this amounts to 37.4 FTE guidance counselors and 73.3
FTE additional student support staff members.
Allocation of human resources for student support roles identified above diverged
from the recommendations included in the Model. The following section focuses on
actual allocation of resources in the student support category.
Counselors. The student-to-counselor ratio varied across the case-study schools
from 326 to 1,926 students per counselor. The average (mean) ratio was 710 students per
counselor and the median was 458.3 students per counselor. The number of FTE
counselors varied from 0.5 at Picasso MS to four at Warhol JHS. These FTEs compare
with Model recommendations of 3.9 counselors at Picasso to 6.9 at Warhol. Table 4.14
illustrates these findings.
180
Table 4.14. Allocation of Guidance Counselors
School Model Recommendation
(Number of FTEs based
on 450:1 student-to-
counselor ratio)
Actual Allocation
(Ratio)
Actual Allocation
(Number of FTEs)
Botticelli 3.7 463.5:1 2.0
Chagall 5.4 453:1 3.0
Degas 4.7 391:1 3.0
Picasso 3.9 1926:1 0.5
Renoir 7.0 869:1 2.0
Rodin 2.6 326:1 2.0
Van Gogh 3.2 822:1 1.0
Warhol 6.9 430.5:1 4.0
Total 37.4 Mean: 710.1: 1
Median: 458.3:1
Total: 17.5
Mean: 2.2 per school
Median: 2.0 per school
Nurses and health assistants. The number of nurses varied from zero assigned to
Chagall JHS to a 0.7 FTE licensed vocational nurse at Picasso MS. A total of 3.4 FTE
nurses serve seven of the eight case-study schools. The eighth school contacts the central
office for a nurse when needed. Four schools employ various health assistants with roles
ranging from direct care for medical needs of special-needs students to clerks who
maintain school health records. A total of 3.1 FTE health assistants are employed across
four of the eight case-study schools.
Psychologists and social workers. School psychologists are present at each of the
case-study schools. Their roles consist primarily of testing students for special education
identification and assisting with the development of students’ individual education plans.
The number of FTE psychologists ranged from 0.5 to 1.5 with a total of 6.8 FTE
psychologists across the eight schools. None of the participating schools employs a social
181
worker. Two schools indicated that they currently or frequently host unpaid social work
interns who meet with student groups on campus.
Speech/Occupational/Physical therapists. Staffing in this category ranged from
zero at Renoir IS to 1.2 FTE speech therapists at Rodin IS. A total of 3.1 FTE specialists
provide speech therapy service across the eight schools. Principals unanimously
indicated that the district office provides occupational and/or physical therapists on an as-
needed basis. Occupational and physical therapists are variously district employees or
contracted by the districts.
Community liaisons/parent advocates. Five of the eight case-study schools
employ parent advocates or include that role within another job. The number of FTE
parent advocates varies from zero at three schools to 1.3 at Botticelli JHS. Principals at
three schools elaborated on this position. Van Gogh MS utilizes Title I funds to employ
one Vietnamese-speaking and one Spanish-speaking liaison for three hours daily to
communicate with parents in the community. Two classified positions, Social Services
Assistant and Healthy Start Coordinator, each contribute to the parent
advocate/community liaison role at Botticelli JHS. Renoir’s principal indicated that the
parent liaison role is a small portion of the attendance technician’s job responsibilities.
Attendance/dropout technicians. Case-study schools approach this role in a
variety of fashions. Four schools employ attendance technicians. Two include
attendance follow-up in the registrar’s job expectations. The other half of the Social
Services Assistant and Healthy Start Coordinator positions’ roles at Botticelli JHS is
focused on dropout prevention and attendance. Picasso MS allocates no resources toward
182
this position. Warhol JHS utilizes its Discipline and Attendance Review Team (DART)
consisting of two assistant principals, all counselors, the school psychologist, and the
special education case carrier (if appropriate) to address needs of extreme attendance and
behavior issues. The DART meets weekly for two hours and counselors manage most of
the follow-up. A total of 8.2 FTE attendance/dropout positions were identified across the
eight case-study schools.
Supervisory aides. The Evidence-Based Model suggests allocation of supervisory
aides at a rate of two per 400-500 students (calculated at 1 per 225 students). If allocated
as per the Model, 41.6 FTE supervisory aides would assist with such tasks as bus, break,
and lunchtime supervision as well as hallway monitoring and restroom checking
throughout the school day. This contrasts with an actual allocation of 7.9 FTE
supervisory aides across the eight case-study schools. Three case-study schools employ
no supervisory aides. The FTE non-instructional aides across the other five schools
ranges from 0.5 to 3.8.
Other student services. Two schools identified additional support-team members
outside the categories identified above. Chagall JHS has a 0.4 FTE Special Education
Program Specialist, a district office employee with school psychologist status who
provides oversight for school-level special education programs. Rodin IS shares an At-
Risk Advocate, a central-office employee, with its feeder elementary school. These two
positions equal 0.9 FTEs dedicated to student support services.
Summary. As illustrated in Table 4.15, case-study schools allocate roughly half
the student support resources recommended by the Evidence-Based Model. If resourced
183
in accordance with the Model, students would be surrounded by substantially more adults
throughout the school day and across the campus. The number of recommended
supervisory aides (41.6 FTEs) relative to actual allocations in this category (7.9 FTEs)
highlights a key point of contrast between the Model and real-life allocations. These
recommended aides are in addition to the 73.4 FTE student support personnel across the
eight schools.
Instead of supervisory aides, schools appear to allocate resources toward
instructional aides that assist teachers in the classroom. The various student support roles
encompassed in the Model appear to reflect the types of roles that actually exist on the
campuses of case-study schools. As will be revealed in the next section, some roles (e.g.,
attendance support and dropout prevention) are distributed across multiple expenditure
elements/job titles. For example, administrators, especially assistant principals, are
actively involved in the student support role of encouraging good attendance along with
other positive student behaviors.
184
Table 4.15. Student Support Resource Allocation
Category Category Elements Model
Recommendation
(FTEs)
Actual
Allocation
(FTEs)
Counselors 37.4 17.5 (47% of Model
Recommendation)
Student
Support
Staff Members
73.4 39.8 (54% of Model
Recommendation)
Nurses 3.4
Health Assistants 3.1
Psychologists 6.8
Speech Therapists 5.5
Physical & Occupational
Therapists
0.0 (contact district as
needed)
Community/ Parent Liaisons 4.0
Attendance/Dropout Technicians 8.2
Other 0.9
Supervisory
Aides
41.6 7.9 (19% of Model
Recommendation)
Administration
The administrative staff includes school administrators, secretarial and clerical
staff, and coordinators of school services. This combination of certificated and classified
front office and behind-the-scenes staff establishes and manage the school’s direction.
The Evidence-Based Model suggests that a prototypical middle school of 450 students be
staffed with one principal and 0.5 assistant principals. It further recommends two full-
time secretaries for a school this size. Each of the eight case-study schools enrolls more
than the prototypical middle school. As a result, all position recommendations are
increased proportionally and/or per alternative recommendations based on larger school
185
sizes. The presentation of findings that follows reflects current administrative staffing
levels across the eight case-study schools.
Principal and assistant principals. The Model recommends one principal and 0.5
assistant principals for a school of 450 students. For larger schools, the Model suggests
the addition of one FTE assistant principal for every additional 450 students (and
prorated based on a factor of 450 students).
Each of the eight schools employs one FTE principal accompanied by between
one and three assistant principals. Two schools have assigned between 0.4 and 1.05 FTE
more assistant principals than suggested by the Model. The remaining six are staffed
below the Model’s recommendation. The Model recommends a total of 16.8 FTE
assistant principals across the case-study schools compared with an actual aggregate
allocation of 15.5 FTE assistant principals. A summary of the findings is available in
Table 4.16.
Table 4.16. Allocation of Principals and Assistant Principals
School Number of Principals
(= Model
Recommendation)
Model
Recommendation –
Asst. Principals
Actual Allocation
– Asst. Principals
Botticelli 1.0 1.6 1.0
Chagall 1.0 2.5 2.0
Degas 1.0 2.1 2.5
Picasso 1.0 1.6 1.0
Renoir 1.0 3.4 3.0
Rodin 1.0 0.95 2.0
Van Gogh 1.0 1.3 1.0
Warhol 1.0 3.3 3.0
Total 8.0 16.8 15.5
Other administrators. Each of the case-study schools employs additional quasi-
administrative staff members. These include Title I coordinators, bilingual coordinators,
186
and department chairs responsible for budgets, legal compliance, testing, and leadership.
A total of 5.8 FTE “other administrators” complete the administrative compliment across
the eight schools.
Secretaries and clerks. As indicated above, the Model recommends two full-time
secretaries for a prototypical school of 450 students. It makes no recommendation for
clerical staffing allocations (lower-paying and shorter calendars than secretaries)
common in public schools. Based on the Model, student enrollment at each of the eight
schools would generate a total of 41.6 FTE secretaries. This compares with actual
allocation of 11.3 FTE secretaries and 15.8 clerical staff. No school exceeds the Model
recommendation for secretaries. However, Rodin IS, with 4.3 combined FTEs in
secretaries and clerks exceeds the Model’s recommendation of a total of 2.9 FTE
secretaries.
Renoir IS, the largest school in the study would employ 7.7 FTE secretaries if
resourced in accordance with the Model. With one 0.9 secretary and one 0.6 clerk, this
school reflects the greatest discrepancy with the Model. Warhol JHS, a California
Quality Education Investment Act recipient (special funding for selected decile 1 and 2
schools as per California Senate Bill 1133), employs two FTE secretaries and five FTE
clerks. With seven front-office staff, Warhol comes closest to matching Model
suggestions for resource quantity in this area.
Examination of the data reveals that schools regularly decide to employ a greater
proportion of clerical staff as compared with secretarial staff. Central offices calculate
numbers of various staff members for front offices, including a substantial number of
187
less-expensive clerks. The data in Table 4.17 highlight school front-office staffing
paradigms.
Table 4.17. Secretary and Clerk Allocations
School Model
Recommendation
(FTE Secretaries)
Actual
Allocation of
Secretaries
Actual Allocation
of Clerks
Combined
Allocation of
Secretaries &
Clerks
Botticelli 4.1 1.0 2.5 3.5
Chagall 6.0 1.8 1.7 3.5
Degas 5.2 1.8 1.5 3.3
Picasso 4.3 1.0 1.0 2.0
Renoir 7.7 0.9 0.6 1.5
Rodin 2.9 1.8 2.5 4.3
Van Gogh 3.7 1.0 1.0 2.0
Warhol 7.0 2.0 5.0 7.0
Total 41.6 11.3 15.8 27.1
Operations and Maintenance
Staffing allocations that support a clean, safe, and functional school include those
for school security and custodial staff. Findings for these positions indicate that school
security is managed in a variety of ways ranging from police officers to classified
employees who monitor the halls. Custodians were present in each of the case-study
schools. One principal commented that the school has lost one FTE custodian due to
budget cuts in the 2008-2009 school year.
School security. The Evidence-Based Model mentions school security staff to
support a safe and orderly campus. Actual resource use in this area includes 7.3 FTE
personnel ranging from city police officers permanently assigned to particular school
sites to classified employee campus supervisors/security support. Two schools enjoy the
services of police officers assigned to their schools (1.3 FTE school resource officers)
188
and four employ between 0.9 and three FTE campus security personnel. Two schools do
not allocate any staff beyond administrators and supervisory aides toward school security.
Custodians. Data collected include those regarding allocation of custodians to
keep schools clean and safe. A total of 28 custodians are allocated across the eight case-
study schools reflecting an average (mean) of 3.5 custodians per school. The average
(mean) student-to-custodian ratio is 322.5 students per custodian. The Model suggests a
comparable allocation (1 custodian per 325 students) with an additional 0.5 FTE
custodian to support extra-curricular event duties.
Student Activities
The evidence-based approach to school funding includes a $200 per-student funds
allocation for student activities: sports, clubs, and other extra-curricular activities. The
intent is that every child be provided with an equal opportunity to develop skills and
talents.
Data collection consisted of ascertaining the student activities available across the
eight case-study schools. School-sponsored activities range from a lunchtime soccer
league at Degas IS to marbles and magic club at Picasso MS. Additionally, each school
includes at least one leadership class section during the school day and some include a
journalism/yearbook class. A compilation of school activities in Table 4.18 assists in
identifying focal points and in some cases, mechanisms for providing valued extra-
curricular experience.
189
Table 4.18. Student Activities
Botticelli Chagall Degas Picasso Renoir Rodin Van Gogh Warhol
ACTIVITY
ASB Class Yes Yes Yes Yes Yes Yes Yes Yes (2)
Journalism
Class
Yes Yes
Sports (no or
low cost to
student)
Compete in
district
Compete in
district – 13
sports
Lunch -
soccer league
Compete in
district
School-based Compete in
district – 6
sports,
Lunch sports
“Pride
Program”
provides
after-school
athletics on
site.
Compete in
district,
“positive
sports,” &
on-site pro-
grams with
non-profit
agency.
Clubs (no or
low cost to
student)
At lunch At lunch At lunch 10 activities
daily after
school
At lunch At lunch At lunch At lunch
Academic
Events (e.g.,
Spelling Bee)
History Day Spelling Bee
Academic
Pentathlon
Peer
Assistance
Leader-ship
(PAL)
Yes
Yes
190
Table 4.18 continued
Botticelli Chagall Degas Picasso Renoir Rodin Van Gogh Warhol
ACTIVITY
Other
Daily news
broadcast by
ASB students
Fitness center
available
after school
to students
and families.
Musical
Theater
Cheer-leading
Fitness
center
available
after school.
Drumline
SUPPORT
SOURCE
Funding/
Support
Source (if
identified by
principal)
Non-Profit
Agency
Prop 49
GATE
Partner with
Boys & Girls
Club for
athletics
District
contract
includes
sports
stipends.
Non-Profit
Agency
Anaheim
Angels
Non-Profit
Agency
Non-Profit
Agency
191
Sponsorship of lunchtime clubs is a regular feature of each middle school in this
study. One principal indicated that these clubs are “free” (i.e., no additional stipends or
salary are involved). Each school includes at least one student government class and
seven of the eight case-study schools offer after-school sports and recreation programs
including competitive athletics requiring referees and transportation. While Degas does
not offer after-school sports, it has an active soccer program at lunch and its fitness center
is open to students and their families after school four days per week.
Five case-study schools identified partnerships with non-profit agencies as a
mechanism for providing students with site-based after-school recreation programs in
addition to sports provided by the school. After-school programs run until six o’clock
while school-sponsored activities tend to conclude at four o’clock or thereabouts.
In addition coaching stipends for after-school sports, school-day leadership, PAL,
and journalism teachers earn a salary for the course sections they teach. Additionally,
principals mentioned special events such as one school’s haunted house and dance,
another’s holiday concert, and yet another’s stargazing night for families. Each of these
requires staff members ranging from teachers “sponsoring” and chaperoning the events to
custodial staff to administrators.
If resourced as per the Model, the eight case-study schools collectively would
allocate $1,871,200 annually for student activities. Chagall’s principal “guesstimated” an
annual expenditure of $180,000 for student activities. If resourced in accordance with the
Model, Chagall would allocate $271,000 in this area. This school offers after-school
guitar and piano instruction, intermural athletics, an open computer lab, music and drama
192
productions, and academic competitions in addition to a school-day leadership class. At
the time of the interview the principal was unsure as to the number or percentage of
students participating in these activities. Principals were unclear as to anticipated
expenditures in the area of after-school sports, recreation, and events for the year.
Instead, they were conversant regarding the types of services and events available. Data
collected reflect this orientation on the part of the principals.
Technology
The Evidence-Based Model suggests an annual $250 per student expenditure
towards computers, maintenance, and other technology to support learning. Computers
are purchased at a rate of one per three students and distributed at a rate of one per four
students, with the others used for administrative and other staff. Principals indicated
various software programs in use to support data analysis and communication. Current
instructional materials packages include technology components ranging from test
question banks to video snippets to accompany lessons. A summary of computers on
each campus shows that half the schools have more computers than the Model suggests.
Three schools allocate computers at a lower level than recommended by the Model (2 are
close). A final school is almost identical with the Model.
Two schools made special mention of using computer software programs for
helping struggling students catch up with their classmates. It was beyond the scope of
this study to examine computer use for quantity or quality of technology experiences.
193
Table 4.19. Allocation of Computers
Botti-
celli
Chagall Degas Picasso Renoir Rodin Van
Gogh
Warhol
Actual 1075 664 789 132 524 270 278 505
Model 309 453 391 321 579 217 274 574
Compa-
rison
3.5x
Model
1.5x
Model
2.0x
Model
0.4x
Model
0.9x
Model
1.2x
Model
1.0x
Model
0.9x
Model
Expenditure Elements in Aggregate Form
Presentation and discussion of school-level resource use highlights comparisons
across schools as well as with the Evidence-Based Model. A comprehensive picture as
presented in Table 4.20 presents the data in an aggregate view, removing much of the
detail while retaining the essence of resource use within each expenditure element upon
which the Model is based.
194
Table 4.20. Aggregate Average Resource Use in Full Time Equivalencies
Expenditure Element Staffing Category Model
Recommendation
(FTEs unless indicated
otherwise)
Actual
(FTEs unless indicated
otherwise)
Teachers Core Academic
Teachers
374.2 258.8
Specialist/Elective
Teachers
74.8 84.2 (includes World
Languages)
Library Staff Librarians 20.8 2.2
Media Specialists 0 4.6
Library Assistants 0 0.9
Technology Support Certified Technology
Coordinator counts
toward Librarian
allocation.
1.5 Technology
Troubleshooters
(classified position)
Extra Help Tutors 73.4 2.4
ELD Teachers 59.0 9.9
Mild-to-Moderate
Special Education
Teachers
62.4 31.1
Severely Handicapped
Teachers
As needed 29.3
Special Education
Aides
As needed 54.4 SH Aides
9.3 Inclusion Aides
7.1 Resource Aides
Extended Day 61.2 11.3
Summer School 61.2 ≈5.5 (Unclear due to
data-collection issues)
Professional
Development
Instructional
Facilitators
73.4 2.1
Student Support Guidance Counselors 37.4 17.5
Pupil Support Staff 73.4 39.8
Supervisory Aides
41.6 7.9
Administration Principals 8.0 8.0
Assistant Principals 16.8 15.5
Other Administrators 0 5.8
Secretaries 41.6 11.3
Clerks 0 15.8
Operations &
Maintenance
Security
7.3
Custodians 32.9 28.0
Total All Staff 1081.8 ≈730.0
195
In addition to human resources, fungible resources are necessary to purchase
materials and services to support students and staff. While this study focused primarily
on school-level use of human resources, data collection included several anticipated
expenditure items mostly in the area of professional development. Anecdotal responses
from principals as well as data gleaned from sources including weekly meeting schedules
informed tentative statements regarding possible actual planned expenditures in areas
including instructional materials, gifted education, and student activities. These data
reflect dollars and/or human resources allocated to complete the comparison with the
Evidence-Based Model. They assist in informing the Model for future modifications and
break down expenditures in greater detail than found in typical school budgets. Details
follow in Table 4.21.
Table 4.21. Average Resource Use in Dollars
Expenditure Element Expenditure Category Recommendation Actual
Extra Support GATE Funds $25 per student
($233,900)
$31,594
Instructional Materials
& Equipment
Textbooks $120 per student
$1,122,720
Beyond the scope of
this study. However,
state requirements
imply an expenditure
equal or greater than
the Model
recommends.
Library Resources $20 per student
$187,120
Beyond the scope of
this study. However,
School Library
Block Grants have
been reduced in
2008-2009, resulting
in diminished
resources
substantially lower
than Model
recommendations.
196
Table 4.21 continued
Expenditure Element Expenditure Category Recommendation Actual
Formative Assessments $25 per student
$233,900
Beyond the scope of
this study. However,
professional
collaboration time
including time to
develop assessments
implies an
investment on par
with Model
recommendations.
Technology
$250 per student
$2,339,000
Beyond the scope of
this study. However,
anecdotal comments
from principals
indicate substantially
lower resource levels
for technology
maintenance &
purchases than
recommended by the
Model.
Student Activities
$200 per student
$1.871,200
Each school offers
ASB and yearbook
classes as well as
school clubs,
athletics, and
activities.
Professional
Development
$100 per student
$935,600
• Consultants:
$126,000
• Trainers: $102,000
• Travel: $59,500
• Registration Fees:
$45,100
• Substitute Teachers
for Release Time:
$49,200
• Materials: $28,050
• Other PD: $49,223
TOTAL:
$459,073
Total All Items $4,818,430 See individual items
above
197
Summary and Discussion
Four research questions grounded this study and frame its summary and
discussion of findings. Principal interviews revealed school-level instructional
improvement strategies, which were analyzed to determine resource allocation levels to
support those strategies and to compare allocations to the Evidence-Based Model for
adequate school funding. Allocation levels for each element of the expenditure model
were laid out in the presentation of findings. A summary of strategies, resource levels
and Evidence-Based Model comparison follows. A key assumption of the study was that
resource availability affects the development and implementation of a school’s
instructional improvement plan.
Strategies in Place
Each participating school faces increasing federal sanctions as a result of its
Program Improvement status. Each principal indicated a strong sentiment that the
school’s students deserve and need the best possible instruction. In fact, each principal
indicated that improving the quality of instruction is an urgent and critical factor in
making necessary and dramatic increases in student learning and performance. Trends in
school-level instructional improvement strategies across the case-study schools include
data analysis and goal setting, professional learning communities, reworking the school
schedule, leadership teams and shared leadership, and differentiating classes to meet
students’ learning needs.
Smaller class sizes. Class-size reduction emerged as a desire across case-study
schools and four schools have taken action to create smaller classes. Two case-study
198
schools (Degas & Warhol) happen to be QEIA recipients and have used these funds to
reduce class size. Warhol additionally has increased its clerical staff and has provided
additional time for department leaders to support their teams during the school day. Van
Gogh’s teachers voluntarily relinquished half of their prep period time in order to lower
class size and maintain the school’s media assistant in the face of budget cuts. Picasso’s
principal solicited and received additional funding from the district to hire an additional
teacher to provide smaller classes for struggling readers.
School size. Ranging in size from 652 to 1,738 students with an average
enrollment of 1,170 students, no school in the study looked like a prototypical Evidence-
Based Model school. However, Rodin (the smallest school) requested and received
permission to eliminate sixth grade from its campus in order to shrink the size of the
school while creating a more developmentally similar student body. Rodin additionally
has arranged itself into smaller learning communities in which students are distributed
among interdisciplinary teams of teachers. Renoir’s principal indicated that each of the
school’s assistant principals essentially functions as a grade-level principal. The impact
of this in terms of creating smaller learning communities is unclear. Degas (enrollment
1,173) separates its sixth graders from the older students and they are placed in self-
contained classrooms instead of moving from class to class throughout the day.
Examples appeared across the case-study schools of efforts to create more intimate
environments in which students could develop positive relationships with teachers and
199
peers in spite of large school sizes. No school completely had divided itself into smaller
schools.
School schedules. Seven of the eight case-study schools have reworked their
school schedules to include collaborative work and learning time for teachers as well as
to provide students with extra time in language arts and intervention courses. The notion
of elective courses has changed to include a menu of study skills, life skills, reading, and
mathematics intervention courses for designated students. In an effort to provide all
students with the opportunity to participate in enrichment experiences, four case study
schools provide intervention and/or elective courses before or after school, extending the
school day and offering full-participation opportunities for all students.
Extra support. All principals indicated efforts to provide students with extra
support during the school day and seven schools provide before- and/or after-school
academic assistance. California Standards Test, diagnostic, and benchmark assessment
results inform class placement. Additionally, grades and benchmark assessment results
determine participation in short-term “booster shot” classes either during or after school.
Efforts to extend the school year seem more challenging. One principal indicated
that district budget cuts resulted in elimination of summer school for middle school
students. Another explained that the district summer school program was held at another
school and transportation was not included. In this case, counselors identified potential
summer school students by looking at grades and proficiency levels. They encouraged,
but could not require attendance. On a positive note, three case-study schools have
developed their own extended-year programs. Renoir provides a winter inter-session
200
program as well as a CST boot camp on weekends in the spring. Van Gogh’s principal
used Title I funds to provide a credit-recovery summer program for 163 students who had
failed language arts or mathematics during the school year. This principal is exploring
the possibility of utilizing a technology-based school-year mechanism for students to
attain learning goals not met during the regular school day. Warhol’s principal used
QEIA funds to provide a summer program for struggling students. Each of these three
principals had reviewed student achievement data connected to summer school
participation and felt that their programs made a difference.
In the extra support area, school resource allocations fell substantially below the
levels suggested by the Evidence-Based Model. Use of certificated teacher tutors was
essentially non-existent. However, several case-study schools provided study skills and
“mandatorial,” classes in which struggling students attended a short-term class in lieu of
an elective until their performance levels increased.
Goal setting and a resulting emphasis on literacy. Each of the case-study schools
identified literacy as a critical element of student achievement following review of
student achievement data. A range of literacy-related improvement strategies across the
schools includes school-wide implementation of protocols such as Step Up to Writing,
Proficient Paragraph, Cornell notes, and GLAD strategies. All AVID schools utilize
Cornell notetaking schoolwide. Teachers learn these strategies from expert colleagues
during professional collaboration time or through training experiences.
Positive learning environment. Four case-study schools identified school climate
and student behavior as an essential element of a strong learning environment. One
201
school participates in a program called CHAMPS in which students are recognized for
positive behaviors and discipline centers on a school behavior code. The other three
currently participate in PBIS and have leadership teams that attend trainings to learn how
to develop school cultures in which all staff members focus on teaching youngsters to be
good students and citizens. Due to budget cuts, Chagall was compelled to move from a
block schedule back to a traditional one this year. The principal noted that acts of student
violence had increased and believed it had to do with more daily transitions – in the halls
and across subject areas.
Technology for data management and communication. Use of technology to
manage student data and communicate with parents emerged as another trend across
case-study schools. Each of the schools utilizes a subscription-based student assessment
management system (Edusoft and Data Director were the two identified), to analyze data
ranging from California English Language Development Test to common assessment
results. Other student-support technology includes a program called “Integrade,” which
compiles specific “at-risk” data items such as absences, low grades, and discipline
referrals. Through Integrade, counselors have a tool for developing reports of students in
need of additional assistance. Through “Snapgrades,” Van Gogh teachers post grades
regularly and parents can access them on-line. Additionally, Van Gogh utilizes
“Teleparent,” a system by which staff members may communicate with individual
parents or the entire school via phone messages translated into any of 47 languages. Each
school maintains a Web presence. Notable are Van Gogh’s interactive Web site designed
with “Moodle,” an open source program similar to Blackboard, and Warhol’s
202
commitment to Web-based communication represented by providing a teacher with a free
period each day to serve as the school’s Web master.
Parent and community involvement. Each of the schools related its strategies for
involving parents and the community. Warhol provides English as a second language
and computer classes for parents on its campus, offers an institute in which parents learn
how to navigate the American school system, and conducts meetings simultaneously in
both Spanish and English on topics important to parents and the community. Rodin has
created a committee of community organizations and private citizens to rally around the
school and support its programs. Most recently, it received donations of guitars in order
to offer after-school music lessons, and has a cadre of community members serving as
mentors to students in an organized after-school program. Degas invites parents to use
the school’s fitness center and provides nutrition information seminars. At the time of
the interview, Chagall was preparing for a stargazing night with a guest astronomer for
families as well as a student concert. In spite of these examples and many other efforts
made to connect with parents, case-study schools face ongoing challenges in keeping
parents actively involved in their children’s educational progress.
Staffing. Two of the eight case-study schools approached the core teacher staffing
allocations suggested by the Model. One had received a grant making this possible and
the other’s teachers gave up half their prep time so students could have smaller classes.
Allocations in almost every staffing area fell below Model recommendations.
Administrators and custodians appeared to be allocated at Model levels and instructional
203
aides exceeded Model recommendations. It was not uncommon to see the least
expensive type of employee fill a number of staffing positions. For example, library
staffing often emphasized classified library aides as opposed to certificated librarians.
Professional development. Participating principals connected professional
collaboration with professional development. School schedules and weekly calendared
sessions at seven of eight schools pointed toward organized data analysis, planning,
formative assessment development (in nascent stages), and learning. Botticelli’s weekly
professional collaboration schedule connects data-driven school-wide student learning
goals (e.g., increased literacy) with specific strategies (e.g., Step Up to Writing &
Proficient Paragraph), classroom applications (e.g., prompts and scoring rubrics),
benchmark assessments connected to state academic standards, data analysis, and
opportunities for teachers to learn the strategies. This idea of consistent, ongoing, action-
oriented learning repeated across seven case-study schools. Chagall, the one school with
no weekly collaboration time, still provides groups of teachers release time to plan and
analyze data together. Learning from expert colleagues and learning together represent a
theme that emerged in this area.
In addition to using colleagues as teachers of teachers, six case-study principals
identified multiple ways that they are bringing outside expertise into their schools.
Several schools participate in a federal mathematics grant emphasizing content
knowledge and professional collaboration. A variety of other professional development
opportunities include school-wide improvement experts, university-based projects, and
literacy experts hired to provide professional development to teachers and administrators.
204
Use of outside experts unilaterally reflects use of data to determine and fill gaps in
teacher knowledge and skills.
In addition to bringing experts onto campus, schools send emissaries to
conferences and training events. In some cases, approved events tie to the school’s plan
for improvement. In other cases, principals are committed to sending teachers to events
that the teachers feel will best serve them professionally. The three schools involved with
PBIS send specific teams charged with leadership in developing a school culture and
climate plan. One school currently is participating in a two-year school leadership-team
training program. This team is tasked with bringing what it learns back to the campus to
guide development of data-driven decision-making and goal setting school wide. In
addition to there being a variety of rationales and expectations regarding sending teachers
to trainings, the current budget situation has reduced or eliminated such plans for the
2008-2009 school year in five of the eight case-study schools.
The Evidence-Based Model suggests a $100 per-student allocation of funds
toward professional development annually. Reflective of research-based evidence
regarding effective professional development, the Model recommends 10 days of
intensive professional development in the summer and instructional facilitators to provide
ongoing, real-time support throughout the year. Instructional facilitators were nearly
non-existent across the eight schools. There was no school-wide intensive professional
development in the summer as recommended by the Model. However, language arts and
mathematics teachers had previously participated in weeklong mandated instructional
materials trainings. AVID teachers participate in a weeklong summer training in AVID
205
strategies. Individual teachers may attend institutes funded by grants during the summer.
BTSA and PAR coaches from schools within each district provide ongoing support for
program participants framed by program objectives.
In general, schools appear to have moved toward ongoing collaboration as a
means for working and learning throughout the year. Professional development days
included in teacher contracts tend to be consumed by district-driven objectives with
varying connections to a school’s improvement efforts. While accurate time comparisons
proved difficult, weekly collaboration time combined with various experiences with
experts and training events appear to be close to the Model’s 10-day recommendation.
Connecting Resource Availability with School Improvement Efforts
Principals across this study demonstrated ingenuity and perseverance in moving
dollars and soliciting additional resources to improve instruction on their campuses.
Coaches for teachers to improve instruction, increased numbers of teachers to reduce
class sizes, increased parent involvement, additional secretarial support, and more
professional development repeatedly emerged as items principals believe would make
substantial differences in student achievement. Their resource “wish lists” included
“excellent” teachers with the flexibility to respond to students’ learning needs. Two
principals indicated that they had all the professional resources they needed. One of these
wanted more parent involvement and was actively working toward that goal. The other
hoped to recapture the high-achieving students that had “fled” the district’s low
performing school, something that required, in the principal’s opinion, effective
instruction, a clean and safe school, and strong test scores.
206
Items identified by principals each relate directly with elements within the
Evidence-Based Model: instructional facilitators, teachers, secretarial staff, student
support staff, and professional learning opportunities. Ongoing efforts to secure these
resources include writing grant proposals, partnering with universities, recruiting
volunteers, and teachers giving up contract time in order to commit more of each day and
year to instruction. Challenges associated with these efforts include finding the time to
pursue unguaranteed resources as well as the stopgap nature of relying on the good nature
of staff and community members to fill roles intended to be staffed by paid employees.
Participating principals’ desires did not include all the areas considered as being under-
resourced in comparison with the Evidence-Based Model, but were congruent with
research-based school improvement strategies.
Tracking Resource Allocation in California
This study is among the first of its kind in California, a state currently facing a
severe budget shortfall with direct impact on education funding. Participating schools all
were under-resourced in comparison with the Evidence-Based Model. Replicating this
study across schools serving different demographic groups and at varying levels of
student achievement in California provides the possibility of creating a rich description of
the innovative ways schools are attempting to implement research-based school
improvement strategies within the constraints of their various situations. It also provides
the possibility of highlighting areas that currently may not be addressed by California
schools but which hold great potential (e.g., teacher tutors).
207
Each of the schools in this study must achieve ongoing dramatic student
achievement gains in order to achieve NCLB goals of 100% proficiency by 2014. In
addition to adding new schools to the research base, revisiting schools over time could
provide critical information regarding how schools go about making and changing their
resource allocation decisions, as well as how long it takes for school improvement
strategies to take effect on student learning.
This study is intended to inform both practitioners and decision/policy makers.
Assuming that a dam of school funding is not about to be released anytime soon, studies
such as this provide areas to consider for best use of the limited resources public schools
with high levels of historically underserved students.
208
CHAPTER FIVE
Summary and Conclusion
Background
Long legal battles in which state courts have interceded in previously held
legislative territory describe the contentious circumstances in which decisions about
educational adequacy and appropriate funding to support it are made. The notion that
education is the key to individual goal attainment as well as essential to national
economic and political stability resulted in the Elementary and Secondary Education Act
(ESEA) signed by President Johnson in 1965 in what was believed to be a lynchpin battle
in the war on poverty. California’s Serrano Supreme Court Decisions beginning in 1971
further focused on economically disadvantaged students by declaring that public
education represents equal protection under the law. Therefore, public education funds
must be distributed equally. Suddenly, the state eclipsed localities as the chief distributor
of funds to school districts, sparking substantial changes in funding paradigms across the
country.
As states across the nation followed California’s equal distribution precedent,
President Reagan’s National Commission on Excellence in Education published an open
letter to the American people indicting the U.S. education system for its failure to prepare
students for productivity in the American workforce. A Nation at Risk (U.S. Department
of Education, 1983) sparked education reforms leading to the signing of the No Child
Left Behind Act by President G.W. Bush in 2002. This most current iteration of the
ESEA represents the capstone of an ongoing commitment to the belief in public
209
education as an equalizer of opportunity and an accountability movement that resulted
from continually increasing costs for education without commensurate increases in
student achievement.
During the wave of reforms between the 1980s and NCLB in 2002, litigation in
Kentucky (Rose v. Council for Better Education, 1989) resulted in a state Supreme Court
ruling that reframed the idea of educational equity. In ruling that every child has a
constitutional right to an adequate education, Kentucky’s highest court declared an
adequate education to be every child’s constitutional right. Subsequent pivotal litigation
in New York further defined adequacy in education. New York’s highest court, the Court
of Appeals, determined that all children in the state deserved education that prepared
them for civic participation and competitive employment. In 2001, the Court of Appeals
declared the state’s education funding system unconstitutional and demanded an infusion
of billions of additional dollars.
As the Campaign for Fiscal Equity v. the State of New York lawsuit wended
through the New York courts, states across the nation were developing academic content
standards in order to establish common learning expectations across localities within each
state. California’s first approved content standards were in the areas of language arts and
mathematics in 1997, followed by science and history-social science the following year.
The state’s Public Schools Accountability Act, enacted in 1999, resulted in the
establishment of a formula to generate an indicator (Academic Performance Index) of
school success and improvement measured by standards-based assessments of student
learning. The same language arts and mathematics assessment results included (among
210
other scores) in the Academic Performance Index also determine school standing in terms
of federal NCLB proficiency targets.
The development of content standards and proficiency levels refined the
definition of an “adequate” education to one resulting in proficient or advanced scores on
state standards assessments. School improvement researchers now had quantifiable
education goals by which they could measure efforts being undertaken on behalf of
historically underserved students. Additionally, school finance researchers sought
methods for “costing out” an adequate education. The questions were: Is it possible to
attach dollars to proficiency? Can instructional inputs be connected to performance
outputs?
Four chief costing out approaches reflect the current state of providing practical
guidance to school funding decisions. These include the Expert/Professional Judgment
Model, the Cost Function Model, the Successful Schools Model, and the Evidence-Based
or State-of-the-Art Model. As its name implies, the Expert/Professional Judgment Model
relies on a panel of education experts to determine necessary resources. The detailed and
complex Cost Function Model uses a wide range of data to determine varying levels of
funding for schools and districts with different populations. After identifying success
criteria, the Successful Schools Model requires identification of schools meeting those
criteria, then analyzing the resource inputs in place at those schools. The Evidence-
Based or State-of-the-Art Model (Model) is grounded in research and evidence focused
on high-quality instructional programs and improving schools. The Model generates a
211
budget that can be custom built for individual schools based on school size, student
characteristics, and needs.
The Evidence-Based Model for educational adequacy connects key ingredients
(e.g., teachers & extra help for struggling students) with salaries and expenditures to
generate school budgets that ensure the level of resources understood to provide students
with an education resulting in academic proficiency. California’s school finance system
is not based on such a system and connections between resource allocation and student
achievement have not been explored in this state.
California’s education system is impacted by the state’s budgetary distress,
evidenced by daily news reports indicating such happenings as planned closures of
summer school programs (ABC7 News, January 29, 2009, Covina Valley Unified School
District) and state approval of waivers to postpone purchases of required instructional
materials. Cuts in education collide with federal expectations that roughly 11% more
students will be proficient in language arts and mathematics each year until reaching a
100% proficiency target scheduled in 2014. Diminished resources combined with
increased expectations and growing numbers of students with first languages other than
English and economically disadvantaged backgrounds beg the question: Which resources
have the greatest impact on student achievement?
The findings from this study inform the field by showing what schools feeling the
pressure of NCLB and state sanctions for poor performance along with budgetary
constraints actually are doing to improve student achievement levels. Additionally, the
212
findings inform the evidence base for the State-of-the Art approach to school adequacy in
considering essentials for school budgets.
Delimitation of the Problem
This study of resource allocations took place in fall 2008 at eight Title I Program
Improvement middle schools in Orange County, California. It is bounded by time,
location, purpose, and sample type. With its focus on resource use toward instructional
improvement, most non-instructional resources were excluded from the data collection
and analysis process. Only public schools serving students in grades six through eight or
seven and eight were considered for participation. School assessment data collection
focused on grade eight because all four core content areas are included in the battery of
annual tests at this grade level. Participants consisted of principals who responded to an
invitation to participate in this study and consented to an interview in which they (and on
occasion, additional knowledgeable personnel) responded to the questions contained
within the Evidence-Based Codebook. The impact of the state’s budget shortfall in 2008-
2009 limits these findings to the specifics of this time period. The author’s perspective,
grounded in instructional improvement literature and resource use, further delimits this
study. The results of this purposeful sample study are generalizable only to the eight
participating schools. However, they provide information for Program Improvement and
other high-stakes schools to consider.
Limitations of the Study
Generalizability of this study is limited by the sample selection process,
dependence upon self-reporting on the part of interviewees, participant bias, interviewer
213
bias and error (controlled by a pre-set series of questions and analysis protocol), and the
influence of the 2008-2009 budget shortfall on resource allocation decisions. In
California’s education system, standards-based student achievement determines school
quality. Studies such as this are grounded in the research that connects high and
increasing levels of student achievement with specific types and levels of resource
allocations. The current budget shortfall highlights the significance of exploring the state
of the education system, school-by-school, to ascertain what is taking place in schools
and how school-based resource allocation decisions jibe with the research base on school
improvement.
A purposeful sample of eight Title I Program Improvement middle schools
reflects a particular sub-population. Additionally, the homogeneity of student
demographic characteristics across the eight case-study (sample) schools further
influences generalizability to schools with students with different backgrounds.
However, the sample schools reflected the demographics of Program Improvement
middle schools across the county from which they were selected. They also reflected six
of the 10 districts with Program Improvement schools. The inclusion of two pairs of
schools from two districts affords both inter- and intra-district comparisons of resource
use. Additionally, each of these pairs includes one school that is receiving Quality
Education Investment Act (QEIA) funds. These QEIA schools offer unique comparison
points with other schools, the literature describing school improvement findings, and the
Evidence-Based Model.
214
This study relied upon self-reports by principals who voluntarily consented to be
interviewed. School and district documents and California Department of Education
resources (e.g., Data & Statistics page) supported data triangulation for accuracy.
Limitations of principals’ reporting became evident in the areas of summer school and
district support reporting. Summer school student attendance and allocation of teachers
proved challenging in some cases due to the nature in which these data are reported by
various districts. In terms of district support (e.g., leadership and professional
development), a number of principals omitted mention of such district-orchestrated
programs as Beginning Teacher Support and Assessment (BTSA) and Peer Assistance
and Reporting (PAR). These specific state-mandated programs are ubiquitous across
California. One might assume that participants targeted for these programs (i.e., new and
struggling teachers) and the lack of direct connection between these programs and school
improvement goals caused principals to not think of mentioning them. However, it raised
questions regarding what other details inadvertently may have been omitted. The nature
of the Evidence-Based interview protocol generally mitigated the possibilities of
interviewer bias or erroneous reporting.
Possibilities for researcher error/interpretation include the way in which staffing
allocations are counted. For example, while virtually every case-study school has a
substantial population of English learner students, not every school offers English
Language Development (ELD) courses. Instead, some schools provide reading
intervention classes to students who have not attained proficient marks in this area. Yet,
virtually every participating student is an English learner. In this particular case, the
215
researcher identified these as English/language arts classes, not as ELD. Decisions such
as these have the potential to skew the data regarding staffing allocations. Additionally,
case-study schools discovered that one way to provide struggling students with additional
time during the school day is to offer intervention classes in lieu of electives. Again, the
researcher had to decide: Is it an elective or a content-area class? However, the method
for calculating the total number of staff members on any campus provided for accurate
overall counts. Cleaning the data compelled close scrutiny and consistency across
schools.
Interviewer bias includes background knowledge following extensive review via
school-improvement literature that emphasizes the influence of types and quantities of
resources (e.g., number of teachers) on student achievement. This study focused on
instructional improvement efforts with the intention of quantifying them as a starting
point for future research in which student achievement may be attached to school
resource use and to school funding. While the notion of “costing out” methods is
somewhat controversial due to what Rebell refers to as “subjective assumptions” (Mathis,
2007, ¶ 3), use of rational models for costing out budgets for educational adequacy “is
vastly superior to the back-room, political deal-making that has historically determined
the amount and distribution of education funds.”
Research Questions
One purpose of this study is to develop an understanding of site-level resource use
decisions. As such, it is among the first in California to move from more macro-level
studies of school budgets and finance, and toward what is happening in the field.
216
Conducting the study required very little in the way of financial accounting and focused
instead on how people and other resources are distributed. Four research questions
derived from school-improvement literature guided the interview protocol and process as
well as subsequent data analysis:
1. What are the current instructional improvement strategies at the school level?
2. How are resources used to implement the school’s instructional improvement
strategies? (instructional vision, plan, goals, etc.)
3. How are the actual resource patterns at the school sites aligned with or different
from the resource use strategies that are used in the Evidence-Based Model?
4. How does the availability of resources affect the development and
implementation of the instructional improvement plan?
Discussion of Findings
Framing the findings with the study’s research questions provides an opportunity
to weave the qualitative elements of the case-study interviews with the quantitative
findings addressed in Chapter Four. The following discussion includes contextual
elements that influence the work of California middle schools in general as well as case-
study schools in particular.
Question One
What are the Current Instructional Improvement Strategies at the School Level?
Principals responded in depth with information in areas including creative
curriculum design, approaches to intervention with struggling students, assessment
development, special summer and intersession programs, professional development,
217
specialized staff members, distribution of leadership, school-wide implementation of
literacy and critical thinking strategies, and use of the school day to increase student
achievement especially in language arts and mathematics. Strategies either absent or
available at very low levels include instructional facilitators to support teachers in the
continual process of learning in order to effectively meet their students’ learning needs,
tutors to support struggling students, and additional support for English learners.
English learner support stood out as an area of particular need due to the
substantial percentage (from 35% to 87% of case-study schools’ enrollments) and sheer
numbers (from 248 to 1,481) of students classified as English learners across the case-
study schools. The principal of Warhol JHS, reporting that this year’s student enrollment
includes 86% English learners (1,481 students) spoke passionately about this student
group, with comments echoed by other case-study principals as well. The vast majority
of these students have grown up in the vicinity of their current school, receiving special
instruction as English learners since kindergarten. However, most have not moved
beyond conversational levels of English fluency. Close examination of one case-study
school in Table 5.1 highlights a universal trend and makes the point that development of
academic English-language fluency is critical to overall academic achievement.
218
Table 5.1 Warhol JHS – Percentage of Proficient and Advanced Scores on the 2008
Grade Eight English Language Arts CST by Student Subgroup
Student
Group
All Econ.
Disadv.
Hisp./
Latino
Eng.
Only
EL RFEP* EL (< 12
months
in
U.S.)**
EL (> 12
months
in U.S.)
%
Proficient
24% 21% 23% 44% 3% 46% 0% 3%
* Note: Reclassified-Fluent English Proficient (RFEP) is the designation given to EL students who
demonstrate English-language fluency.
**Note: This reflects 13 students.
Reclassified-Fluent English Proficient (RFEP) students tend to outperform other
student groups. The homogeneity of Warhol’s student population represented by large
and similar percentages of three student groups (i.e., 86% EL, 90% economically
disadvantaged, & 92% Hispanic/Latino), makes it safe to assume that most of the
school’s English learners also are represented in its Hispanic/Latino and economically
disadvantaged counts. Yet, its RFEP students achieve at the highest level in the school.
The potential for dramatic student achievement gains associated with targeted support for
this student group points to the value associated with allocating resources in this high-
leverage area. Instead, interview responses indicated a reduction in (and in two cases,
complete elimination of) ELD classes. Schools have moved toward reading intervention
courses and differentiated content-area courses based on student readiness characteristics.
So, English learners may participate in intensive language arts intervention classes and
“sheltered” math, social studies, and science classes in which the language load is
reduced in an effort to enhance conceptual understanding. The overall success of this
approach is illustrated in Table 5.2, which identifies English learner proficiency levels in
core content areas across the eight schools included in this study.
219
Table 5.2. English Learner 2008 CST Proficiency Percentages by Subject Area
Subject Area
ELA Math HSS Science
School Botticelli 4 19 3 16
Chagall 9 20 5 22
Degas 3 20 2 9
Picasso 6 8 10 18
Renoir 3 26 7 28
Rodin 11 29 8 23
Van Gogh 23 23 13 27
Warhol 3 10 5 32
The high numbers of English learners across the eight case-study schools reveal
one of the challenges of meeting students’ instructional needs. Actual allocations of
teachers to schools are based on numbers of students without special weighting or basing
resources on the characteristics of a school’s students. Although beyond the scope of this
particular study, it would be especially interesting to examine state and federal program
funds and pencil out how closely a given school from this study might be able to
implement Evidence-Based Model strategies to attain educational adequacy.
Question Two
How are Resources Used to Implement the School’s
Instructional Improvement Strategies?
Reorganizing time, using specific (i.e., special education) teachers, and using data
in combination with manipulation of site funds support the instructional improvement
strategies in place across the eight schools analyzed in this study. Additionally, a number
of schools utilize district-provided resources to enhance those at the site-level especially
220
in the areas of instructional facilitators, student support, instructional materials and
related professional development, and technology related to student data management
and communication.
Reorganizing Time
Two of the three schools that previously implemented a block schedule were
compelled to return to a traditional schedule due to budgetary reasons in the 2008-2009
school year. In spite of this, six of the eight case-study schools have developed systems
to permit professional collaboration and learning time on a weekly basis. Five schools
accumulate a few extra minutes of instruction on each of four days per week and bank
them for one hour of professional learning and collaboration time weekly. Two schools’
staffs voted against banking minutes and opted for common prep periods for each
department. One of these two successfully has assigned common prep periods to all
teachers within departments. These two schools are allocated quarterly data analysis days
by their district in lieu of district-sponsored professional development.
Three schools require all students to take a double dose of language arts and
mathematics instruction daily. The other five include an array of ELD, reading
intervention, mathematics support, and study skills courses with the result that, generally
speaking, students with CST language arts scores at Basic, Below Basic, and Far Below
Basic levels receive at least two hours of language arts instruction daily. The type of
school schedule (i.e., traditional or block) does not appear to influence the nature of
support available to students. However, the principal at Van Gogh MS commented that
221
professional development connected with effective teaching in an 80-minute class proved
invaluable for the school’s teachers.
Directing instructional time toward high-stakes subjects (i.e., language arts and
mathematics) has resulted in reduced time for science and social studies at three case-
study schools. Due to the large number of variables, it is unclear what, if any, impact this
reorganization of student instructional time has had on academic achievement. However,
the two schools closest to meeting AYP objectives in 2008 also dedicate the most time
overall to instruction each day. Rodin follows a traditional schedule and met its 2008
AYP objectives. Students receive 354 minutes of instruction daily, just under six hours.
Van Gogh, the sole block-schedule school in the study, nearly met 2008 AYP objectives
and has the distinction of providing Algebra instruction to the greatest percentage of its
students. At Van Gogh, students receive 343 minutes of instruction daily.
Repurposing Teachers
Each of the case-study schools has moved its special education “resource”
teachers into the general school population through incorporation of inclusion classes.
Special education teachers have joined the cadre of teachers who instruct reading
intervention classes, provide study skills and life skills classes for students falling behind
in homework and struggling with personal management and behavior, and serve as a
second teacher in specific content-area classes. Additionally, in the schools where
special education resource teachers provide content-area instruction to special education
students, they often instruct students outside their caseloads. For example, a student from
a self-contained class may take a social studies class with resource students. This shift
222
toward inclusion supports lower class sizes across schools and addresses a growing trend
toward mainstreaming and exiting special education students. Instead of serving select
students based on designations they bear, the goal is to provide every student with
necessary support to achieve academically.
Using Data for Instruction
While use of data to inform the practice of teachers in the classroom is spotty
and specific to a few departments at several schools, its use for goal setting and student
placement decisions is universal across the case-study schools. Schools throughout the
study use CST, CELDT, publisher-developed placement tests, and district benchmark
exam results to determine placement of students in appropriate content-area courses.
Additionally, each case-study school has some means for teachers to
collaboratively gather and analyze data. Six schools include weekly professional learning
and collaboration time with data analysis as a regular feature on the calendar. One school
provides teachers within departments with common prep periods so they may analyze
data and plan. The eighth school has quarterly data-analysis days and the principal
provides release time to departments as requested for the purpose of developing common
assessments, analyzing data, setting goals, and making instructional plans. Each case-
study school is striving to develop common course assessments to gather data more
frequently with the intended result of modified instructional practices. However,
common assessments require common agreements regarding what all students must learn
and know how to do, and therefore common agreements regarding the content of the
rigorous and relevant curriculum.
223
Principals at Van Gogh and Renoir identified their mathematics departments as
uniquely committed to common instructional goals reflected in common assessments, use
of data, changes in instruction, and student achievement. The same held true for science
departments at Chagall, Renoir, Van Gogh, and Warhol. Principals at Botticelli and
Warhol commented that their language arts and social studies teachers struggled with
data use and its satellite components. Headline: Where data use is happening at a
meaningful level, it appears that student achievement is growing.
Allocating Funds to Ensure School Improvement Resources
Title I, Title IIb, Title III, EIA/LEP, Proposition 49, lottery, and QEIA funds
variously underwrite school improvement strategies ranging from additional teachers to
community liaisons. As mentioned above, the SPSA reflects the consolidation of these
resources and their collective translation into school improvement resources. Examples
found in Table 5.3 illustrate the means by which various schools within this study fund
some of their school improvement strategies.
Table 5.3. Resources for School Improvement Strategies
School Strategy Source
All Schools Increase number of teachers Title I
QEIA (Degas & Warhol)
Botticelli
Chagall
Picasso
Rodin
Van Gogh
Warhol
School-sponsored extended day
academic program
Title I
All Schools External agency sponsored
extended academic and
recreation program
Prop 49
ACES
Endowments
Community partners
224
Table 5.3 continued
School Strategy Source
Renoir Increase number of
administrators
Title I & district counselor
allocation
Van Gogh
Warhol
School-sponsored summer
school program
Title I
QEIA (Warhol)
Renoir School-sponsored summer
academic strategies, winter
intersession, and test-prep
weekend programs
Title I
All Schools Instructional Aides (primarily
for special education classes)
Title I
Title III
EIA/LEP
Picasso
Renoir
Warhol
Non-instructional Aides District allocation
Title I
QEIA (Warhol)
Botticelli
Van Gogh
Community Liaison EIA/LEP
Healthy Start Grant
Title I
Botticelli Language! (reading intervention
program)
EIA/LEP
All Schools Professional Development Title I
QEIA (Degas & Warhol)
Various grants
Title IIb (Picasso)
District support
Degas
Warhol
Van Gogh
Instructional Coaches QEIA (Degas & Warhol)
District-office language arts
coaches (Van Gogh)
All Schools Student Data Management
Software (e.g., Data Director)
District support (typically
Title I funds)
Picasso
Degas
Rodin
Positive Behavioral
Interventions & Systems (PBIS)
Title I
All Schools Current State Board of
Education Approved
Instructional Materials in 4 Core
Content Areas
District purchases and
distributes
225
Creative use of funds includes Warhol and Van Gogh’s decisions to operate their
own summer programs when the district-supported programs (a) did not meet their
students’ needs and (b) were not available due to budget constraints. Warhol has used
QEIA funds to free up department chairs during the school day so they may provide
coaching to department members. Every school uses Title I resources to add to its
number of teachers and provide for instructional aides. Botticelli utilized site funds to
purchase reading intervention course materials.
The illustration above highlights several common instructional strategies,
including district-supported student data management systems. During the past five
years, this expenditure has shifted from sites to district offices reflecting its importance in
the instructional improvement process. Additional district-supported improvement
resources are described in the section that follows.
District-supported Services and Resources for School Improvement
District-provided school improvement resources and services include the data
management software previously identified. As principals responded to interview
questions, they identified several resources that appear to be unique to specific districts
and in response to unique situations at particular schools. Van Gogh’s principal indicated
that the district provides a Student Achievement Teacher (SAT) to coach teachers and
model language arts improvement strategies. The D’Orsay USD provides Rodin IS with
a half-time at-risk counselor. Chagall receives a district psychologist specialist who
oversees and troubleshoots the Individual Education Program process with the school’s
psychologists. At Picasso, the district allocates a certain number of noon aides based on
226
student enrollment. Additionally, the city provides Picasso with a school resource police
officer.
Intentional use of resources to operationalize the schools’ various instructional
improvement strategies includes a large dose of reorganizing and repurposing that make
critical data-driven conversations and effective student placement possible. Case-study
schools receive a variety of additional state and federal funds, which they direct toward
the classroom in the form of teachers and aides. As described above, these funds also
underwrite a variety of activities and personnel ranging from translators and technology
to facilitate communication between school staff and students’ families to providing
summer school and other extended year programs when such are not available through
the district. Partnerships with outside agencies support after-school programs that often
operate in tandem with school-sponsored academic programs. Each principal depends
upon every penny of state and federal support, evidenced in the use of federal funds to
support special education classroom assistants. An area to watch will be principals’
responses to students’ needs for additional time and support in light of districts
considering elimination of summer school in 2009.
Question Three
How are the Actual Resource Patterns at the School Sites Aligned with or Different from
the Resource Use Strategies that are Used in the Evidence-Based Model?
Chapter Four lays out the findings that resulted from eight Program Improvement
middle school case studies. A consistent theme that emerged is one of under-resourcing
relative to the Evidence-Based Model in nearly every expenditure area. Resources are
227
directed toward the classroom, mostly as teachers and aides (mostly to special education
students). Beyond classroom teachers and administration, the less-expensive options
frequently seem to be selected when possible (e.g., clerks as opposed to secretaries &
media specialists as opposed to librarians).
The Evidence-Based Model designates resource allocation quantities to key
instructional improvement components: core teachers, specialist and elective teachers,
extra support for struggling students, specialized programs (including special,
career/technical, & gifted/talented education programs), and professional development
and support for teachers. Students use materials (e.g., textbooks & computers)
appropriate to attain rigorous standards-aligned learning goals. These human and
material resource allocations take place in a context in which there is time within the
regular school day to provide extra support to students in need and goal-oriented, data-
driven planning and collaboration time for teachers. The context additionally includes
such social, physical, and developmental elements as extracurricular activities and
athletics in which all students may participate. In essence, the Model quantifies the
resources that operationalize the critical features of dramatically improving schools (e.g.,
rigorous curriculum, professional learning communities, & goal orientation).
Identifying the Model elements and essential improvement features in place at the
case-study schools required a return to the qualitative responses made by principals in
concert with consideration of the quantitative findings reported in Chapter Four. An
interpretative analysis of both qualitative and quantitative data results in a range of
similarities and differences between the intent and depth of current practice and that of
228
the Model and related features. The researcher placed results in subjectively determined
clusters based on the strength of their combined intent and resource allocation level
relative to the quantity and quality features in the Evidence-Based Model.
Strongly Similar Resource Use
Both the Evidence-Based Model and essential features of improving schools hold
a number of similarities with current practice across the eight case-study schools.
Similarities exist with regard to an intention to provide students with extra help as
needed, extend the school day and year, restructure the school day to provide students
with time to learn and teachers with time to work effectively, use professional learning
communities to set and work toward school improvement goals, provide teachers with
relevant and effective professional development, and equip schools with adequate
numbers of teachers and administrators.
School-day extra help. Extra help during the school day takes different forms. At
Van Gogh, Renoir, and Degas, all students receive daily extra help in the form of school-
wide double-blocked classes in language arts and mathematics. Principals at these
schools indicated that all students need extra time in these courses. At Chagall and Van
Gogh, short-term, school-day extra support takes the form of language arts, mathematics,
and study skills courses that replace electives (e.g., music, woodshop) until student
achievement reaches desired levels. Chagall and Van Gogh’s principals shared that the
language arts and mathematics departments have agreed to base student grades on
mastery. This includes substituting assessment grades from the “Reteach” classes for the
original grades.
229
Implications include willingness to change grades, philosophical commitment to
mastery, and use of common assessments and grading practices within departments.
Chagall’s extra help includes an added dimension in that a special education teacher
provides instruction in its Reteach classes. During the three-week sessions in which
students participate, this teacher works to fill in content mastery gaps and informally
observes students for possible special education placement testing recommendations.
Each of the case-study schools includes reading intervention courses, adding an
extra period of instruction in this area for students more than two years below grade level.
Picasso’s principal noted the school’s next scheduling challenge: creating time during the
school day for the additional instructional time recently added by the State Board of
Education for school-day extra help in mathematics. Overall, schools made extra help
during the school day a priority and implemented it by reducing time in electives and/or
science and history-social science courses.
Extending the school day and year. Seven of the eight case-study schools have
extended the school day with before- and after-school intervention and tutoring programs.
These range from Warhol’s zero period language arts intervention courses offered so that
students do not have to miss out on electives during the school day to Botticelli’s after-
school “Academic Empowerment Club,” mandatory for students failing to meet
benchmarks in language arts and mathematics. Picasso turns Warhol’s plan on its head,
offering zero period orchestra class for students who need academic support during the
school day. Other than at Botticelli and Warhol, participation in extended day academic
support is voluntary. Degas, the one school with no school-provided extended day
230
academics, relies on the after-school program provided by an outside agency housed on
its site to provide these services.
Principals at Renoir, Van Gogh, and Warhol have moved beyond district-
orchestrated extended-year offerings. Renoir offers intensive weekend and holiday-time
academic programs prior to the beginning of the school year, during winter break, and
prior to testing in the spring. Van Gogh and Warhol each designed their own summer
school programs in 2008 with the intention of “catching up” those students who had not
met mastery goals by the end of the school year. Van Gogh and Warhol’s principals
commented on the overwhelming success of their programs and on the importance of
providing summer school as a learning experience, not as punishment for misbehavior
during the school year. Student achievement results were so dramatic at Van Gogh that
the principal intends to include the computer-based components introduced in the
summer throughout the school year to support students when they begin to fall behind in
class.
The disconnect between district-organized summer school programs and student-
achievement goals was most evident in comments made by principals at Warhol, Chagall,
and Rodin. Rodin students must provide their own transportation across town to attend
summer school. While school counselors meet with students in need of summer
instruction to encourage their attendance and to help them with the on-line registration
process, the school receives no feedback regarding attendance or achievement. Due to
budget constraints, the summer program was so limited within the Getty Union High
School District that virtually none of Chagall’s students was permitted to participate
231
(excluding special education students). Warhol, in the same district as Chagall, benefits
from Quality Education Investment Act (QEIA) funds, which it used to operate its own
summer program in 2008. In general, principals were uninformed as to student
participation in or the effects of district-sponsored summer school programs.
Restructured school day. Each case study school has changed its school-day
structure within the last two years. In Botticelli, Warhol, and Chagall’s cases, the change
was a budget-driven return to a traditional schedule following a more costly four-by-four
block schedule. Six schools have utilized the restructured day to support professional
learning and collaboration time at least once weekly. Teachers at Warhol opted for
common prep periods within departments instead of school-wide common meeting time
once per week. Otherwise, both Warhol and Chagall are limited to quarterly school-wide
professional collaboration days with a focus on benchmark assessment results.
While Van Gogh is the only of the eight schools currently to have a four-by-four
period block schedule in place, each of the other schools uses its more traditional
schedule to create blocked periods for language arts and reading intervention instruction,
and for mathematics instruction. As mentioned above, a number of case-study schools
use elective time to provide extra support for students in need. After considering the
amount of content and the significance of the annual California science exam in eighth
grade relative to mathematics, Picasso’s principal has decided that next year science will
be reduced to two trimesters, with one trimester available for additional mathematics time
for students struggling in this area. Schedule decisions across the eight schools rely on
budgets, federal proficiency targets, and state exam goals.
232
Results were so significant following implementation of the four-by-four block
schedule at Van Gogh that the entire staff voluntarily opted to take only half their prep
time, making the schedule financially viable for this year. The principal commented on
the importance of investing in teachers through professional development that permits
them to be effective in longer classes. On the other hand, Warhol’s staff resisted the
addition of two minutes per class period in order to create time for one hour of common
meeting time weekly. Whether it is to maximize student-learning opportunities or to
provide professional learning and collaboration time, schedules are a key vehicle.
Professional learning communities and professional development. As mentioned
above, one intention behind restructuring the school day has been to provide teachers
with time during the regular school day to examine student achievement data, set goals,
plan, and learn together. Five schools provide this kind of time by teaching a few minutes
longer during each of four days each week to bank once-weekly school-wide professional
meeting and learning time. Picasso MS begins the instructional day at nine o’clock.
Teachers share a common prep time beginning at eight in the morning. Once per week,
the entire staff meets for professional development and departments are encouraged to
use one additional prep period weekly for the work of professional learning communities.
Professional development is included in this section because of the trend across
case-study schools to combine professional learning communities and professional
development. While no school engages in 10-day intensive summer institutes as
recommended by the Evidence-Based Model, each works to provide teachers with
instructional improvement objectives and mechanisms to learn strategies, share results,
233
and consider next steps throughout the school year. Additionally, the various
professional development programs engaging the efforts of staff members all include
professional learning communities as a critical vehicle for collectively inquiring into
student data, considering students’ instructional needs, determining requisite teacher
knowledge and skills, and planning accordingly. From PBIS to a federal math and
science grant to work with content-area consultants and school improvement firms,
participants come together as professional learning communities in which teacher
learning includes a cycle of data-analysis techniques, targeted professional learning based
on emerging needs, and curriculum/assessment design.
Professional development through professional learning communities reflects
characteristics of effective professional development as articulated by Elmore (2002),
Garet et al. (2001), Odden et al. (2002), the National Staff Development Council (2008),
and No Child Left Behind (cited in Yoon et al., 2007). These criteria describe
professional development that is (a) focused on improving student learning, (b) sustained,
intensive, and content-focused, (c) standards-aligned, (d) intended to increase teacher
knowledge of subject matter, (e) designed to improve and increase teacher knowledge of
scientifically based pedagogical practices, and (f) evaluated for effects on teacher
effectiveness and student achievement.
As the 2008-2009 school year wears on, district freezes on professional
development include registration fees, travel and lodging, and substitute coverage. As a
result, on-site professional development through professional learning communities has
become increasingly significant. One hour per week amounts to up to 36 hours
234
committed to professional collective inquiry and learning each school year. Professional
development is only worth the student achievement improvement it creates, a reminder
that all professional development efforts should be evaluated for effectiveness.
Teachers and administrators. Across the eight schools in this study principals
emphasized the absolute necessity of high quality teachers. In terms of administration,
each school’s resource allocation was close to the Evidence-Based Model target. Six of
the eight schools provide core instruction in classes with sizes ranging from roughly 36 to
42 students. Two schools (Degas and Warhol) became recipients of QEIA funds in 2007
and immediately began to use these dollars to add teachers to lower class sizes. Van
Gogh’s staff, mentioned above, made a triple commitment to maintain its block schedule,
its lower class sizes, and a media specialist without extraordinary resources. The
commitment was great enough for teachers to rescind half their prep time to make this a
reality.
Across the eight schools, principals mentioned efforts to reduce class sizes in
intervention and study skills classes. Picasso’s principal in particular solicited the district
for an additional reading intervention teacher to maintain a ratio of one teacher to 20
students in this class. At Van Gogh, the goal is to have no more than 12 students in an
intervention class. This focus of smaller classes for especially struggling students
provides a new slant on the Model’s suggestion of 25 students per class. Perhaps there
are specific (types of) classes in which these lower class sizes make a significant
difference.
235
Capturing the intent of instructional improvement strategies results in a variety of
manifestations and combinations across the case-study schools. School leaders have
combined professional development, professional learning communities, and restructured
school days to address professional needs. Schools extend the day and year with
extensive menus to provide flexibility in addressing student needs while partnering with
outside agencies to maximize resources directed toward students. As resources become
available, schools increase the ranks of their teaching staffs and work to maintain an
administrative leadership corps to sustain and facilitate ongoing improvement efforts.
Even when the numbers do not reach Model recommendations, understanding of the
intent appears evident.
Gray Areas
Several areas emerged in which principals’ statements revealed resource decisions
with weaker connections to the intention of the Evidence-Based Model and the essential
features of improving schools. Whereas the similarities apply relatively consistently
across the case-study schools, connections to the Model in this gray area tend to be more
idiosyncratic. Instructional improvement areas in which the connections appear weaker
include rigorous curriculum, goal setting for student learning, use of data, technology for
learning, effective leadership, and class size.
Rigorous, relevant, and engaging curriculum. Improving instruction at several
case-study schools includes teaching at grade level. Botticelli’s principal related the
school’s collective epiphany that using state-approved instructional materials did not
automatically translate to teachers instructing at grade level. Principals at Chagall,
236
Warhol, Renoir, Rodin and Van Gogh made a point of emphasizing that the curriculum
does not equal the instructional materials. They pointed to areas of dramatic student
achievement gains (e.g., science and mathematics), indicating that teachers had identified
learning goals and had created their own curricula, using district-adopted, state-approved
materials when appropriate.
On the other hand, as a result of their Program Improvement status, the language
arts and mathematics teachers at each of the case-study schools have participated in a 40-
hour training followed by an 80-hour practicum to develop expertise in using their
instructional materials. Degas’ principal shared that these trainings developed teacher
awareness about the materials, but did not result in teachers becoming masters either with
the content, standards, learning objectives, or the materials themselves.
When asked about the curriculum in each core content area, some principals
responded with a list of the instructional materials programs they have purchased (e.g.,
Glencoe social studies or Language!). Others explained the ongoing and iterative process
of curriculum development taking place in various departments, their desire to see this
spread across their campuses, and challenges associated with this effort with different
departments and content areas.
Marzano (2003) describes curriculum as everything the students are expected to
learn and how that learning will be achieved. DuFour and Eaker (1998) concur with
Marzano, explaining that teachers must decide what it is that they want their students to
learn before they effectively can teach them. While each of the principals explained
impressive school-wide instructional strategies in place to improve student achievement
237
scores, it was unclear the extent to which teachers actually included those strategies in
their daily work, and if or how they attached learning goals to those strategies. It was
beyond the scope of this study to examine the relevance or level of engagement provided
by any school’s curriculum. However, the entire area of curriculum is one in which
schools across the study have varied levels of success and understanding.
Goal setting. Strongly connected to curricular choices, interview data revealed
different levels of understanding of the purpose and value of establishing a vision and
setting goals. Each school identified California Standards Test achievement (and related
API and AYP objectives) as driving forces if not their actual goals for student
achievement. Rodin’s principal (and the poster in the principal’s office) identified
Rodin’s “big, hairy, audacious” goal of becoming a California Distinguished School
within four years. This vision requires that every numerically significant student group
meet state and federal proficiency objectives and implies the school’s expectation that
teachers will figure out how to increase achievement for all student groups. Its tamer
goal is to exit Program Improvement in 2009. Rodin incidentally happened to be the
single school that met its federal AYP targets in 2008. The school’s vision and goals
have mobilized the staff to focus on data and what they personally are doing to improve
instruction.
No school has met the academic needs of its special education students. Yet, just
two of the eight schools indicated a move toward setting academic achievement goals for
this student group. Goal setting ties directly to the belief that all students can learn and
238
drives the specifics of the work of effective schools. This area appeared fuzzy in the
majority of the schools studied.
Use of data. Data are essential to robust, effective professional learning
communities and to schools with a singular focus on student achievement. Each of the
participating schools uses data to place students in core classes and to determine
intervention needs. Chagall’s counselors utilize a computer software program to combine
data regarding “at-risk” characteristics (e.g., absences, grades, disciplinary referrals) in its
effort to ensure student success. The science teachers at Chagall, Renoir, Degas, and Van
Gogh meet regularly to examine common assessment data, then modify instructional
plans accordingly. The principals at Renoir and Van Gogh attribute student achievement
in mathematics to their departments’ foci on data and learning goals. Use of data as
described above requires a commitment to create and use (at the same time) common
assessments, which further means a departmental commitment to an agreed-upon set of
learning objectives for students.
Principals repeatedly commented that this package of common goals, common
assessments, and use of data was emerging slowly outside science and mathematics.
They commented that language arts and history-social science teachers found this to be
especially challenging. Principals’ speculation into the difficulty included the density of
the curriculum in each of those two subject areas, the pre-packaging of language arts
programs resulting in a disconnect for teachers between what they are doing and why
they are doing it, the feeling of being overwhelmed on the part of history teachers
(especially at the eighth grade and especially in schools with reduced history class time)
239
because of the pressure to prepare students for a three-year cumulative exam at the end of
eighth grade. The schools identified above as being successful with their use of data in
mathematics and science also recognize that their teachers participated in a federal math
and science grant with a focus on professional learning communities (including use of
data).
With the exception of some departments at some schools, data use does not exist
at the level called for to dramatically improve student learning. Degas, in working with
the School Management Program (SMP) at UCLA, has begun to use a classroom
walkthrough process to collect data focused on instruction. Teachers receive immediate
feedback regarding school-wide “non-negotiables” in terms of instruction. The principal
sees changes in teacher actions as a result, and looks forward to examining the longer-
term effects of these walkthrough visits.
Principals recognize the value of data use and see areas in which it is making a
difference in teachers’ instruction and student achievement. Schools such as Chagall
serve as a reminder that inviting teachers to collaborate using data may not be sufficient.
Although the principal has extended an open invitation to all departments or collaborators
to use release time during the school day for development of common assessments, data
examination, and planning, only the science department and a newly formed
interdisciplinary group of teachers has taken advantage of this opportunity.
It appears that there is tremendous potential in using data if combined effectively
with a purpose for student learning and an expectation that it will result in teachers
changing the way they instruct. The weekly professional learning community schedules
240
provided by five of the case study principals clearly indicate time for data analysis,
common assessment development, and planning. Moving some departments beyond
arguing about what students need to know and be able to do requires more than providing
the time to do so. Effective goal setting and use of data rely on leaders’ expectations and
development of requisite capacities for staff members to traverse this territory
successfully. It is not surprising then, that effective leadership surfaced as a school
improvement feature in this gray area.
Leadership. Six of the eight case-study schools appear to utilize a system of
distributed leadership including a focus on developing the instructional and leadership
capacities of staff members. One is moving from a top-down leadership model to a more
distributed style while the approach of the other principal is less clear. Two principals
stand out in terms of the leadership they have provided to their staffs, resulting in vision
and goal statements that appear to have mobilized and energized faculty members. One
has relied on bringing in outside sources to tell staff members what to do and is slowly
moving toward less micromanagement and more capacity building. The Single Plan for
Student Achievement, a document that each school must complete each year, contains the
plan for accomplishing the goals set by each school. One principal overtly bases
leadership efforts on this document while another plans to connect more directly with it
in the future.
Rodin and Van Gogh’s principals elaborated on the idea that school improvement
would be impossible without a complete team dedicated to that effort. At 652 and 822
students respectively, Rodin and Van Gogh also are the two smallest schools within the
241
sample, with implications for effectively leading a staff. These two stand out in terms of
their effectiveness in leading their staffs to create meaningful goals and an achievable,
desirable vision beyond simply meeting state and federal requirements. Botticelli’s
principal has a personal vision for the school and is working toward developing a
common vision school-wide.
The principals that articulated a sense of trust in staff members to do the hard
work of instruction and have rallied their teachers around a common vision are closest to
achieving state and federal growth objectives. They balance support with expectations
and focus on relationships both within and outside the schoolhouse. Van Gogh met and
exceeded both state and federal objectives in all areas with the exception of its special
education student group. Rodin met federal AYP goals in 2008 and plans to exit Program
Improvement in 2009. This intangible difference is evident in the aura of each school, in
the nature of data use, and the way that data inform school practices broadly and in the
classroom. The power of collectively determining a vision, then effectively and
continually communicating and focusing on it seems to be a critical component of
successful leadership that extends beyond case-study schools.
Technology for learning. Rodin, the one school to meet 2008 federal AYP
targets, was the one case-study school resourced substantially below the level suggested
by the Evidence-Based Model. At the other end of the spectrum, Botticelli launched a
one-to-one laptop program three years ago, resulting in the school exceeding Model
recommendations in this area. Botticelli’s principal, a self-described technophile, related
the school’s decision to pull back from its one-to-one laptop program because
242
achievement data did not justify the time and expense. The principal elaborated,
explaining that technology adds a layer of complexity to the act of teaching. Effective
use requires a tremendous amount of thought and knowledge about both the content area
and the technology. This year, a small group of teachers is working with a cohort of 70
students to develop effective ideas for using computers for more than electronic
worksheets. The lesson learned at Botticelli is that vision and effective decisions
regarding technology use are critical in using it effectively for learning.
Smaller class size. Degas, Warhol, and Van Gogh each have found means beyond
Title I funds to reduce class size. Degas and Warhol accomplished this through use of
QEIA funds allotted to 488 of the state’s lowest performing schools through a lottery
from accepted proposals for use of the funds. Van Gogh relies upon teachers to
voluntarily relinquish half their prep time in order to teach so that class sizes may stay
low in a block schedule environment.
Even with the infusion of funds, Warhol’s core class size (average of 37 students
per class) remains higher than at Renoir (36 students per class), a school without the
QEIA funding source. If Spanish language classes are included as core courses, the
average (mean) class sizes at Warhol and Renoir are nearly identical at 35.8 and 35.7
students per core class respectively at each school. The average (mean) class size at
Warhol drops to 27.5 students per class when elective and specialist classes are part of
the count. At Renoir, overall class size decreases to 31.6 students per class. Meanwhile,
Degas has used QEIA funds to reduce its core class size to 27 students per class and
overall class size to 23.5 students per class.
243
An illustration of the class sizes at each of the schools directing extraordinary
resources toward class size reduction highlights the extent toward which resources
actually are directed for this purpose. It alludes to a resource-use decision Warhol’s staff
made with regard to use of QEIA funds. Unlike Degas, Warhol has distributed its QEIA
funds to support a variety of areas including class-size reduction, increased non-
instructional time for department leaders, time for a staff member to work on the school’s
Web site, and increased office staff. In Warhol’s case, the principal feels that staff morale
depends upon a sense that there is ample time to accomplish all that the school needs to
do. By supporting morale through a combination of time and somewhat smaller classes,
the principal and leadership team anticipate continued dramatic student achievement
gains. In contrast, Degas chose to direct nearly all QEIA funds toward reducing class size
on the belief that smaller classes will result in greater student achievement gains while
easing classroom management and grading stresses on teachers.
Two headlines emerge here: Schools make their own decisions about where and
why resources might have the greatest influence. Additionally, short-term solutions such
as QEIA funds or dependence on the good will of teachers will not permanently provide
for smaller class sizes.
Table 5.4. Average (Mean) Class Sizes at Schools Using Extraordinary Resources to
Reduce Class Sizes
Degas Van Gogh Warhol
Core Classes 27.1 27.7 37.2
All Classes 23.5 20.8 27.5
244
While focusing on the three schools that are using special resources to reduce
class sizes provides insight as to where they direct their teacher allocations, it is essential
to remember the five schools without these particular resources. Picasso’s principal was
successful in adding one teacher in a strategic area in order to minimize class size in
reading intervention courses. Additionally, both Chagall and Van Gogh’s principals
remarked about their commitments to keep their various intervention course classes
small. Both accomplished this by including intervention classes as elective courses.
They are able to move students in and out of short-term interventions as needed by
offering relatively large-sized elective classes balanced by small intervention classes
typically with 12 to 15 students. These leadership actions do not reflect on the averages,
but highlight efforts to maximize resources where most needed.
Clearly, principals at case-study schools are considering smaller class sizes in
their efforts to increase student achievement. However, schools such as Warhol provide
a reminder that some see this as one of numerous possible strategies. California QEIA
funds provided for smaller class sizes, professional development, and counselors
beginning with the 2007-2008 school year. Both Degas and Warhol experienced
substantial API growth at the end of 2008, even with the difference in the way they have
used these funds. After its third year of smaller class sizes (and block schedules), Van
Gogh’s growth placed the school above the state average API. Other than in special
education, all student groups at Van Gogh met or exceeded federal AYP proficiency
targets. This information adds depth to Table 5.5, which illustrates growth in schools’
APIs and proficiency percentages over one year.
245
Table 5.5. 2007 – 2008 Achievement Gains at Schools with Reduced Class Sizes
2008 API 2007-2008 API
Increase (measured
in API points)
2007 - 2008
Increase in%
Proficient in 8
th
Grade Language
Arts
2007 - 2008
Increase in
% Proficient
in 8
th
Grade
Mathematics
Degas 603 27 1% 17%
Van Gogh 750 41 5% 14%
Warhol 665 41 6% 11%
If indeed smaller class sizes have contributed to Van Gogh’s school improvement
gains, it might be then predicted that Degas, with its nearly identical class sizes, might be
in line to make similar growth. While this oversimplifies the situation, it reflects the
possibilities of following individual schools and seeking correlations between specific
expenditure elements and school improvement.
Smaller class sizes fell into the “gray area” category because only three schools
have access to extraordinary resources other than already utilized Program Improvement
funds to ensure smaller classes for their students. Also, one school with resources to
substantially reduce class sizes has opted to dilute this by investing in a number of
different expenditure areas across the school. The unique approaches to providing
especially struggling students with small classes demonstrate a valuing of this resource
even when it is not possible school-wide.
Areas more idiosyncratically and more superficially addressed across the eight
case-study schools include provision of a rigorous and relevant curriculum, goal setting
for student learning, use of data to inform instruction, effective leadership, and creation
of smaller class sizes. While these areas were identified as important by principals, it
246
appears that teacher knowledge and a willingness to target specific school-improvement
approaches at the expense of others are necessary components of implementation of
improvement strategies with the quality and focus necessary to realize the difference they
are capable of making. Beyond these gray areas in which schools are dabbling with
powerful strategies, a number of other areas emerged in the analysis in which schools
diverge from the Model.
Differences
The differences between case-study schools and the Evidence-Based Model lie in
quantitative resource-use decisions. These choices appear to reflect decisions made to
substitute less-expensive replacements in lieu of trained professional staff members,
redirect rather than increase staffing allocations, treat some professional positions as
dispensable luxury items, redefine elective and specialist classes, and maintain large
schools. They appear to be the trade-offs necessary to maintain the elements that are in
place.
Substitutions. Trained librarians and secretaries exist in small numbers across the
eight case-study schools. Instead, schools provide less-expensive substitutes in the form
of media specialists and clerks. This results in libraries being used simply to check out
materials instead of teachers and students having the opportunity to learn how to become
effective consumers of information.
Rodin is the one case-study school that met its federal AYP objectives in 2008. It
also has the only full-time librarian. Librarians contribute knowledge of how to find the
most meaningful resources. They have the capacity to inform teachers regarding
247
effective use of technology for student learning and should be part of the team that
develops any school’s technology vision and plan. Schools without librarians either go
without this knowledge entirely or are subject to whatever teachers and administrators
bring to schools with regard to being effective filters of information and connoisseurs of
print. Data regarding the significance of librarians to influence student achievement
gains would further inform the decisions schools (and their districts) make with regard to
selecting a lower-expense media specialist.
As expressed at one school, low secretarial allocations results in the principal by
default becoming an extremely high paid secretary instead of having the time to serve as
a true instructional leader. Warhol has responded to this shortage by using QEIA funds
to hire lower-paid clerks to support the existing secretarial staff members. The variety of
titles assigned to front-office staff implies a hierarchy with implications for effectiveness
and role limitations. This finding resulted in many questions and no strong conclusion.
Because front-office staff members are the hubs for school information, studies of
effectiveness of school systems and appropriate staffing naturally should include them.
The documentation and information processing loads at public schools, especially those
receiving state and federal funds, is enormous. Yet, no school in this study contained
what the Model would deem adequate numbers of secretaries to manage those
responsibilities.
Schools utilize instructional aides, a common feature on the school landscape that
is not supported by the Evidence-Based Model. Most aides are placed in special
education classes, and one principal described their purpose in connection with reducing
248
the ratio of students to adults in the classroom. This implies that aides may be used as
less-expensive substitutes for teachers.
For mildly handicapped students and those with specific learning disabilities, the
Model suggests the addition of more teachers rather than use of aides. Principals
indicated that special education aides often are funded through Title I resources, pointing
to an effort to support the learning needs of some students at the expense of others also in
need. Additionally, the largest number of aides works in self-contained classes with
more severely handicapped and medically fragile students. It is unclear if they actually
serve as instructional aides or student support assistants. Becoming clear regarding the
various roles played by classroom aides and evaluating their effectiveness could
transform their use in schools.
Redirection of resources. While the Model suggests additional teachers to
support English learners, case-study schools have moved away from ELD classes and
toward reading intervention and content classes with the needs of English learners in
mind. Two case-study schools have eliminated ELD altogether. Instead of adding staff
members to support the needs of English learners, schools have developed differentiated
courses to support students with differing levels of readiness, including those developing
English language fluency. Across the eight schools, English learners trailed their English
only and English fluent classmates in state exams as discussed earlier. However,
concerted efforts to help English learners become fluent English speakers yield dramatic
achievement gains. Supporting even a targeted number of English learners could support
Program Improvement schools in meeting their AYP objectives.
249
Title I schools have the opportunity to attain modified NCLB achievement targets
through an alternate measure referred to as “safe harbor.” By increasing the percentage
of proficient and advanced scores within a given student group by 10%, that group is
considered to have attained its AYP objective even if the percentage falls below the
annual national proficiency objective. For example, although only 11% of Rodin’s
English learners were proficient in language arts in 2008, by increasing the number of
proficient students by at least 10% over the previous year, Rodin met its AYP objective.
The dark side of this good news is that schools may engage in a game of prioritizing
resource use with the intention of incremental improvement as opposed to adequate
proficiency levels for all students. Regardless of the motivation (i.e., educational
adequacy or meeting AYP goals) or the strategy (i.e., reading intervention or ELD),
English learners do not appear to be experiencing instruction that influences their
academic achievement. Redirecting existing resources has not met their learning needs.
Luxury items? Data in the generally under-resourced area of counselors and
student support appear to indicate that schools make a trade-off between administrators
and counseling/student support staff. It is common across the case-study schools to find
administrators responsible for student discipline, attendance, and related counseling in
addition to their roles as instructional leaders. Grants and special resources (e.g., Title I &
EIA/LEP funds) support community liaison and at-risk counselor positions, implying the
impermanence of these jobs. Even the case-study school that serves as the district’s
service point for visually impaired and orthopedically handicapped students has only
itinerant nurse support. District centralization of nurses and occupational and physical
250
therapists across the case-study schools reflects efficiency and further research would be
necessary to determine its effectiveness.
Principals repeatedly mentioned instructional facilitators as highly desired
resources in supporting teachers as they continually learn how to teach in order that
students may learn. However, this non-classroom position was available primarily at
schools with QEIA funds or with district-supported instructional coaches assigned to
schools. Warhol’s decision to add an additional student-free period to department chairs’
schedules in order that they might coach department members includes no training or
preparation to support the quality of department chairs’ work in this area. Both quality
and quantity count and it is unclear if quality exists even when coaches are present.
Tutors. Case-study schools provide virtually no teacher tutors for struggling
students. AVID students, the higher-performing students at case-study schools, receive
tutoring support from college students. Small Reteach, study skills, and “Mandatorial”
intervention classes across the eight case-study schools may serve as substitutes for
tutors. Special education “shadow” teachers provide on-the-spot support to students in
content-area classes. Schools seek mechanisms to provide students with effective and
flexible support. The effectiveness of these various methods has not been evaluated or
compared with use of teacher tutors for struggling students.
Redefining electives. Elective and specialist courses tended to be over-resourced
relative to the Evidence-Based Model. A trend across case-study schools is to add
academic intervention classes to schools’ lists of elective options. While the Model
includes physical education and non-core elective courses as a means to provide all
251
teachers with prep and planning time during the regular school day, the use of this time to
support students during the regular school day redefines the purpose of elective classes.
Schools such as Van Gogh and Chagall have created systems in which students move in
and out of support classes throughout the school year based on achievement in language
arts and mathematics classes, or based on factors such as turning in homework in any
core class.
In addition to electives focused on student assistance, case study schools include
academically oriented electives as well. AVID and Spanish language classes are features
of schools with rapid API increases. Seven of the eight schools in this study have
adopted the AVID program as part of their commitment to prepare students for success in
high school and beyond. This program includes an elective course for participating
students in which they learn reading comprehension, notetaking, personal management,
and study skills. Warhol offers Spanish language classes for both Spanish and non-
Spanish speakers. Students have the opportunity to enter high school ready for Spanish II
or honors Spanish II, college-prep classes that can lead to Advanced Placement exams
and support college readiness.
It appears that non-core courses have the potential to directly and indirectly
support academic achievement. Flexible mechanisms for making a variety of courses
available to students may meet Model recommendations for small-group tutoring as well
as provide specific content with the potential of improving student learning in all classes.
Electives prove to be as much a release valve for students as for teachers and intentional
use of this time to meet social, developmental, and academic needs of students is equally
252
important to providing teachers with planning time. This may inform schools that have
eliminated all elective course offerings in order to provide two hours of language arts,
two hours of mathematics, a shared hour of science and social studies, and one hour of
physical education daily.
Summary. In addition to instructional aides, case-study schools employ Title I and
bilingual coordinators. These positions lie outside Model recommendations. Many case-
study schools are comparatively over-resourced in the area of elective and specialist
courses, something easy to do in a school with a six or seven period day with one class
period devoted to physical education and another for an elective. However, elective
options have moved toward academic support classes in addition to technology, arts, and
vocationally related classes.
In expenditure areas identified by the Evidence-Based Model, many
comparatively under-resourced areas include core teachers, EL support teachers, support
teachers without classroom assignments (i.e., teacher tutors & instructional facilitators),
and the network of credentialed and classified staff members that support student well-
being and school operations. As a result, administrators assume the roles of social
workers, nurses, counselors, community liaisons, and secretaries within a taxed system.
Question Four
How Does the Availability of Resources Affect the Development
and Implementation of the Instructional Improvement Plan?
Two case-study principals indicated that they have all the resources they need.
However, all eight remarked that they need flexible, responsive, committed, creative
253
teachers who are willing to “do what it takes” to meet students where they are and take
them where they need to go. No school within this study is resourced at the levels
suggested by the Evidence-Based Model. However, observation of the choices schools
make reflects the strategies they believe to hold the most potential.
Decisions made by Warhol and Degas with regard to QEIA funds highlight two
different paths selected by similarly resourced schools. Warhol has distributed funds
across numerous expenditure areas while Degas has focused on core class size reduction.
When push came to shove, Van Gogh’s teachers volunteered to increase their teaching
time in order to provide students with lower class sizes and block-schedule instructional
time because of the success they have experienced during the past two years with these
strategies. Van Gogh, Warhol, and Renoir invest resources in extended year programs
customized to meet their students’ academic needs. Picasso has organized its school day
to provide teachers with common collaboration and learning time while providing the
smallest class size possible for the students with the greatest literacy development
challenges.
Before being forced to abandon its block schedule, Chagall had begun to provide
departments with common prep periods to support data analysis and planning. While the
traditional schedule makes this task more challenging, the principal continues to work
toward this objective. Every school but Degas has invested in some form of a school-
sponsored extended day program ranging from an American culture class for ELD
students at Chagall to Botticelli’s mandatory Academic Empowerment Club for failing
students.
254
While principals voiced desires for additional resources in the form of
instructional coaches and front-office support staff, they expressed a universal wish for
highly effective and professional teachers. Some of the many resource allocation
decisions they make reflect district constraints and tradition. Others reflect tremendous
ingenuity. Resource availability has resulted in one school not being able to send any
students to summer school while two others created their own. It has resulted in every
school increasing its focus on data and nearly every school vigorously working to create
robust professional learning communities. Where funds exist and the political climate
permits, schools continue to invest in professional development services.
Recognition that the case-study schools generally are resourced at levels lower
than those suggested by the Model, a condition unlikely to change in the near future in
California, led to the question: Which Model elements and which essential features drive
the work across these schools? The responses to this question appear in the instructional
improvement decisions described above and hold practical value for educators and
leaders in schools in need of improvement in California today. They additionally reflect
a convergence of research and evidence that middle schools turn to in their school
improvement efforts.
Guiding Forces for California Middle Schools Alongside the Evidence-Based Model
The California Department of Education publishes research-based documents
intended to guide efforts toward instructional excellence and academic achievement in
elementary, middle, and high schools. Schools being considered for California’s
Distinguished School designation complete an application in which they describe how
255
their efforts correspond with the 12 factors identified in Taking Center Stage – Act II
(California Department of Education, 2008f), considered essential to middle school
effectiveness. The guidelines within this document are intended to help middle schools
close the achievement gap by ensuring that all students learn. While Taking Center Stage
– Act II (TCSII) is silent on resource allocation recommendations, the similarities
between it and the Evidence-Based Model and essential features of improving schools are
worthy of mention and hold implications for both the Model and for instructional
improvement efforts. A summary of the 12 key recommendations of TCSII juxtaposed
against similar summaries for the Evidence-Based Model and the essential features of
improving schools triangulates the rationale for the Model.
Table 5.6. Comparison: Taking Center Stage – Act II, Evidence-Based Model, and
Essential Features of Improving Schools
Taking Center Stage –
Act II
Evidence-Based Model Essential Features of
Improving Schools
Rigor &
Expectations
Academic rigor/
high expectations
based on state
standards
Based on state and
federal expectations
High expectations &
goal setting
Instruction,
Assessment, &
Intervention
Appropriate
interventions &
enrichment
Extra help
GATE funds
Instructional materials
Rigorous, engaging,
relevant curriculum
Extra help for struggling
students
Time Time for learning and
collaboration
Extended day
Summer school
Professional learning
communities
Professional
development
Restructured school day
and school year
256
Table 5.6 continued
Taking Center Stage –
Act II
Evidence-Based Model Essential Features of
Improving Schools
Relevance Technology, arts, real-
world connections,
interdisciplinary
connections, & school
participation
Elective courses
Career & technology
courses
Activities & sports
Relationships Advisory & smaller
learning communities
25:1 student-teacher
ratio
450 student
prototypical school
Smaller class sizes
Transitions Articulated transitions
(between elementary
and high school)
Planning and prep time
Professional
development
Professional development
Access Qualified teachers,
quality instruction,
interventions,
leadership &
recognition
opportunities, sports,
& clubs
Professional
development
Effective curriculum
Activities & sports
Technology
Library & media
resources
Professional development
Rigorous, relevant
curriculum
Safety, Resilience,
& Health
Mentors, counseling,
positive discipline, &
civic education.
Pupil support staff
Leadership Distributed leadership
that relies on use of
data
Principal & asst.
principal allocation
Use of data
Distributed leadership
Professional
Learning
Professional Learning
Communities
Prep & planning time
during regular school
day.
Restructured school day
Accountability Time, training, talent,
& resources to support
student learning
Accountability for
moving all students
toward proficiency
Based on state and
federal expectations
High expectations
Partnerships Partnerships with
local and larger
community
Parent/Community
liaison
Parent & community
involvement
257
While TCSII does not say it outright, none of its 12 elements exists without
human and/or material resources. TCSII identifies resources through mention of high
quality teachers, time, mentors, and school structures as critical features of excellent
middle schools. It includes specific middle-school recommendations such as creating
seamless transitions for elementary students entering middle school and for graduates as
they enter high school as well as the form that discipline (i.e., positive) should take. The
alignment between the Model and TCSII has the potential to further inform middle
school and district leaders in terms of quantifying their efforts. The primary difference
appears to be TCSII’s suggestion that schools look to partnerships to support resource
efforts instead of expecting those resources from the state.
Challenges and potential identified for English learners highlight implications
within the TCSII category labeled appropriate interventions and enrichment. The
Evidence-Based Model provides resource allocation recommendations to support
academic achievement of this student group. Currently, the case-study school that met its
AYP targets in 2008 did so without meeting the educational adequacy needs of its
English learners (or special education students for that matter). What might happen if
Rodin’s already mobilized staff was increased by 2.5 FTE teachers dedicated to the
reclassification of the school’s English learners?
Emerging Insights
The conviction that all children will learn and the commitment to find ways to
make it happen resonated throughout every case-study interview. The eight principals
enthusiastically shared about their schools’ improvement journeys and their efforts to
258
develop systems, relationships, and professional capacities throughout their schools to
provide their students with excellent instruction. Insights emerging from the composite
of eight school-improvement stories lie in considerations about quality in combination
with quantity of resources, schools’ focal points, and the impact of current resource
allocations on staff and students.
Recombining and Redirecting Resources
In the face of budget restrictions, contract agreements regarding teacher-student
ratios and work schedules, case-study schools have enlisted their creative repertoires and
sacrificed personally and professionally to generate time for students to meet proficiency
expectations in high-stakes courses. They have taken existing resources (e.g., teachers &
time) and redistributed them based on goals for school improvement. Some of the new
combinations appear to be effective while in other cases, recombining existing resources
does not appear to be adequate. Professional development and professional learning
communities have come together in some form at each case study school. The success of
this combination appears in the mathematics and science successes described earlier.
However, repackaging services for English learners by minimizing ELD classes and
placing them in classes with struggling readers has not yielded great success as measured
by state English language arts assessment results. An emerging insight is the significance
of evaluating the effectiveness of instructional efforts and being responsive to the results
of those evaluations.
259
District Support of Instructional Improvement Efforts
Case-study principals unilaterally shared that they and their leadership teams were
the engineers of their schools’ plans for school improvement. District involvement with
the school-improvement process was more evident in the interviews with Degas and
Picasso’s principals. For example, Picasso is required to use the services of a consultant
that the district has hired to work with its Program Improvement schools. Even in this
case however, Picasso has determined the goals and direction for working with the
consultant. Although they provide a variety of professional development and other
support options, district offices do not seem closely connected to support that is directly
relevant to the efforts to improve instruction that are taking place on school sites (e.g.,
developing effective professional learning communities or providing programs to develop
effective coaches). In fact, as a result of the current budget shortfall, at least two districts
have completely frozen professional development budgets, regardless of the funding
source, because of concerns that the public will perceive such expenditures as frivolous
and wasteful. For the most part, solving the student achievement puzzle seems to be the
work of principals, leadership teams, and teachers in the classroom.
Finding Out What Really Works and Under What Circumstances
Analysis of summer school, instructional facilitator, and classroom aide data
offers insights into quality, quantity, and availability of resources. Other than at the two
schools that offered their own summer school programs, schools appeared to have no real
sense of any benefits resulting from the district programs their students attended.
Similarly, where instructional coaches are available, it is unclear how much their services
260
are utilized or how effective they are. Classroom aides are so much a part of the school
landscape that decisions to consider their contributions to student achievement would be
as likely as measuring the contribution of the principal.
While it is unrealistic to measure the effectiveness of each and every resource
element, those connected to direct services for students as well as those provided by
decreasing direct services for students (e.g., department chairs with an extra free period
daily for coaching) may warrant close scrutiny. Are coaches as implemented making a
difference? How do classroom aides influence student achievement? Does voluntary
summer school attendance at a distant campus make a difference?
Limitations and Benefits of Partnerships
While TCSII recommends seeking partnerships to support student achievement,
such efforts require time to develop relationships and establish common objectives.
Some principals voiced tremendous appreciation for their non-profit and community
partners while others voiced frustration. At present, none of the eight case-study schools
has a means by which student achievement may be compared with participation in after-
school programs provided by external agencies. It is highly possible that the real benefits
from these programs derive from the fact that participating students are in a safe and
supervised environment with access to resources and recreational opportunities.
Each of the insights described above could serve as an area for future research in
its own right. Additional considerations for subsequent research are identified next.
261
Considerations for Future Research
Implementation of the Quality Education Investment Act in 2007 resulted in a
large infusion of additional funds to 488 of California’s lowest-performing schools.
These dollars augment other state and federal funds allocated to support the needs of
economically disadvantaged and special needs students and create a special category of
highly resourced schools within the state. Close examination of how state and federal
funds are used to augment school resources in general and in comparison with QEIA
schools in particular has potential to inform the extent to which evidence-based practices
are being applied to instructional improvement efforts.
Any of the eight schools that participated in this study would be of interest to
researchers looking at longitudinal effects of particular instructional improvement efforts.
Because English learner student achievement emerged as a high-leverage improvement
area, and because this group is significant across Program Improvement schools, deeper
exploration of how resources are allocated to serve this student population could provide
practical possibilities for district and school administrators.
Observing generally lower proficiency levels in language arts and history-social
science sparks interest in examining schools achieving at high levels in these areas.
Additionally, deductions drawn from principals’ statements regarding mathematics and
science achievement gains as a result of vibrant professional learning communities would
be worthy of follow up through targeted research of schools with high achievement rates.
262
Concluding Remarks
Schools are expected to make remarkable achievement gains during the next five
years and have reoriented their efforts in terms of distributed leadership, professional
learning communities, and support for struggling students. Case-study schools have
embraced systems change as evidenced by collaboration and use of data. They have
reworked schedules in efforts to make better use of the regular school day. If the
successes in mathematics and science that emerged in this study are an indication of the
next growth area for schools, the quality of professional learning communities is likely to
become a focal point in the near future. Not surprisingly, the focus continues to be on
instruction.
In shedding light on school decisions regarding key resources, this study
illuminated the many areas in which resources are minimal and foreshadowed areas in
which cuts may be looming. One cannot help but wonder what the possibilities for
student achievement might be if the Evidence-Based Model were to be implemented, and
then what the will of the people might be if the Model succeeded. Thompson and
Crampton (2002) concluded that the state’s budgetary and political entities might be
incapable of committing to both equality and excellence at the same time for all children.
That said, coming to agreement on what constitutes adequate funding permits an
informed debate over educational adequacy to move forward.
263
REFERENCES
American Federation of Teachers (n.d.). Benefits of small class size. Retrieved May 24,
2008, from http://www.aft.org/topics/classsize/.
American Institutes for Research (AIR). (n.d.). A quality education: How much does it
cost? Palo Alto, CA: Author.
Andrews, M., Duncombe, W., & Yinger, J. (2002). Revisiting economies of size in
American education: Are we any closer to consensus? Economics of Education
Review, 21(3), 245-262.
Archibald, S. (2006). Narrowing in on educational resources that do affect student
achievement. Peabody Journal of Education, 81(4), 23-42.
Argyris, C. (1990). Overcoming organizational defenses: Facilitating organizational
learning. Needham, MA: Allyn Bacon.
Barnett, M. (2008). Academic plan for 2008-2009 (unpublished).
Bodilly, S. & Berends, M. (1999). Necessary district support for comprehensive school
reform. In G. Orfield & E.H. DeBray (Eds.). Hard work for good schools: Facts,
not fads, in Title I reform (pp. 111-119). Cambridge, MA: Harvard University,
The Civil Rights Project.
Bonstingl, J. (2004). Expanding learning potential for all students. Leadership
Magazine. Retrieved May 22, 2008, from
http://www.acsa.org/publications/pub_detail.cfm?leadershipPubID=1415.
Bransford, J Brown, A., & Cocking, R. (Eds). (2000). How people learn: Brain, mind
experience, and school: Expanded edition. Washington, DC: The National
Academies Press.
Brown, K., Morris, D. & Fields, M. (2005, Spring). Intervention after grade 1: Serving
increased numbers of struggling readers effectively. Journal of Literacy
Research, 37(1), 61-94.
Borman, G. & Boulay, M. (Eds.). (2004). Summer learning: Research, policies,
programs. Mahwah, NJ: Lawrence Erlbaum & Associates.
California Department of Education (2007). 2006-07 Accountability progress reporting
system: 2006 base Academic Performance Index report. Retrieved April 6, 2008,
from http://www.cde.ca.gov/ta/ac/ap/documents/infoguide06b.pdf.
264
California Department of Education (2008b). The Williams case – an explanation.
Retrieved June 20, 2008, from http://www.cde.ca.gov/eo/ce/wc/wmslawsuit.asp.
California Department of Education (2008c, January 10). News release: State schools
chief Jack O’Connell comments on governors proposed budget for fiscal year
2008-09 (Release #08-8).
California Department of Education (2008d, March 14). News Release #08-31: State
Schools Chief Jack O'Connell, Teachers, Support Staff, Administrators
Announce More Than 20,000 Teachers and Support Staff Getting Layoff Notices
Due to Budget Crisis.
California Department of Education (2008d). Fact book 2008: Handbook of education
information. [Electronic version]. Sacramento, CA: CDE Press.
California Department of Education (2008e). Schoolwide programs: Title I vision for
school reform in California. Retrieved April 6, 2008, from
http://www.cde.ca.gov/sp/sw/rt/.
California Department of Education (2008f). Taking center stage – Act II: Ensuring
success and closing the achievement gap for all of California’s middle grades
students. Retrieved February 8, 2009, from http://pubs.cde.ca.gov/tcsii/.
Campaign for Fiscal Equity, Inc. (2008). CFE v. State of New York: Ensuring every New
York child their constitutional right to a sound basic education. Retrieved April
15, 2008, from http://www.cfequity.org/background.html.
Carroll, S., Krop, C., Arkes, J., Morrison, P., & Flanagan, A. (2005). California’s K-12
public schools: How are they doing? Santa Monica, CA: RAND Corporation.
Chambers, J., Levin, J., & DeLancey, D. (2006). Efficiency and adequacy in California
school finance: a professional judgment approach [Electronic Version]. Palo
Alto, CA: American Institutes for Research. Retrieved April 18, 2008, from
www.schoolfunding.info/states/ca/CA-AIR-3-07.pdf.
Cohen, D., Hill, H. (2001). Learning policy: When state education reform works. New
Haven, CT: Yale University Press.
Collins, J. (2001). Good to great: Why some companies make the leap and others don’t.
New York: Harper Collins Publishers Inc.
Collins, J. (2005). Good to great and the social sectors: A monograph to accompany
good to great. New York: Harper Collins Publishers Inc.
265
Crane, E. & Boaz, D. (Eds.). (2005). Cato handbook on policy (6
th
ed.). Washington,
D.C.: Cato Institute.
Creswell, J. (2003). Research design: Qualitative, quantitative, and mixed methods
approaches (2
nd
ed.). Thousand Oaks, CA: Sage Publications.
Cuban, L. (1993). How teachers taught: Constancy and Change in American classrooms
1890-1990 (2
nd
ed.). New York: Teachers College Press.
Darling-Hammond, L. & Bransford, J. (Eds.). (2005). Preparing teachers for a changing
world: What teachers should learn and be able to do. Indianapolis, IN: Jossey-
Bass.
Darling-Hammond, L. & Friedlaender, D. (2008). Creating excellent and equitable
schools. Educational Leadership, 65(8), 14-21.
Dee, T. (2002). Where does new money go? Evidence from litigation and a lottery. In
W. Fowler (Ed.). Developments in School Finance, 1999-2000 (129-141).
Washington, D.C.: National Center for Education Statistics, U.S. Department of
Education.
DuFour, R. & Eaker, R. (1998). Learning communities at work: Best practices for
enhancing student achievement. Bloomington, IN: National Educational Service.
Ed-Data (2008). Glossary of terms. Retrieved April 20, 2008, from, http://www.ed-
data.k12.ca.us/glossary.asp.
EdSource (2008a). California school finance: California’s school finance system:
California school finance history. Retrieved April 2, 2008, from
http://www.californiaschoolfinance.org/FinanceSystem/History/tabid/68/Default.a
spx.
EdSource (2008b). Accountability: Related data: Title I schools in Program
Improvement. Retrieved April 5, 2008, from
http://www.EdSource.org/sch_acc_title1.cfm.
Education Commission of the States. (2005). Prisoners of time: Reprint of the 1994
report of the National Education Commission on Time and Learning. Denver,
CO: Author.
266
Education Data Partnership (2009). School technology 2007-2008. Retrieved December
29, 2008, from http://www.ed-
data.k12.ca.us/Navigation/fsTwoPanel.asp?bottom=%2Fprofile%2Easp%3Flevel
%3D07%26reportNumber%3D16.
Education Data Partnership (2008). Schools by type: State of California, 2006-2007.
Retrieved April 5, 2008, from http://www.ed-
data.k12.ca.us/Navigation/fsTwoPanel.asp?bottom=%2Fprofile.asp%3Flevel%3D
04%26reportNumber%3D16.
Elmore, R. (2002). Bridging the gap between standards and achievement: The imperative
for professional development in education. Washington, D.C.: The Albert
Shanker Institute.
Fermanich, M., Mangan, M., Odden, A., Picus, L., Gross, B., & Rudo, Z. (2006).
Washington learns: Successful district study. North Hollywood, CA: Lawrence
O. Picus and Associates.
Forum for Education and Democracy (2008). Democracy at risk: The need for a new
federal role in education policy. Amesville, OH: Author.
Fullan, M. (2002). The change leader. Educational Leadership, 59(8), 16-20.
Fullan, M. (2005). Leadership and sustainability: System thinkers in action. Thousand
Oaks, CA: Corwin Press.
Garet, M., Porter, A., Desimone, L., Birman, B., Yoon, K.S. (2001). What makes
professional development effective? Results from a national sample of teachers.
American Educational Research Journal, 38(4), 915-945.
Garfield, R.R., Garfield, G.J., & Willardson, J.D. (2003). Policy and politics in American
education. Atlanta, GA: St. Barthelemy Press, LTD.
Garmston, R. & Wellman, B. (1999). The adaptive school: A sourcebook for developing
collaborative groups. Norwood, MA: Christopher-Gordon Publishers, Inc.
Gerber, S.D., Finn, J.D., Achilles, C.M., Boyd-Zacharias, J. (2001). Teacher aides and
students’ academic achievement. Educational Evaluation and Policy Analysis,
23(2), 123-143.
Goertz, M. & Stiefel, L. (1998). School-level resource allocation in urban public schools.
Journal of Education Finance, 23(4), 435-446.
267
Greenwald, R., Hedges, L.V., & Laine, R.D. (1996). The effect of school resources on
student achievement. Review of Educational Research, 66(3), 361-396.
Grissmer, D. (1999). Conclusion: Class size effects: Assessing the evidence, its policy
implications, and future research agenda. Educational Evaluation and Policy
Analysis, 21(2), 231-248.
Guthrie, J., Hayward, G.C., Smith, J.R., Rothstein, R., Bennett, R.W., Koppick, J.E., et
al. (1997). A proposed cost-based block grant model for Wyoming school
finance. Davis, CA: Management Analysis and Planning Associates, LLC.
Retrieved July 5, 2008, from
http://legisweb.state.wy.us/schoolx/cost/final/final.htm.
Hannaway, J., McKay, S., Nakib, Y. (2002). Reform and resource allocation: National
trends and state politics. In W. Fowler, Jr. (Ed.). Developments in School
Finance, 1999-2000: Fiscal Proceedings from the Annual State Data Conference,
July 1999 and July 2000 (57-76). Washington, D.C.: National Center for
Education Statistics, U.S. Department of Education.
Hannaway, J. & Nakib, Y. (2002). How school districts allocated their resources in a
period of reform: Changes during the 1990s. School Business Affairs, 68(5), 12-
19.
Hannaway, J. & Sharkey, N. (2004). Does profit status make a difference: Resource
allocation in EMO-run and traditional public schools. Journal of Education
Finance, 30(Summer), 27-50.
Hanushek, E. (1979). Conceptual and empirical issues in the estimation of educational
production functions. The Journal of Human Resources, 14(3), 351-388.
Hanushek, E. (1989). The impact of differential expenditures on school performance.
Educational Researcher, 18(4), 45-65.
Hanushek, E. (1997). Applying performance incentives to schools for disadvantaged
populations. Education and Urban Society, 29(3), 296-316.
Hanushek, E. (1998). Conclusions and controversies about the effectiveness of school
resources. Economic Policy Review – Federal Reserve Bank of New York, 4(1),
11-27.
Hanushek, E. (1999). Evidence on class size. In S. Mayer & P. Peterson (Eds.) Earning
and learning: How schools matter (131-168). Washington, D.C.: Brookings
Institution Press.
268
Hanushek, E. (2006). Science violated. In Eric Hanushek (Ed.). Courting Failure: How
School Finance Lawsuits Exploit Judges’ Good Intentions and Harm Our
Children (257-311). Stanford, CA: Hoover Institution Press.
Hanushek, E. & Rivkin, S. (1997). Understanding the twentieth century growth in U.S.
school spending. Journal of Human Resources, 32(1), 35-68.
Hartman, W., Bolton, D., & Monk, D. (2001). A synthesis of two approaches to school-
level financial data: The accounting and resource cost model approaches. In W.
Fowler, Jr. (Ed.) Selected Papers in School Finance, 2000-01 (NCES 2001-378)
(77-119). Washington, DC: National Center for Education Statistics, U.S.
Department of Education.
Haycock, K. (2001). Closing the achievement gap. Educational Leadership, 58(6), 6-11.
Health & Education Research Operative Services, Inc. (2003). Project STAR. Retrieved
April 17, 2008, from http://www.heros-inc.org/star.htm.
Hedges, L., Laine, R., & Greenwald, R. (1994). An exchange: Part I: Does money
matter? A meta-analysis of studies of the effects of differential school inputs on
student outcomes. Educational Researcher, 23(3), 5-14.
Hickock, E. (2004, February 6). Letter to Dr. Steven O. Laing, Superintendent of Public
Instruction, Utah State Office of Education. Retrieved April 13, 2008, from
http://www.ccsso.org/content/pdfs/USDEdLettertoUtah.pdf.
Hord, S. (1997). Professional learning communities: What are they and why are they
important? Issues…About Change, 6(1). Retrieved July 6, 2008, from
http://www.sedl.org/change/issues/issues61.html.
Hoxby, C. (2000). The effects of class size on student achievement: New evidence from
population variation. The Quarterly Journal of Economics, 115(4), 1239-1285.
http://www.idra.org/IDRA_Newsletter/March_2004_Self_-
_Renewing_Schools_Early_Childhood_Education/Equity_and_Adequacy_Conce
pts_as_Considered_in_School_Finance_Court_Cases/.
Huebner, T., Corbett, G., Phillippo, K. (2006). Rethinking high school: Inaugural
graduations at New York City’s new high schools. San Francisco, CA: WestEd.
Kahlenberg, R. (2001, June 22). Learning from James Coleman. The Public Interest,
144(Summer), 54-72.
269
Kahlenberg, R. (2008). Albert Shanker and the future of teacher unions. Phi Delta
Kappan, 89(10), 712-720.
Kauffman, A. (2004, March). Equity and adequacy concepts as considered in school
finance court cases. IDRA Newsletter. Retrieved April 2, 2008, from
http://www.idra.org/IDRA_Newsletter/March_2004_Self_-
_Renewing_Schools_Early_Childhood_Education/Equity_and_Adequacy_Conce
pts_as_Considered_in_School_Finance_Court_Cases/.
Kocivar, C. & Halladay, S. (2005, June 7). Breaking the cycle of underfunded schools.
San Francisco Chronicle, B 7.
Krathwohl, D. R. (1993). Methods of educational and social science research: An
integrated approach. New York: Longman.
Krueger, A. (2003). Economic considerations and class size. The Economic Journal,
113(February), F34-F63.
Lake, R., Hill, P., O’Toole, L., Celio, M. (1999). Making standards work: A case study
of Washington state. Washington, D.C.: Thomas B. Fordham Foundation.
Lankford, H. & Wyckoff, J. (1995). Where has the money gone? An analysis of school
district spending in New York. Educational Evaluation and Policy Analysis,
17(2), 195-218.
Lawrence, B., Bingler, S., Diamond, B., Hill, B., Hoffman, J., Howley, C., Mitchell, S.,
Rudolph, D., Washor, E. (2002). Dollars and sense: Cost effectiveness of small
schools. Cincinnati, OH: Knowledge Works Foundation.
Lee, V. & Smith, J. (1997). High school size: Which works best, and for whom?
Educational Evaluation and Policy Analysis, 19(3), 205-228.
Levin, H., Belfield, C., Muennig, P., & Rouse, C. (2007). The costs and benefits of an
excellent education for all of America’s children. New York: Teachers College
Press.
Lindseth, A. (2006). The legal backdrop to adequacy. In E.A. Hanushek (Ed.) Courting
Failure: How School Finance Lawsuits Exploit Judges’ Good Intentions and
Harm Our Children (33-78). Palo Alto, CA: Hoover Press.
Loeb, S., Bryk, A., Hanushek, E. (2007). Getting down to facts: School finance and
governance in California summary. Retrieved June 20, 2008, from
http://irepp.stanford.edu/projects/cafinance.htm.
270
Mangan, M. (2007). School-level resource in Arkansas: A statewide study. Unpublished
doctoral dissertation, University of Wisconsin-Madison.
Manzo, K. (2008, January 25). Iowa moves on content standards for core subjects.
Education Week. Retrieved March 20, 2008, from
http://www.edweek.org/login.html?source=http%3A%2F%2Fwww.google.com%
2Fsearch%3Fhl%3Den%26q%3Dmanzo%2Biowa%2Bstandards%26btnG%3DSe
arch&destination=http%3A%2F%2Fwww.edweek.org%2Few%2Farticles%2F20
08%2F01%2F30%2F21iowa.h27.html&levelId=2100&baddebt=false.
Marzano, R. (2003). What works in schools: Translating research into action.
Alexandria, VA: Association for Supervision and Curriculum Development.
Marzano, R., Pickering, D., Pollock, J. (2001). Classroom instruction that works:
Research-based strategies for increasing student achievement. Alexandria, VA:
Association for Supervision of Curriculum and Development.
Marzano, R., Waters, T., & McNulty, B. (2005). School leadership that works: From
research to results. Alexandria, VA: Association for Supervision and Curriculum
Development.
Mathis, W. (2007). After five years: Revisiting the cost of the No Child Left Behind Act.
Retrieved April 18, 2008, from http://www.vsse.net/node/269.
McREL (2005, November 2). Final report: High-needs schools – What does it take to
beat the odds? (Technical Report). Aurora, CO: Mid-Continent Research for
Education and Learning.
Merrow, J. (2003). The Merrow report: Serrano v. Priest (1971 and 1976). Retrieved
May 25, 2008, from http://www.pbs.org/merrow/tv/ftw/serrano.html.
Merrow, J. (Executive Producer) (2004, February 2). First to Worst. New York:
Learning Matters, Inc.
Mid-Continent Research for Education and Learning (2008). Instruction. Retrieved June
22, 2008, from http://www.mcrel.org/topics/Instruction/.
Miles, K. (2000). Critical issue: Rethinking the use of educational resources to support
higher student achievement. NCREL. Retrieved May 8, 2008, from
www.ncrel.org/sdra/areas/issues/envrnmnt/go/go600.htm.
271
Miles, K. & Darling-Hammond, L. (1998). Rethinking the allocation of teaching
resources: Some lessons from high-performing schools. Educational Evaluation
and Policy Analysis, 20(1), 9-29.
Miles, K., Odden, A.R., Fermanich, M., & Archibald, S. (2004). Inside the black box of
school district spending on professional development: Lessons from five urban
school districts. Journal of Education Finance, 30(1), 1-26.
Mizell, M. (1999). Thirty and counting (Remarks, April 29, 1999, Southern Region
Education Board Middle Grades Conference). Retrieved April 5, 2008, from
http://www.middleweb.com/HMsouth.html.
Monk, D., Roellke, C., & Brent, B. (1996). What education dollars buy: An examination
of resource allocation patterns in New York state public school systems. Final
report prepared for the Finance Center of the Consortium for Policy Research in
Education, University of Wisconsin-Madison.
Mosteller, F. (1995, Summer/Fall). The Tennessee study of class size in the early school
grades. The Future of Children, 5(2), 113-127.
Murnane, R. & Levy, F. (1996). Teaching the new basic skills: Principles for educating
children to thrive in a changing economy. New York, NY: The Free Press.
Nakib, Y. (1995). Beyond district-level expenditures: Schooling resource allocation and
use in Florida. In Lawrence Picus and James L. Wattenbarger (Eds.), Where Does
the Money Go? Resource Allocation in Elementary and Secondary Schools (pp.
85-105). Thousand Oaks, CA: Corwin Press.
National Access Network (2008). Resource center: A costing out primer. Retrieved May
26, 2008, from
http://www.schoolfunding.info/resource_center/costingoutprimer.php3.
National Center for Education Statistics (2005). The nation’s report card: National
Assessment of Educational Progress. Retrieved April 13, 2008, from
http://nces.ed.gov/nationsreportcard/about/expenditures_teacher_ratio.asp.
National Center for Education Statistics (n.d.) Navigating resources for rural schools.
Retrieved April 14, 2008, from
http://nces.ed.gov/surveys/ruraled/TablesHTML/1bealedistricts.asp.
National Dissemination Center for Children with Disabilities (ND). IDEA. Retrieved
April 13, 2008, from http://www.nichcy.org/idea.htm.
272
National Education Association (2008). Special education and the Individuals with
Disabilities Education Act. Retrieved April 13, 2008, from
http://www.nea.org/specialed/index.html.
National Staff Development Council (2008). NSDC’s standards for staff development.
Retrieved June 22, 2008, from http://www.nsdc.org/standards/index.cfm.
Oakes, J. (n.d.). Education Inadequacy, Inequality, and Failed State Policy: A Synthesis
of Expert Reports Prepared for Williams v. State of California. Los Angeles:
UCLA.
Oates, W. (1999). An essay on fiscal federalism. Journal of Economic Literature, 37(3),
1120-1149.
Odden, A.R. (2000). The costs of sustaining educational change through comprehensive
school reform. Phi Delta Kappan, 81(6), 433-439.
Odden, A.R. (2003). Equity and adequacy of school finance today. Retrieved June 12,
2008, from http://www.EdSourceonline.com/pdf/sfequityadequacy.pdf.
Odden, A.R. (2004). Summary and reflections on 14 years of CPRE school finance
redesign research. Madison, WI: Consortium for Policy Research in Education.
Odden, A. (2007, Summer). Findings on doubling student performance from research in
school finance adequacy analyses. Poster presented at the ELPA Conference on
Doubling Student Performance, Madison, WI.
Odden, A.R., Archibald, S., Fermanich, M. & Gallagher, H.A. (2002). A cost framework
for professional development. Journal of Education Finance, 28, 51-74.
Odden, A.R., Archibald, S., Fermanich, M., & Gross, B. (2003). Defining school-level
expenditure structures that reflect educational strategies. Journal of Education
Finance, 28(3), 323-356.
Odden, A.R., Goertz, M., Goetz, M., Archibald, S., Gross, B., Weiss, M., & Mangan, M.
(2008). The cost of instructional improvement: Resource allocation in schools
using comprehensive strategies to change classroom practice. Journal of
Education Finance, 33(4), 381-406.
Odden, A.R., Mangan, M., & Picus, L.O. (2006, June 15). Level and use of resources in
Arkansas: Are use patterns consistent with doubling student performance? North
Hollywood, CA: Lawrence O. Picus and Associates.
273
Odden, A.R., Monk, D., Nakib, Y., & Picus, L.O. (1995). The story of the education
dollar: No Academy awards and no fiscal smoking guns. Phi Delta Kappan,
77(2), 161-169.
Odden, A.R. & Picus, L.O. (2008). School finance: A policy perspective (4th ed.). New
York: The McGraw-Hill Companies, Inc.
Odden, A.R., Picus, L.O., Archibald, S., Goetz, M., Mangan, M., & Aportela, A. (2007).
Moving from good to great in Wisconsin: Funding schools adequately and
doubling student performance. Madison, WI: Consortium for Policy Research in
Education.
Odden, A.R., Picus, L.O., Goetz, M., Mangan, M., & Fermanich, M. (2006). An
evidence-based approach to school finance adequacy in Washington. North
Hollywood, CA: Lawrence O. Picus and Associates.
Odden, A.R., Picus, L.O., Mangan, M., Goetz, M., & Aportela, A. (2008, January 29).
The use of educational resources in Wyoming: Updated report to the Wyoming
legislature’s joint interim education committee. North Hollywood, CA:
Lawrence O. Picus and Associates.
Pan, D., Rudo, Z, Schneider, C., & Smith-Hansen, L. (2003). Examination of resource
allocation in education: Connecting spending to student performance: Executive
summary. Austin, TX: Southwest Educational Development Laboratory.
Patton, M.Q. (1987). How to use qualitative methods in evaluation. Newbury Park, CA:
Sage Publications.
Patton, M.Q. (2002). Qualitative research and evaluation methods. Thousand Oaks,
CA: Sage Publications.
Phelan, R. (n.d.). Assessment, curriculum, and instruction: Overview of the standards
movement. Retrieved January 3, 2008, from
http://www.sonoma.edu/users/p/phelan/423/standards.html.
Picus, L.O. (1991). Cadillacs or Chevrolets? The effects of state control on school
finance in California. Los Angeles: Center for Research in Education Finance.
Picus, L.O. (1993). The allocation and use of educational resources: District level
evidence from the schools and staffing survey [Working paper]. Los Angeles,
CA: Consortium for Policy Research in Education.
274
Picus, L.O. (1994, April). The $300 billion question: How do public elementary and
secondary schools spend their money? Paper presented at the annual meeting of
the American Educational Research Association, New Orleans.
Picus, L.O. (2000). In search of more productive schools: A guide to resource allocation
in education. Eugene, OR: ERIC Clearinghouse on Educational Management.
Picus, L.O., & Blair, L. (2004). School finance adequacy: The state role. Insights on
Education, Practice, and Research, 16(March), 1-12. Southwest Educational
Development Laboratory.
Picus, L.O. & Fazal, M. (1996). Why do we need to know what money buys? Research
on resource allocation patterns in elementary and secondary schools. In L. Picus
& J. Wattenberger (Eds.), Where does the money go? Resource allocation in
elementary and secondary education (1-19). Thousand Oaks, CA: Corwin Press,
Inc.
Picus, L.O., Tetreault, D. & Murphy, J. (1996). What Money Buys: Understanding the
Allocation and Use of Educational Resources in California. Final report prepared
for the Finance Center of the Consortium for Policy Research in Education,
University of Wisconsin-Madison.
Posner, P. & Wrightson, M. (1996). Block grants: A perennial, but unstable, tool of
government. Publius, 26(Summer), 87-108.
Price, M. (2006, April). Legal and policy barriers to redesigning California schools.
Seattle, WA: Center on Reinventing Public Education.
Raywid, M.A. (1997, December; 1998, January). Small schools: A reform that works.
Educational Leadership, 55(4), 34-39. Retrieved July 5, 2008, from
http://www.ascd.org/publications/ed_lead/199712/raywid.html.
Rebell, M. (2007). Professional rigor, public engagement and judicial review: A proposal
for enhancing the validity of education adequacy studies. Teachers College
Record, 109(6), 1303-1373.
Rebell, M. (2008). Equal opportunity and the courts. Phi Delta Kappan, 89(6), 432-439.
Reynolds, A.J. & Wolfe, B. (1995). Special education and school achievement: An
exploratory analysis with a central-city sample. Educational Evaluation and Policy
Analysis, 21(3), 249-269.
275
Rivkin, S., Hanushek, E., & Kain, J. (2005). Teachers, schools, and academic
achievement. Econometrica, 73(2), 417-458.
Romeo, J. (2007, July 31). Special education state funding income research paper.
Bakersfield, CA: CBO Mentor Project. Retrieved January 22, 2009, from
wwwstatic.kern.org/gems/fcmat/11specialed6.pdf.
Rose, H., Sonstelie, J., Reinhard, R., & Heng, S. (2003). High expectations, modest
means: The challenge facing California’s public schools. San Francisco, CA:
Public Policy Institute of California.
Rosenholtz, S.J. & Smylie, M. (1984). Teacher compensation and career ladders. The
Elementary School Journal, 85(2), 149-166.
Rothstein, R. & Hawley Miles, K. (1995). Where’s the money gone? Changes in the
level and composition of education spending. Washington, D.C.: Economic
Policy Institute.
Schmoker, M. (2004). Tipping point: From feckless reform to substantive instructional
improvement. Phi Delta Kappan, 85(6), 424-432.
Schrag, P. (2005). Final test: The battle for adequacy in America’s schools. New York:
The New Press.
Slavin, R.E. & Madden, N.A. (Eds.). (2001). Success for all: Research and reform in
elementary education. Mahwah, NJ: Lawrence Erlbaum and Associates, Inc.
South Coast Air Quality Management District (2001, June 1). Nine local “clean air
heroes” to be honored at AQMD’s 13
th
annual clean air awards ceremony
(Promotion of environmental stewardship). Retrieved April 13, 2008, from
http://www.aqmd.gov/news1/2001/2001_Clean_Air_Awards_winners.htm.
Spillane, J., Halverson, R., & Diamond, J., (2001). Investigating school leadership
practice: A distributed practice. Educational Researcher, 30(3), 23-28.
Stake, R. (2002). Case Studies. In N.K. Denzin and Y. S. Lincoln (Eds.). Handbook of
Qualitative Research. Thousand Oaks, CA: Sage Publications.
Stern, S. (2006). The march of folly. In E.A. Hanushek (Ed.) Courting Failure: How
School Finance Lawsuits Exploit Judges’ Good Intentions and Harm Our
Children (1-32). Palo Alto, CA: Hoover Press.
276
Stringfield, S., Ross, S.M., Smith, L. (Eds.). (1996). Bold plans for school restructuring:
The new American schools designs. Mahwah, NJ: Lawrence Erlbaum Associates,
Inc.
The Education Trust (2005). The power to change: High schools that help all students
achieve. Washington, DC: Education Trust.
The National Center for Public Policy Research (2008). Supreme Court of the United
States: Brown v. Board of Education, 47 U.S. 483 (1954) (USSC+). Retrieved
April 13, 2008, from http://www.nationalcenter.org/brown.html.
Theobald, N. & Picus, L. (1991). Living with equal amounts of less: Experiences of
states with primarily state-funded school systems. Journal of Education Finance,
17(Summer), 1-6.
Thompson, D. & Crampton, F. (2002). The impact of school finance litigation: A long
view. Journal of Education Finance, 27(Winter), 133-172.
Tilly, W.D. (2006). Response to intervention: An overview: What is it? Why do it? Is
it worth it? The Special Edge, 19(2), 1-10.
U.S. Department of Education (1983). Findings. A nation at risk: The imperative for
educational reform. Retrieved May 25, 2008, from
http://www.ed.gov/pubs/NatAtRisk/findings.html.
U.S. Department of Education (1995). Systemic reform in the professionalism of
educators: Executive summary. Retrieved May 24, 2008, from
http://www.ed.gov/pubs/SER/ProfEd/execsum.html.
U.S. Department of Education (2004). Comprehensive school reform program: About
CSR. Retrieved April 20, 2008, from
http://www.ed.gov/programs/compreform/2pager.html.
U.S. Department of Education (2005). Elementary and secondary education: Part A –
Improving basic programs operated by local educational agencies. Retrieved
April 13, 2008, from
http://www.ed.gov/policy/elsec/leg/esea02/pg2.html#sec1116.
U.S. Department of Education (2006, February 6). Press Releases: Fiscal year 2007
budget request advances NCLB implementation and pinpoints competitiveness.
Retrieved April 13, 2008, from
http://www.ed.gov/news/pressreleases/2006/02/02062006.html.
277
U.S. Department of Education (2007). Overview: 10 facts about k-12 education funding
[Electronic version].
U.S. Supreme Court Center (n.d.). Crawford v. Los Angeles Board of Educ., 458 U.S.
527 (1982). Retrieved April 13, 2008, from
http://supreme.justia.com/us/458/527/case.html.
Vandell, D., Pierce, K., Brown, B. (2002). Study of promising after-school programs.
Wisconsin Center for Education Research. Retrieved January 23, 2009, from
http://childcare.wceruw.org/des3.html.
Waits, M., Campbell, H., Gau, R., Jacobs, E., Rex, T., Hess, R. (2006). Why some
schools with Latino children beat the odds…and others don’t. Phoenix, AZ:
Center for the Future of Arizona.
Weissert, C. & Schram, S. (1996). The state of American federalism, 1995-1996.
Publius, 26(Summer), 1-26.
Williams, J. (2008, February 27). Schwarzenegger wants interventions for failing school
districts. Press Enterprise. Retrieved March 20, 2008, from
http://www.pe.com/ap_news/California/CA_Failing_Schools_329456C.shtml#.
Yoon, K.S., Duncan, T., Lee, S., Scarloss, B., & Shapley, K. (2007). Reviewing the
evidence on how professional development affects student achievement. Issues
and Answers Report, REL 2007-No. 033. [Electronic version]. Washington,
D.C.: U.S. Department of Education, Institute of Education Sciences, National
Center for Education and Regional Assistance, Regional Laboratory Southwest.
Young, B. (1999). Characteristics of the 100 largest public elementary and secondary
school districts in the United States: 1997-98. Washington, DC: U.S. Department
of Education, National Center for Educational Statistics (NCES 1999-318).
278
APPENDIX A
Informed Consent
September 25, 2008
The undersigned voluntarily consents to participate in the study conducted by Deborah Granger titled
“Allocation of Educational Resources to Improve Student Learning: Case Studies of California Schools.”
Abstract: California education professionals today face increasing pressure to improve student learning of
their students. Despite a number of performance gains in recent years, more needs to be done. The release
of the studies comprising the Getting Down to Facts project has focused attention on the question of school
finance, and in particular how much money schools and districts need to provide educational programs that
will give all California school children an opportunity to meet our state’s high performance standards.
Using current models to estimate school finance adequacy, the researcher will identify a group of up to ten
schools that meet established criteria related to student characteristics and performance. Using qualitative
and quantitative methodologies that include in-person interviews and document analysis, the researcher will
seek to ascertain how the resources available at each school compare with what a selected adequacy model
recommends to achieve California’s educational performance standards. The researcher also will assess
how those schools utilize their resources (personnel and dollars) to achieve results. This study will add to
body of knowledge of surrounding California school finance as well as provide practitioners with practical
knowledge of how educational resources might be allocated to help all students succeed.
Participation in this study entails responding to a series of structured interview questions as well as
providing documents to assist with collection of data regarding allocation of (primarily) human resources at
schools in California. The researcher hopes to contribute to the field of school finance adequacy in efforts
to determine if a dollar amount can be attached to student academic achievement and proficiency as
determined by the state and federal government.
All participants, their schools, and their districts will remain anonymous. This school and its district have
been assigned aliases.
The principal will direct the researcher to appropriate individuals and resources as needed to obtain a
complete data set.
With the principal’s permission, the investigator will contact the district superintendent to thank the district
for its support of this academic research project.
Upon conclusion of the study, the researcher will send the principal a copy of the findings.
All participation in this research study is voluntary. Participants receive no compensation for their
participation.
_______________________________________________________________
Principal Date
279
APPENDIX B
Invitation to Participate
June 12, 2008
MEMO
To: School Resource Allocation Use Study – Potential Participants
From: Deborah Granger, Researcher
RE: Research Study Information and Site Permission Letter
To everyone with whom I have had the opportunity to personally speak, THANK YOU for your preliminary
agreement to participate in my dissertation study. This memo provides a brief description of the topic, the study,
and the site permission letter (template enclosed).
I have the privilege of working with Larry Picus, Ph.D. at the University of Southern California, a nationally
recognized expert in the area of school finance adequacy – the topic of my dissertation. The basic premise is that
we have a body of research about effective instructional practices along with measurable academic achievement
standards, definitions of academic proficiency, and an expectation that all students will achieve those levels of
proficiency. At the same time, we have school funding systems that were not based on any of these performance
expectations. I hope to contribute to research efforts that have begun paint a picture of what an adequate
education looks like by identifying resource use (e.g., FTEs teaching English, instructional minutes, FTEs
serving as classroom aides, etc.) at Program Improvement middle schools. The ultimate hope is to figure out
how much it costs to provide every child with the resources s/he needs in order to achieve proficiency.
This fall I’d like to conduct a structured interview with you and/or appropriate staff members (e.g., Title I
coordinator, assistant principal(s), department chairs – as you see fit). The interview follows a specific format
and includes examination of a variety of documents (e.g., master schedule, bell schedule, school improvement
plan, budgets). I hope to conduct the data-collection interviews in October and November. All schools remain
anonymous. The data collected are public record types of information.
In order to receive approval for this study, I must turn in “Site Permission Letters” for each potential site to the
Institutional Review Board. Enclosed, please find a template for the letter. Would you be so kind as to transfer
the text to your school’s letterhead, sign the letter, and return to me in the return envelope also enclosed? If
you would like me to send you the text electronically, please be so kind as to send me an email and I will send it
to you right away.
After confirming receipt of the “Site Permission Letter,” I do not anticipate troubling you again until end-of-
summer/early fall to set an appointment. However, if you would like to schedule earlier, please let me know
your preferred date(s). As I mentioned in our conversations, the interview is lengthy – part of the reason you
may be inclined to include additional staff members – I know and respect how busy you are!
Thank you again for this initial commitment to help me complete my Ed.D. while contributing to research that
turns out to be extremely timely.
Please do not hesitate to contact me with any thoughts or questions! I can be reached as follows: email:
dgranger@usc.edu, mobile phone: XXX-XXX-XXXX, work phone: XXX-XXX-XXXX. It will be an honor and
a privilege to work with you! Many thanks!
280
APPENDIX C
School Visit/Interview Dates
School
Visit/Interview Date
Van Gogh Middle School September 16, 2008
Renoir Intermediate School September 25, 2008
Chagall Junior High School October 3, 2008
Warhol Junior High School October 10, 2008
Degas Intermediate School October 13, 2008
Botticelli Junior High School October 21, 2008
Picasso Middle School October 24, 2008
Rodin Middle School October 29, 2008
281
APPENDIX D
List of Documents and Artifacts Provided by Participating Schools
• Master Schedule of Courses and Teachers
• Bell Schedules
• Meeting and Professional Development Calendars
• District Professional Development Catalogues/Schedules
• Classified Staff Lists – Roles and Hours
• School Plans for Student Achievement
• Categorical Budget and Expenditure Items
• Program Improvement Update Presentations Provided to School Boards
• Principal’s Research and Rationale for Instructional Improvement Strategies
• Extracurricular/Club Sponsorship Lists
• Extracurricular Club/Sports Schedule
282
APPENDIX E
Data Collection Codebook
This Codebook identifies data collection items and their definitions. This document is
organized according to the corresponding Data Collection Protocol and the Web portal
for data entry (www.lopassociates.com).
Each data item has a place for notes. This section is meant to be used for any
notations that you would like to record as a personal reminder. Notes fields will
not be used in data analysis.
I. School Profile
A. School Name: Each researcher has developed his/her own criteria to identify
California schools to include in this study.
B. School State ID: This is the CDS identification number that the state has
assigned the school.
C. Address Line 1: Street address of the school
D. Address Line 2: (optional) Second line of street address of the school
E. City: City of the school
F. State: “CA”
G. Zip: Postal zip code of the school
H. Phone: Main office phone number for the school
I. Fax: Main office fax number for the school
J. Website: School’s official Website
I. School Contacts
This section is for recording the contact people at the school. This will include
the principal, and most likely the secretary. Anyone else you interview should
also be recorded here. Any notes you’d like to make about this person (e.g.,
phonetic spelling of their name) should go in the notes sections, as well as what
the data source is.
1. Title: The job title of the person who you interview from the school
2. Honorific: Mr., Mrs., Ms., Dr., Rev., etc.
3. First Name: Formal first name of school staff member (e.g., Michael
instead of Mike)
4. Initial: (optional) Middle initial of school staff member
5. Last Name: Surname of school staff member
6. Suffix: (optional) Jr., etc.
7. Phone #: Direct phone number to the school staff member
8. Fax #: Fax number for the school staff member
9. Email Address: Preferred email address of the school staff member
283
10. Mail Address: Street address of the contact person
11. Address Line 2: (optional) Second line of street address of the contact
person
12. City: City of the contact person
13. State: “CA”
14. Zip Code: Postal zip code of the contact person
15. Zip + 4: Four digit extension of the zip code
II. District Profile
1. District Name: This is the name of the district where the school is located.
2. District State ID: This is the identification number that the state has assigned
to the district within which the school resides.
III. District Contacts
This section is for recording the contact people at the district office. This will
include the superintendent, and possibly an assistant superintendent and/or
director of curriculum and instruction. Anyone else you interview should also be
recorded here. Any notes you’d like to make about this person (e.g., phonetic
spelling of their name) should go in the notes sections, as well as what the data
source is.
1. Title: The job title of the person who you interview from the school
2. Honorific: Mr., Mrs., Ms., Dr., Rev., etc.
3. First Name: Formal first name of school staff member (e.g., Michael
instead of Mike)
4. Initial: (optional) Middle initial of school staff member
5. Last Name: Surname of school staff member
6. Suffix: (optional) Jr., etc.
7. Phone #: Direct phone number to the school staff member
8. Fax #: Fax number for the school staff member
9. Email Address: Preferred email address of the school staff member
10. Mail Address: Street address of the contact person
11. Address Line 2: (optional) Second line of street address of the contact
person
12. City: City of the contact person
13. State: “CA”
14. Zip Code: Postal zip code of the contact person
15. Zip + 4: Four digit extension of the zip code
IV. School Resource Indicators
School resource indicators should be collected for the 2008-09 school year. Enter
personal notations pertaining to the data in the yellow notes fields.
284
1. Current Student Enrollment: Headcount of students enrolled at the school
on the day of the site visit minus any pre-kindergarten students.
2. Pre-kindergarten Student Enrollment: Headcount of students enrolled in
any pre-kindergarten programs at the school on the day of the site visit.
These students should not be included in the previous category, Current
Student Enrollment. Make sure to also ask this question at secondary
schools.
3. Grade Span: Range of grades (e.g., K-5).
4. Number of ELL/Bilingual Students: As of the day of the site visit, the
number of students eligible for services as an English language learner
(ELL) as defined by the federal No Child Left Behind Act of 2001
(http://www.k12.wy.us/FP/title3/Wy_ELD_ELA.pdf).
5. Number of Students Eligible for Free- or Reduced-Price Lunch (FRL):
Number of enrolled students who are eligible for the federal free- and
reduced-price lunch program.
6. Total number of Special Education Students (IEPs): As of the day of the
site visit, number of students in the school with an Individualized
Education Plan (IEP) indicating their eligibility for special education
services (This will most likely be a larger number than the number of
students who are in a self-contained special education classroom.). Does
not include gifted and talented students.
7. Number of Special Education Students (self-contained): Number of
students in the school with an Individualized Education Plan (IEP)
indicating their eligibility for special education services and who learn
primarily (at least 60% of the school day) outside of a regular education
classroom.
8. Total Length of School Day: Number of minutes per day that students are
required to be present at school. If multiple grade spans are present for
different amounts of time, report the average length (e.g., If the school day
begins at 8:30 AM and ends at 3:15 PM, then the total length of the school
day is 405 minutes.).
9. Length of Instructional Day: Number of minutes per day that students are
present for instruction. This information should be available from the
school bell schedule or a school staff member. Subtract recess, lunch, and
passing periods time from the total minutes in the school day. This
calculation is different from how the state measures the “instructional
day” (e.g., If the length of the school day is 405 minutes, and the students
have 20 minutes for lunch and 25 minutes for recess, then the length of the
instructional day is 360 minutes.).
10. Length of Mathematics Class: Number of minutes of mathematics class
periods per day. These include periods when students are specially
grouped for extended mathematics instruction. Report an average per day
length.
285
11. Length of Reading/English/LA Class: Number of minutes of reading,
English, and language arts (LA) class periods. These include periods when
students are specially grouped for extended literacy instruction (e.g.,
reading, writing, comprehension). Report an average per day length.
12. Length of Science Class: Number of minutes of science class periods per
day. These include periods when students are specially grouped for
extended science instruction. Report an average per day length.
13. Length of Social Studies Class: Number of minutes of social studies and
history class periods per day. These include periods when students are
specially grouped for extended history or social studies instruction. Report
an average per day length.
14. AYP: This is a measure as to whether the school made Adequate Yearly
Progress (AYP) during the previous school year (2007-08). Enter “Y” or
“N” or “NA.”
V. Core Academic Teachers
The classroom teachers primarily responsible for teaching a school’s core
academic subjects of reading/English/language arts, mathematics, science,
history/social studies, and foreign language. In elementary schools, core
academic teachers consist of the teachers in the self-contained regular education
classrooms. Some elementary schools may also departmentalize certain core
subjects such as math or science, especially in the upper grades. These teachers
are also to be included as core teachers. In middle schools, high schools, or any
other departmentalized school, core teachers consist of those teachers who are
members of the English/language arts, mathematics, science, social studies, and
foreign language departments along with special education or ESL/bilingual
teachers who provide classes in these subjects. The teachers should be entered as
full-time equivalents (FTEs), which may include decimals (e.g., a half-time
teacher would be entered as 0.5). If teachers are assigned to multiage
classrooms, divide up the FTEs weighted by students per each grade. Enter each
teacher’s name that corresponds to the FTEs entered in the corresponding notes
fields. Indicate in parentheses if the teacher is not a 1.0 FTE in that category.
Example:
Grade 1: Matthew Perry (0.5), Lisa Kudrow, Jennifer Aniston;
Grade 2: David Schwimmer (0.25), Courteny Cox Arquette (0.33), Matt
LeBlanc
A. Grades K-12: Number of FTE licensed grade-level teachers who teach the
core subjects. The FTEs should not duplicate those in the individual
subject categories.
B. English/Reading/LA, History/Social Studies, Mathematics, Science, and
Foreign Language: Number of FTE licensed subject-specific teachers who
teach the core subjects. The FTEs should not duplicate those in the grade
categories.
286
VI. Specialist and Elective Teachers
This expenditure element consists of teachers who teach non-core academic
classes, and usually provide planning and preparation time for core academic
teachers. The teachers should be entered as full-time equivalents (FTEs), which
may include decimals. In the notes sections, enter each teacher’s name that
corresponds to the FTEs entered in the related fields. Indicate in parentheses if the
teacher is not a 1.0 FTE in that category.
1. Art/Music/PE: Number of FTE specialist teachers, such as art, music, and
physical education (PE) teachers, who usually provide regular classroom
teachers with planning and preparation time.
2. Drama/Technology/Health: Number of FTE teachers who provide
instruction in a subject area that represents a special academic focus.
3. Career & Technical Education: Number of FTE vocational education
teachers
4. Driver Education: Number of FTE drivers education teachers.
5. Study Hall: Number of FTE teachers who monitor study hall.
6. Athletics: Number of FTE teachers who coach an athletic team during the
school day. This does not include time spent as an athletic director, which
would be captured under the Administration section.
7. Other: Number of FTE specialist teachers who are not specifically listed
above.
8. Other Description: Indicate the subject area that the “Other” specialist
teacher(s) instruct.
VII. Library Staff
Library staff should be entered as full-time equivalents (FTEs), which may
include decimals. Enter each staff member’s name that corresponds to the FTEs
entered in the related fields. Indicate in parentheses if the staff member is not a
1.0 FTE in that category.
1. Librarian/ Library Media Specialist: Number of FTE licensed librarians or
media specialists who instruct students
2. Library Aide: Number of FTE library aides who help instruct students
VIII. Extra Help Staff
This category mainly consists of licensed teachers from a wide variety of
strategies designed to assist struggling students, or students with special needs, to
learn a school’s regular curriculum. The educational strategies that these teachers
deploy are generally supplemental to the instruction of the regular classroom.
Extra help staff should be entered as full-time equivalents (FTEs), which may
include decimals. Do not include volunteers in the FTE counts. Enter each staff
287
member’s name that corresponds to the FTEs entered in the related fields.
Indicate in parentheses if the staff member is not a 1.0 FTE in that category.
1. Certified Teacher Tutors: Number of FTE tutors who are licensed teachers
and provide help to students one-on-one or in small groups of 2-5.
2. Non-Certified Tutors: Number of FTE tutors who are not licensed teachers
and provide help to students one-on-one or in small groups of 2-5.
3. ISS Teachers: Number of FTE licensed teachers who monitor/teach In-
School Suspension (ISS) students.
4. ISS Aides: Number of FTE Title I funded aides who monitor/teach In-
School Suspension (ISS) students.
5. Title I Teachers: Number of FTE non-special education teachers who
provide small groups of students with extra help as a function of the Title I
program.
6. Title I Aides: Number of FTE non-special education aides who provide
small groups of students with extra help as a function of the Title I
program.
7. ELL Class Teachers: Number of FTE licensed teachers of English as a
second language (ESL) who work with non-English speaking students to
teach them English.
8. Aides for ELL: Number of FTE aides of English as a second language
(ESL) classes who work with non-English speaking students to teach them
English.
9. Gifted Program Teachers: Number of FTE teachers who instruct students in
the gifted program.
10. Gifted Program Aides: Number of FTE aides who instruct students in the
gifted program.
11. Gifted Program Funds: Dollar amount budgeted for the gifted program for
the 2008-09 school year
12. Other Extra Help Teachers: Number of FTE teachers who provide
supplemental instructional assistance to students to learn the school’s
curriculum (Use this category sparingly.).
13. Other Extra Help Teachers Description: Indicate what the “Other” extra
help staff do.
14. Other Extra Help Classified Staff: Number of FTE classified staff who
provide supplemental instructional assistance to students to learn the
school’s curriculum (Use this category sparingly.)
15. Other Extra Help Classified Staff Description: Indicate what the “Other”
extra help classified staff provide.
288
16. Special Ed. Teacher (Self-contained for students with severe disabilities):
Number of FTE licensed teachers who teach in self-contained special
education classrooms and work with “severely” disabled students for most
or all of the school day. These teachers may teach a modified version of a
school’s curriculum or other learning goals required by their students’
Individualized Education Programs (IEPs).
17. Special Ed. Inclusion Teachers: Number of FTE licensed teachers who
assist regular classroom teachers with mainstreamed students who have
physical or mental disabilities, or a learning problem. These students
generally have “less severe” disabling conditions.
18. Special Ed. Resource Room Teachers: Number of FTE licensed special
education teachers who provide small groups of students in special
education with extra help in specific areas.
19. Special Ed. Self-contained Aides: Number of FTE aides who assist in self-
contained special education classrooms and work with “severely” disabled
students for most or all of the school day.
20. Special Ed. Inclusion Aides: Number of FTE aides who assist regular
classroom teachers with mainstreamed students who have physical or
mental disabilities, or some learning problem. These students generally
have “less severe” disabling conditions.
21. Special Ed. Resource Room Aides: Number of FTE special education aides
who provide small groups of students in special education with extra help
in specific areas.
22. Number of Extended Day Students: Number of students who participate in
the extended day program.
23. Minutes per Week of Extended Day Program: Number of minutes per week
that the extended day program is offered.
24. Teacher Contract Minutes per Week: Number of work minutes per week in
the teacher contract.
25. Extended Day Teachers: Number of FTE licensed teachers who provide
students with extra instructional time to achieve to the standards in the
regular curriculum after school.
26. Extended Day Classified Staff: Number of FTE staff members who provide
students with extra instructional time to achieve to the standards in the
regular curriculum after school.
AA. Description of Extended Day Classified Staff: Description of classified
staff members’ roles in the extended day program.
BB. Minutes Per Week of Summer School: Number of minutes per day
multiplied by the number of days per week that students attend summer
school.
289
a. Length of Session: Number of weeks that summer school is in session.
b. School’s Students Enrolled in the Summer School Program: Number
of students from the individual school who are enrolled in the summer
school program (a subset of the following item).
c. All Students in Summer School: Total number of students enrolled in
the summer school program.
d. Summer School Teachers: Number of FTE teachers who provided
students with extra instructional time to achieve to the standards in the
regular curriculum during summer 2008.
CC. Summer School Classified Staff: Number of FTE classified staff who
provided students with extra instructional time to achieve to the standards
in the regular curriculum during summer 2008.
IX. Other Instructional Staff
Included here are instructional staff members that support a school’s instructional
program, but do not fit in the previous categories. Other instructional staff should
be entered as full-time equivalents (FTEs), which may include decimals. Enter
each staff member’s name that corresponds to the FTEs entered in the related
fields. Indicate in parentheses if the staff member is not a 1.0 FTE in that
category.
1. Consultants (other than PD contracted services): Dollar amount for all
other consultants other than professional development contracted services.
2. Building Substitutes: Number of FTE permanent substitutes.
3. Other Teachers: Number of FTE teachers who instruct, but were not
included in previous categories.
4. Other Instructional Aides: Number of FTE aides who assist instruction, but
were not included in previous categories.
5. Funds for Daily Subs: Daily rate for certified teacher substitutes who
replace sick teachers. (This is not for substitutes who replace teachers who
are participating in professional development.)
X. Professional Development Staff & Costs
This expenditure element includes spending on the professional development of a
school’s staff and the staffing resources necessary to support it. Professional
development staff should be entered as full-time equivalents (FTEs), and cost
figures should be entered as a dollar amount, both of which may include decimals.
Enter each staff member’s name that corresponds to the FTEs entered in the
related fields. Indicate in parentheses if the staff member is not a 1.0 FTE in that
category.
290
1. Number of Professional Development Days in the Teacher Contract:
Number of days the teacher contract specifies for professional
development.
2. Substitutes and Stipends (teacher time): Dollar amount budgeted for
substitutes and stipends that cover teacher time for professional
development. For time outside the regular contract day when students are
not present before or after school or on scheduled in-service days, half days
or early release days, the dollar amount is calculated by multiplying the
teachers’ hourly salary times the number of student-free hours used for
professional development. For planning time within the regular contract,
the dollar amount is calculated as the cost of the portion of the salary of the
person used to cover the teachers’ class during planning time used for
professional development. For other time during the regular school day,
including release time provided by substitutes, cost is calculated with
substitute wages. For time outside the regular school day, including time
after school, on weekends, or for summer institutes, the dollar amount is
calculated from the stipends or additional pay based on the hourly rate that
the teachers receive to compensate them for their time.
3. Instructional Facilitators/Coaches: Number of FTE instructional facilitators
and coaches. This may include on-site facilitators and district coaches
(though only the FTE for the specific school should be recorded). Outside
consultants who provide coaching should be captured in an estimated FTE
amount depending on how much time they spend at the school.
4. Trainers/Consultants: Dollar amount for outside consultants who provide
training or other professional development services. If trainers are from
the district, convert to a dollar amount.
5. Administration: Number of FTE district or school-level administrators of
professional development programs. (Again, only the FTE for the specific
school should be recorded).
6. Travel: Dollar amount of the costs of travel to off-site professional
development activities, and costs of transportation within the district for
professional development.
7. Materials, Equipment, and Facilities: Dollar amount of the materials for
professional development including the cost of classroom materials,
equipment needed for professional development activities, and rental or
other costs for facilities used for professional development.
8. Tuition & Conference Fees: Dollar amount of tuition payments or
reimbursement for college-based professional development, and fees for
conferences related to professional development.
9. Other Professional Development: Either FTEs or Dollar amount for other
professional development staff or costs. (Use this category sparingly.)
10. Other Description: Specify what the “Other” professional development is,
and indicate whether it is a FTE or dollar amount.
291
XI. Student Services Staff
This expenditure element consists of school-based student support staff, as well as
school expenditures for extra-curricular activities and athletics. Student services
staff should be entered as full-time equivalents (FTEs), which may include
decimals. Enter each staff member’s name that corresponds to the FTEs entered in
the related fields. Indicate in parentheses if the staff member is not a 1.0 FTE in
that category.
1. Guidance: Number of FTE licensed guidance counselors.
2. Attendance/dropout: Number of FTE staff members who manage
attendance and report dropouts.
3. Social Workers: Number of FTE licensed school social workers.
4. Nurse: Number of FTE registered nurses or nurse practitioners
5. Parent advocate/community liaison: Number of FTE staff members who
serve as the parent advocate and/or community liaison, often working with
parents to get their children to attend school.
6. Psychologist: Number of FTE licensed school psychologists or educational
diagnosticians.
7. Speech/OT/PT: Number of FTE licensed speech, occupational (OT), and
physical therapists (PT) who provide services to the school’s students
8. Health Asst.: Number of FTE health assistants
9. Non-teaching aides: Number of FTE non-teaching aides. (e.g., Lunchroom
aides, Aides who help students board buses; DO NOT include cooks – the
defining difference is whether the staff member is supervising students or
not.)
10. Other Student Services: Number of FTE other student services staff. (Use
this category sparingly.)
11. Other Description: Indicate what the “other” student services staff member
does.
XII. Administration
This expenditure element consists of all staffing resources pertaining to the
administration of a school. Administrators should be entered as full-time
equivalents (FTEs), which may include decimals. Enter each staff member’s name
that corresponds to the FTEs entered in the related fields. Indicate in parentheses
if the staff member is not a 1.0 FTE in that category.
1. Principal: Number of FTE licensed principals.
2. Assistant Principal: Number of FTE assistant principals.
3. Other Administrators: Number of FTE other administrators (Use this
category sparingly.).
4. Other Description: Indicate what the “Other” administrators’ duties are.
292
5. Secretary: Number of FTE Secretaries. (12 month employees)
6. Clerical Staff: Number of FTE clerical staff members (10 month
employees).
7. Technology Coordinator: Number of FTE technology coordinators and IT
staff.
8. Security: Number of FTE security staff.
9. Custodians: Number of FTE staff who provide custodial services
XIII. Elementary Class Sizes (We are NOT collecting this data for middle and high
schools.)
Sometimes it is easiest to get this information when you get the staff list, but other
times the secretary can just copy the sheet that tells them how many students are
in each classroom (we don’t want student names). You want a (preferably
electronic) copy of the master class schedule to enter this data. When entering
the data online, make sure to enter the class size for every class that is taught at
the school. Click on the Class Size option from the main menu and a new menu
will be displayed on the left. This menu will have options for grades Pre-8 plus
Special Education. When you click on a grade, the page with that grade's sections
will be displayed where you can enter the individual class sizes.
293
APPENDIX F
OPEN-ENDED DATA COLLECTION PROTOCOL – SCHOOL SITES
Following are open-ended questions intended to capture each school’s strategies for improving
student performance. Ask the questions in the order that they appear on this protocol. Record
the principal’s answers as s/he gives them and focus on getting the key elements of the
instructional improvement effort with less emphasis on the process aspect.
I. Tell me the story of how your school improved student performance.
A. What were the curriculum and instruction pieces of the strategy?
1. What has been the content focus of your improvement process?
(e.g., Reading, Math, Reading First, Math Helping Corps, etc.)
2. What curricula have you used during your instructional improvement effort?
(e.g., Open Court reading, Everyday Math, etc.)
• Is it aligned with state standards?
• How do you know it is aligned? (e.g., District recent review for alignment)
3. What has been the instructional piece of your improvement effort?
• Does your staff have an agreed upon definition of effective teaching?
4. What is the instructional vision for your improvement effort?
(e.g., Connecticut standards or the Danielson Framework)
5. Have assessments been an integral part of your instructional improvement
process?
• If so, what types of assessments have been key? (e.g., formative,
diagnostic, summative)
• How often are those assessments utilized?
• What actions were taken with the results?
6. What type of instructional implementation has taken place as a part of your
reform efforts? (e.g., Individualized instruction, differentiated instruction, 90
minutes of uninterrupted reading instruction)
• Were teachers trained in a specific instructional strategy?
• How did you know that the instructional strategies were being
implemented?
294
B. What were the resource pieces of the strategy? How long have the
resources been in place?
1. Early Childhood program: Is it half or full day? Number kids? Staffing
ratios? Eligibility? *
2. Full Day Kindergarten
• If yes, how long have they had full day kindergarten?*
3. Class Size Reduction
• Reduction Strategy (e.g., 15 all day K-3 or reading only with 15)*
4. Professional Development: (e.g., Summer Institutes, Inservice Days)
• What is the focus of the professional development?
• When are the professional development days scheduled?
• Do you have instructional coaches in schools? Were there enough
coaches? (Did they need more but couldn’t afford it?)
5. “Interventions” or Extra Help Strategies for Struggling Students:
• Tutoring: Specify 1:1, small groups (2-4), or medium groups (3-5)
• Extended day: How frequently (Number minutes & Number of times per
week), Academic focus, Who instructs (certified teachers or aides), Who
participates
• Summer school: How Frequently (Number hours a day, Number weeks),
Who instructs (certified teachers or aides), Who participates
• ELL
• Scheduling: (e.g., double periods in secondary schools)
6. Parent outreach or community involvement
7. Technology
C. Was the improvement effort centrist (central office orchestrated) or bottom
up?
D. What type of instructional leadership was present?
E. Was there accountability built into this improvement plan? (e.g., School
Board report which helped solidify focus)
F. What additional resources would be needed to continue and expand your
efforts?
* Not applicable at the middle school level.
295
APPENDIX G
Data Collection Protocol
School Profile
School Name School
Pseudonym
Address
City State Zip
CA
Phone Fax
Website
NOTES:
School Contact (1)
Title
Principal
Honorific First Name Last
Phone # Fax #
Email Address
NOTES:
296
School Contact (2)
Title
Honorific First Name Last
Phone # Fax #
Email Address
NOTES:
School Contact (3)
Title
Honorific First Name Last
Phone # Fax #
Email Address
NOTES:
District Profile
District Name
District State ID
297
District Contact (1)
Title
Superintendent
Honorific First Name Last
Phone # Fax #
Email Address
NOTES:
District Contact (2)
Title
Honorific First Name Last
Phone # Fax #
Email Address
NOTES:
District Contact (3)
Title
Honorific First Name Last
Phone # Fax #
Email Address
NOTES:
298
School Resource Indicators
Current Student Enrollment
Pre-Kindergarten Student Enrollment
Grade Span
Number of ELL Students
Number of Students Eligible for Free- and Reduced-
Price Lunch (FRL)
Total Number of Special Education Students (IEPs)
Number of Special Education Students (self-contained)
Total Length of School Day
Length of Instructional Day
Length of Mathematics Class
Length of Reading or English/LA Class
Length of Science Class
Length of Social Studies Class
Length of Foreign Language Class
AYP
NOTES:
299
Core academic teachers
(Self-contained/
Regular Education)
FTEs
Kindergarten
(Full day program)
Grade 1
Grade 2
Grade 3
Grade 4
Grade 5
Grade 6
Grade 7
Grade 8
Grade 9
Grade 10
Grade 11
Grade 12
English/Reading/L.A.
History/Soc. Studies
Math
Science
Foreign Language
NOTES:
300
Specialist and Elective Teachers/
Planning and Prep
FTEs
Art
Music
PE/Health
Drama
Technology
Career & Technical Education
Drivers Education
Study Hall
Athletics
Other Specialist & Elective Teachers
Other Specialist & Elective Teachers
Description:
NOTES:
Library Staff FTEs
Librarian
Library Media Specialist
Library Aide
NOTES:
301
Extra Help I FTEs or Dollars ($)
Certified Teacher Tutors
Non-Certified Tutors
ISS Teachers
ISS Aides
Title I Teachers
Title I Aides
ELL Class Teachers
Aides for ELL
Gifted Program Teachers
Gifted Program Aides
Gifted Program Funds
$
Other Extra Help Teachers
Other Extra Help Classified Staff
NOTES:
302
Extra Help II FTEs
Special Ed. Teacher (Self-contained for severely
disabled students)
Special Ed. Inclusion Teachers
Special Ed. Resource Room Teacher
Special Ed. Self-contained Aides
Special Ed. Inclusion Aides
Special Ed. Resource Room Aides
NOTES:
Extra Help III
Number of Extended Day Students
Minutes per Week of Extended Day Program
minutes
Teacher Contract Minutes per Week
minutes
Extended day Teachers
Extended Day Classified Staff
Description of Extended Day Classified Staff
Minutes per Week of Summer School
minutes
Length of Session (# of Weeks)
weeks
School’s Students Enrolled in Summer School
All Students in Summer School
Summer School Teachers
Summer School Classified Staff
NOTES:
303
Other Instructional Staff FTEs and Dollars ($)
Consultants
(other than PD contracted services)
$
Building substitutes and other substitutes
Other Teachers
Other Instructional Aides
Funds for Daily Subs
$
NOTES:
Professional Development
Dollars ($) and FTEs
Number of Prof. Dev. Days in Teacher
Contract
Substitutes and Stipends (teacher time)
$
Instructional Facilitators/Coaches
Trainers/Consultants
$
Administration
Travel
$
Materials, Equipment and Facilities
$
Tuition & Conference Fees
$
Other Professional Development
$
Other Professional Development Staff
Funded with Federal Dollars:
NOTES:
304
Student Services
FTEs
Guidance
Attendance/Dropout
Social Workers
Nurse
Parent advocate/community liaison
Psychologist
Speech/O.T./P.T.
Health Asst.
Non-teaching aides
Other Student Services
Description Of Other Student Services
Staff:
NOTES:
305
Administration
FTEs
Principal
Assistant principal
Other Administrator
Description of Other Administrator:
Secretary
Clerical staff
Technology Coordinator/ I.T.
Security
Custodians
NOTES:
Elementary School Class Sizes
Section 1 Section 2 Section 3 Section 4
Special
Education
Grade 1
Grade 2
Grade 3
Grade 4
Grade 5
Grade 6
Grade 7
Grade 8
306
APPENDIX H
Analysis Process
Data analysis occurred in stages. After cleaning the data, the first stage focused
on analysis of data for each individual school. The second stage highlighted cross-school
findings and facilitated response to the four research questions that guided this
dissertation. Both stages relied on analysis of quantitative and qualitative data.
Cresswell’s (2003) six steps (transcribe, read for meaning, create categories, generate
descriptions with themes, prepare a narrative, attach meaning and identify
themes/challenges/future considerations) for analyzing qualitative data guided that part of
the process at each stage of the data-analysis process.
The first stage in quantitative data analysis focused on a comparison of current
resource allocations in a given school with recommendations based on the Evidence-
Based Model (EBM). Comparing quantitative data with EBM suggestions and by
relating qualitative interview data with research- and evidence-based understandings
regarding school improvement facilitated consideration of implications for each school.
The second stage following completion of all interviews, inputting of all data, and
final cleaning of all data entered. The researcher exported all entered data in the form of
an Excel spreadsheet and conducted additional calculations to analyze resource use
across the eight schools included in the case-study series. Further explanation of the
process and calculations necessary for each stage of the quantitative analysis follows
sequential headings below.
307
Stage I: School – EBM Comparison Calculations
In order to compare each case study with resource allocation recommendations
based on the EBM, the researcher referred to data collected during interviews and follow-
ups. Application of the calculations described below resulted in suggested resource
allocation numbers in terms of Full Time Equivalents (FTEs) or dollars. Additional data-
based calculations also described below resulted in identifying current resource
allocations for side-by-side comparison with EBM suggestions as illustrated in the
comparison table included in each case study write-up.
All resource-allocation calculations based on the EBM are grounded in the
research and evidence undergirding the model. This research and evidence base is
described at length in Chapters Two and Three of this dissertation.
Current-allocation calculations requiring an explanation of the rationale include a
short explanatory note immediately following the calculation.
Calculations
1. School Size - EBM Comparison: Actual Number of Students ÷ 450
2. Core Class Size: Number of Students ÷ Number of (English/language arts +
Mathematics + History-social science + Science FTE Teachers). This does
not include self-contained special education classes. This does include
language arts intervention and English language development class teachers.
3. Specialist and Elective Teachers: Number of FTE Elective Teachers (i.e..,
AVID, foreign language, computers, physical education, study skills, teen
skills, etc.).
4. Total Teachers: Core FTE Teachers + Specialist/Elective FTE Teachers.
5. Instructional Coach Recommendation: Number of Students ÷ 200. Rounded
to nearest 10
th
.
6. Tutor Recommendation: Number of Students Participating in Free/Reduced
Price Lunch Program ÷ 100.
7. English Learner Teacher Recommendation: Number of English learners ÷
100. Rounded to nearest 10
th
.
308
8. Summer School Recommendation: (Number of Students Participating in
Free/Reduced Price Lunch Program ÷ 2) ÷ 15 ÷ 4* = FTE. (*4 represents a
proxy for ¼ of a school year, the approximate amount of time represented by
a recommended summer school session.)
9. Summer School Teacher Calculation: (Summer School Minutes ÷ Contract
Minutes) x (Length of One Session in Weeks ÷ 18)* x Number of Teachers =
FTE. (*One course taken during a single session of summer school typically
represents an equivalent course taken in a traditional school semester.)
10. Services for Gifted Students: $25 x Number of Students.
11. Career/Technical Education: $7000 x FTE = Dollar Amount to Add to
Teacher Contract. Class loads were not counted in this study. The class size
calculation may be made as follows: Average class size ÷ 1.3 =
Recommended CTE Class Size.
12. Guidance Counselors: Number of Students ÷ 250 = FTE.
13. Pupil Support Staff (Nurse, Health Assistant, Psychologist,
Parent/Community Liaison, Attendance/Dropout, Social Worker,
Speech/Occupational/Physical Therapist) Recommendation: Number of
Free/Reduced Price Lunch Participants ÷ 100 = FTE.
14. Non-Instructional Aide Recommendation: Number of Students ÷ 250 = FTE.
15. Librarian/Media Specialist Recommendation: Number of Students ÷ 450.
16. Assistant Principal Recommendation: (Number of Students ÷ 450) x .5*. (*.5
reflects the recommendation for a school with 450 students.)
17. School Secretary Recommendation: Number of Students ÷ 225 = FTE. This
calculation is based on the recommendation for two secretaries in a school
with 450 students).
18. Technology Recommendations: Number of Students ÷ 3 = Recommended
number of computers to have on campus. Number of Students ÷ 4 =
Recommended per-student distribution as well as computers for staff
members. Number of Students x $250 = Annual Maintenance, Replacement,
etc. Costs.
19. Instructional Materials Recommendation: Number of Students x $120.
20. Formative Assessments Recommendation: Number of Students x $25.
21. Library Materials Recommendation: Number of Students x $20.
22. Student Activities Recommendation: Number of Students x $200 to support
staffing and equipment/materials, transportation, etc.
Stage II: Cross-School Data Analysis
Use of a 78-page Excel spreadsheet reflecting the totality of quantitative data as
well as interview notes facilitated analysis across the eight schools included in this study.
Inclusion of calculations from the School – EBM Comparison Table for each school
309
added depth to the analysis. Statistical tools employed primarily included averages and
ratios. Because of the size difference across schools (652 students at the smallest school
compared with 1,738 students at the largest), calculation of staff- or dollar-to student
ratios facilitated the process of meaning in the data.
310
APPENDIX I
Botticelli Junior High School: A Case Study of Instructional
Improvement and Resource Use
Botticelli Junior High School is a grade seven and eight school with 927 students
in an urban district in Orange County, California. The district serves approximately
10,200 students in grades kindergarten through eight. Botticelli Junior High School is
one of three junior high schools in the district. Data released by the California
Department of Education in September 2008 indicate that Botticelli (along with 75% of
California Title I middle schools) did not attain the minimum percentages of students
with scores at or above “proficient” in language arts (35.2%) and mathematics (37%) to
meet federal Adequate Yearly Progress goals. This Title I-assisted school has been
identified as a Year Four Program Improvement school for the 2008-2009 school year.
A number of state and federal classifications assist in describing the population
served at Botticelli Junior High School. Although proportions attached to various
classifications are reported individually in California, most Botticelli students fit multiple
descriptors. For example, many English learners are also Latino as well as free and
reduced price lunch program participants.
Roughly 86% of the students at Botticelli Junior High School are classified as
“minority.” Comprising 76% of Botticelli’s students, the predominant ethnic student
subgroup considered numerically significant for state and federal reporting purposes
consists of Hispanic/Latino students. Student race and ethnicity proportions at Botticelli
diverge from those of the district overall. Notably, Botticelli’s Hispanic/Latino
311
population is 30% higher than that of the district while at 14%, its White population is
13% lower than seen across the district.
Approximately 74% of Botticelli Junior High School students are eligible for free
and reduced price lunch, as compared with 41% of students district-wide. These numbers
mirror those for Hispanic/Latino students at the school and in the district. Roughly 14%
of Botticelli’s students have been identified to receive special education services as
contrasted with 11% across the district. While 21% of the district’s overall population is
Asian, this group represents just 4% of students at Botticelli. Students identified as
English learners comprise 39% of Botticelli Junior High School’s student population, a
rate that is 7% higher than the district’s overall English learner proportions. Table 1
summarizes comparisons between the school and district.
Table 1. Botticelli Junior High School and Guggenheim School District Comparison of
Student Demographic Groups
Student Group
% at Botticelli % in District
African American 2.5 1.8
Asian 4 21
Economically Disadvantaged 74 41
English Learner 39 32
Hispanic/Latino 76 46
Special Education 13.5 10.5
White 14 27
312
Botticelli’s Academic Performance Index (API) has increased 100 points from
591 in 1999 to 691 in 2008 with its greatest growth (29 points) between 2005 and 2006.
In 2005 California added English Learner and Special Education student categories to its
disaggregated data system. These and any other student classifications containing at least
100 representatives at a school Botticelli’s size are considered significant for federal No
Child Left Behind reporting purposes. Inclusion of these categories reveals a
performance gap between English learners and special education students when
compared with their schoolmates. Similarly, a performance gap exists for every student
group when compared with their peers throughout the district. Figure 1 provides an
illustration of student achievement between 2005 and 2008 for significant student groups
both at Botticelli and across the Guggenheim School District. The difference is
somewhat muted by the inclusion of Botticelli’s API scores in the district numbers.
Figure 1. 2008 Federally significant student group API comparison.
Since the inception of the Academic Performance Index in 1999, Botticelli has
been identified with the lowest academic scores among the three middle schools in the
district.
313
Figure 2 illustrates the achievement differential across student groups in the areas
of language arts, Algebra, general mathematics, social studies, and science at the eighth
grade. This is the first year that students are tested in the four content areas of English-
language arts (ELA), mathematics, history-social science (H-SS), and science. Persistent
gaps between White and all other student groups exist in all areas outside mathematics. It
is significant to note that at the eighth grade students take either a general mathematics,
Algebra, or geometry exam to measure mathematics proficiency. The mathematics
course in which a student is enrolled determines the examination he or she takes. The
California State Board of Education decision to test all eighth graders in Algebra by 2010
as well as competitive college admissions expectations highlights the significance of this
as a gatekeeper course. Annually increasing NCLB proficiency expectations in
combination with the new state mandate magnify the importance of effectively teaching
Algebra to all students.
314
Figure 2. Botticelli Junior High School 2008 California Standards Test proficient and
advanced scores for significant student subgroups as defined by No Child Left Behind.
California’s measurement of academic achievement, the Academic Performance
Index (API) was instituted in 1999. The state provides schools with growth goals based
on the difference between the most recent API score and a goal score of 1000. Schools
compare schools from one year to the next to measure instructional improvement as
reflected in student test results. In 2003, the exams all became California academic
standards based. The language arts and mathematics portions of the battery of eighth-
grade California Standards Tests determines NCLB Adequate Yearly Progress goal
attainment. At present, the API calculation at the eighth grade includes scores from
history and science exams as well. The API and percentages of proficient and advanced
*Note: Percentages noted in the legend reflect the participation rate of each significant group that
took the Algebra test in 2008 to measure mathematics proficiency. Percentages are rounded to the
nearest whole number. For example, 2% of the 8
th
-grade students in the special education group
took the test. Hispanic/Latino and economically disadvantaged comprised the two largest
demographic groups among the 8
th
-grade class of 2008.
315
scores for each component exam and significant student group provide a glimpse into
student achievement at Botticelli J.H.S.
Student achievement as measured by the API at Botticelli Junior High School has
remained relatively flat during the six years between 2003 and 2008. The principal has
compared Botticelli’s student achievement with demographically similar schools in the
area and has noted that with its 100 Academic Performance Index (API) point growth
between 1999 and 2008, the school far underscores those in its similar-schools
comparison group. The elementary schools that feed into Botticelli all have attained
APIs over 700 while Botticelli remains in the 600s. At Botticelli, fluctuations in
proficiency levels in Algebra reflect those in other mathematics measures (e.g., general
mathematics) at the school. In spite of Algebra being offered to students who meet
course entry criteria (e.g., successful completion of pre-Algebra), student achievement
has seesawed from 29% proficient and advanced one year to 70% another. Figure 3
depicts Botticelli’s overall student achievement and growth in the core content areas
tested at the eighth grade between 2003 and 2008: language arts, mathematics (with a
focus on Algebra), history-social science, and science (added in 2006).
316
Note: Percentages noted above Algebra proficiency scores reflect the percentage of eighth-grade students
who took the Algebra test to demonstrate mathematics proficiency. Other eighth-grade students took a
general mathematics test measuring sixth- and seventh-grade standards.
Figure 3. Botticelli Junior High School Eighth Grade California Standards Test
proficient and advanced scores in core content areas from 2003 to 2008.
Since attaining Program Improvement status in 2005, Botticelli Junior High
School has concentrated school improvement efforts on language arts instruction,
instructional strategies to maximize student learning, specific mathematics
understandings and skills (e.g., multiplication), specialized interventions to improve
reading and mathematics skills, increased time for struggling learners in language arts
and mathematics, development of pacing guides in each content area, teaching at grade
level, positive school environment, professional collaboration and use of data, distributed
leadership, student use of technology, and moving first to well managed then to highly
effective classrooms. The purpose of this case study is to tell the story of efforts to
improve student learning at Botticelli Junior High School and to identify the resources
allocated toward those efforts.
34% 27.6%
24.7%
20.9%
30.6%
22.2%
317
Efforts to Improve Instruction at Botticelli Junior High School
The Evidence-Based Model informs the structured interview protocol that guided
data collection that subsequently framed this case study. Program Improvement middle
school principals responded to a series of quantitative and qualitative questions regarding
their 2008-2009 school improvement plans and strategies. Principals’ responses provided
quantitative resource allocation data for comparison with the Evidence-Based Model and
shared the strategies they employed toward attainment of their school improvement goals.
The “what, how, and why” part of the story follows.
Impetus for Change
The principal joined Botticelli Junior High School in fall 2006 as Botticelli was
entering year two of Program Improvement. Teacher morale, level of instruction, and
student behavior were sub-par. A group of school leaders participated in a two-year
School Leadership Team training and development series designed for Program
Improvement schools. Faculty complained about discipline problems with students.
Teachers instructed below grade-level expectations as determined by California standards
in each content area.
In its effort to attract students and to retain the 200-to-300 students electing to
attend other, higher-achieving schools in the district, Botticelli moved to a block schedule
and launched a school-wide one-to-one laptop computer program. Each student received
a laptop and every teacher was expected to utilize the possibilities of technology to
enhance learning their particular disciplines. Student achievement levels did not rise as
anticipated and student misconduct limited teachers’ efforts to teach.
318
While impressive on the surface, the one-to-one laptop program did not yield
anticipated student achievement results. The principal turned to research to determine a
path toward increased student learning.
Adopting New Curricula to Achieve Goals
In addition to using the state-approved instructional materials in math, science,
language arts, and history-social science, Botticelli invested in research-based
intervention programs. These include Language!, REWARDS, and Six-Minute Solution
(reading intervention programs), grades seven and eight Advancement Via Individual
Determination (AVID) curricula, and Accelerated Math (math intervention program).
Botticelli also has adopted school-wide use of Cornell Notetaking, a strategy commonly
used in AVID classes. In addition to academic curricula Botticelli invested in
Conversation, Help, Activity, Movement, and Participation (CHAMPS), a positive school
climate program developed by Safe and Civil Schools. Addressing academic skills
weaknesses, developing a positive and college-going attitude among students, and
teaching at grade level influenced curricular selections.
Implementing the Research-based Curricula.
Implementation of the augmented curricula occurs during and after the regular
school day. The school has identified four basic student schedule options based on
student academic readiness and need as measured by the California Standards Tests.
Students struggling with decoding, the most basic component of reading enroll in two
periods of language arts. One class period is dedicated to an intervention program using
the REWARDS program to teach students to decode text. When the reading challenge
319
centers on fluency (i.e., ability to read smoothly with appropriate pronunciation, tone, and
responses to punctuation), students use the Six Minute Solution during a second period of
language arts instruction. Students needing to learn basic math facts enroll in a basic-
level math class accompanied by a period of Accelerated Math, an intervention
emphasizing essential math skills and facts. At the next level of proficiency students take
a math class with a companion class to enhance and support learning. As student
achievement increases electives replace interventions. Figure 4 illustrates the four basic
schedules.
320
Far Below Basic & Below Basic Students
English/
Language
Arts
Reading
Intervention
with
REWARDS
(decoding)
Mathe-
matics
Mathe-
matics
Intervention
with
Accelerated
Math
P.E. History Science
Basic – to – Low Proficient Students
English/
Language
Arts
Reading
Intervention
with Six
Minute
Solution
(fluency)
Mathe-
matics
Mathe-
matics
Companion
Class
(reteach,
pre-teach,
tutoring)
P.E. History Science
Proficient in Mathematics and Language Arts
English/
Language
Arts
Reading
with Six
Minute
Solution
Mathe-
matics
Elective P.E. History Science
Proficient and Advanced in Mathematics and Language Arts
Advanced
English/
Language
Arts
Elective Advanced
Mathe-
matics
Elective P.E. History Science
Figure 4. Botticelli student schedule options.
After school academic support takes place three days each week in a mandatory
eighth-period class for students failing language arts and mathematics classes called the
“Academic Empowerment Club.” Every Monday, Tuesday, and Thursday nearly 200
students participate in an hour-long remediation program funded by California High
School Exit Exam (CAHSEE) preparation and remediation dollars.
321
Botticelli participates in the national AVID program and has two AVID class
sections, one for seventh and one for eighth grade. A key curricular component of the
AVID program is use of Cornell Notes. While AVID students use this dialectical
notetaking system that includes a summary of learning each day, at Botticelli all students
are taught and reinforced in the use of Cornell Notetaking in all content classes.
Other research-based instructional strategies implemented throughout Botticelli
Junior High School include identifying similarities and differences, recognizing effort,
setting objectives, and providing feedback. These strategies have been identified through
Marzano’s meta-analysis as being especially effective tools for student learning.
Teachers in all subject areas have received training in Step Up to Writing (SUTW) to
instruct students to write coherent paragraphs. Reciprocal Teaching in which students
negotiate collective understanding of text through structured small group discussions and
Schmoker’s reading comprehension protocol in which students read, write, and talk about
their reading add to the research-based instructional strategies directed toward reading
comprehension.
Every student receives a Botticelli planner, which includes calendar-style pages to
list daily assignments and homework, bell schedules, school policies, a description of
CHAMPS, and instructions in school wide instructional strategies: SUTW, Venn diagram,
and Reciprocal Teaching. Teachers participate in ongoing professional development to
support student learning goals through strategies such as those delineated in this section.
322
Support Teachers with Professional Development
Most professional development is tied to Botticelli’s goals for student learning.
In addition, the district supports teachers and administrators with both compulsory
professional development programs and incentives for continued learning. Specific
programs such as AVID provide annual and periodic conferences and seminars.
Much of Botticelli’s professional development takes place on campus through
ongoing collaboration within departments and through use of outside consultants working
with different departments. In addition to three contractual professional development
days each school year, Botticelli faculty participate in weekly data-driven, student-
learning collaborative meetings. A Curriculum Committee establishes quarterly goals to
support attainment of annual student learning and disposition goals. These quarterly and
annual goals inform possible strategies and instructional approaches, which further
inform knowledge and skills teachers need.
Teachers new to the profession as well as those facing instructional and classroom
management challenges receive support through compulsory district-organized programs.
The Guggenheim School District offers salary increases to administrators who complete
learning commitments related to district goals such as technology.
The district underwrites and provides personnel for Beginning Teacher Support
and Assessment (BTSA) and Peer Assistance and Review (PAR) – state-mandated
programs to support teachers new to the profession and experienced teachers facing
instructional and classroom management challenges. AVID teachers attend both state
and local trainings to develop and enhance their delivery of AVID strategies.
323
Struggles to develop consistency in mathematics achievement has led the school
to hire a consulting firm to work with the department on benchmark assessments and
standards-based instruction. This work began during the 2007-2008 school year and will
continue throughout 2008-2009.
The earlier-mentioned two-year School Leadership Team training provided the
principal, fellow administrators, departmental and other leaders opportunities to examine
school data and consider both protocols and goals to bring back to the campus staff.
During this time the school adopted CHAMPS and sent a team to observe the effects of
this positive school culture program in action at a school with demographics similar to
Botticelli’s. Upon adoption of this program the entire staff received training to
implement this new way of teaching students to behave appropriately at school. All
teachers received SUTW training and the language arts department worked with a
consultant to learn to use Reciprocal Teaching. All language arts and mathematics
teachers are trained in the use of their instructional materials. Additionally, all teachers
have been trained to use Data Director to analyze student achievement results on
benchmark and common assessments. During the 2008-2009 school year Action
Learning Systems, a consulting firm, will identify what is being implemented well at
Botticelli using the school as a norm.
The instructional day ends shortly after two o’clock each Wednesday and teachers
meet in collaborative groups to examine data, establish goals, and identify colleagues to
share how they are meeting student learning needs. The faculty has been trained to
follow a 30-minute data-driven meeting protocol, which includes a document identifying
324
the upcoming goal and strategies to achieve it. As briefly alluded to above, Botticelli
J.H.S. has modified its school day to support student learning and professional needs.
Restructured School Day
Botticelli employed a block schedule until the 2008-2009 school year. Recent
budget cuts led to a return to a traditional schedule. The principal reports this as a
positive change. The seven-period day affords students needing additional language arts
and mathematics support with more instructional time than they had with the 85-minute
block in each subject. Additionally, teachers see all students daily as opposed to every-
other-day with the block schedule. The school day extends until 4:10 PM with the eighth-
period Academic Empowerment Club serving 180 students three days each week. The
extended school day is one example of the school’s coordinated efforts to help struggling
students.
Coordinated Help for Struggling Students
Botticelli has developed a number of responses to address student academic and
behavioral concerns. California Standards Test scores, benchmark exam results,
individual education plans (IEPs), and grade point averages feature in class placement.
As mentioned above, Botticelli has adopted a school wide positive educational climate
program (CHAMPS) in which all students are taught appropriate campus behaviors
ranging from voice volume to requesting teacher assistance. A teacher on special
assignment conducts academic counseling meetings with the school’s lowest-achieving
students and their parents.
325
To support students at different levels of mathematics proficiency Botticelli
provides a range of offerings. Beyond Algebra and pre-Algebra, students may be placed
in a number of classes at both lower and higher levels. For example, eighth grade
students whose previous year’s math scores are less than proficient are enrolled in
Algebra Readiness. These eighth-grade students are taught seventh-grade math
standards. These students also enroll in Algebra Concepts Skills, a companion class in
which key concepts are pre- and re-taught. Skill development is supported through use of
Accelerated Math and Accelerated Reader available during the regular school day. In
2009-2010 all eighth graders will take Algebra. The principal expects to implement
additional supports for students struggling with this course.
The Academic Empowerment Club operates three days each week for one hour
after school ends. Students identified as “Far Below Basic,” “Below Basic,” and with a
grade point average below 2.0 are required to enroll in this intervention course. The
curriculum includes REWARDS, a language arts intervention emphasizing decoding
skills. REWARDS Plus soon will be added to support reading comprehension and content
learning in social studies and science.
In addition to support for struggling students, Botticelli Junior High School
utilizes its extensive technology resources to provide 70 students with take-home laptop
computers. One cohort each of seventh- and eighth-grade students take content and
elective classes together with a team of teachers representing all content areas and
interested in utilizing computer technology to maximize student learning and creativity.
This modified school-within-a-school serves as a pilot for effective use of technology.
326
Botticelli’s response to student academic and social needs is captured in Figure 5.
Each “tier” includes parent information, outreach, and involvement. More detail on
efforts to communicate with, educate, and include parents follows in the next section.
Figure 5. Response to student academic and social needs at Botticelli JHS.
Ongoing Parent Communication and Education
Parent outreach at Botticelli includes parent education nights in English and
Spanish, weekly phone calls, and active parent governance and support organizations on
campus. Parent conferences run from 12:30 through six o’clock during conference week
to maximize parent opportunities to meet with teachers. A teacher on special assignment
holds a quasi-administrative role and meets with the school’s most struggling students
and their parents to explain the significance of improving academically and point out how
parents can partner with the school to support their children’s success.
Step Up to Writing
Cornell Notes
Reciprocal Teaching
Benchmark Assessments
Anytime Anywhere (1:1
laptop program)
“Marzano” Strategies
(e.g., Similarities and
Differences)
CHAMPS
Accelerated Reader
After School Athletics
Student Planner
Student Success
Assemblies
Tier I
Academic
Empowerment Club
Attendance Outreach
Parent Education
Courses
Support/Companion/
Inclusion Classes
“Healthy Start”
(resources for students
in need)
Tier II
Student and Parent
Counseling
Intervention Classes
Tier III
327
In addition to a frequently updated Web site, Botticelli also uses weekly phone
calls through its “Connect Ed” phone system to notify parents of upcoming events.
Families receive a school newsletter periodically as well.
The school has contracted with the Latino Educational Advancement Foundation
to provide a six-week parent education series entitled “Ten Education Commandments
for Parents.” The course is offered in both Spanish and English. Parents of students at
feeder elementary schools are invited to attend as well. Course content includes
information on college-prep classes and homework expectations. Parents who complete
the series receive a certificate of completion at a culminating ceremony.
An active PTA, English Learner Advisory Committee, and School Site Council
provide support to school initiatives including the “Pathways to Success” assemblies in
which students are recognized for academic achievement and improvement. Parent and
community support are part of the school’s vision for student academic and social
success. This vision guides leadership efforts at Botticelli.
Instructional Leadership as a Foundation for Change
Leadership and guidance for instructional improvement occurs primarily and
predominantly at the school site level with support from the district. The principal
emphasizes research-based strategies, a focus on grade-level standards, ongoing
monitoring of student achievement with data, and institutionalization of essential
practices school wide. By developing a strong curricular leadership team and addressing
longstanding concerns regarding student misconduct, the principal at Botticelli Junior
328
High feels that 2008-2009 will be the year that the school moves from now being well
managed to becoming highly effective.
The principal describes Botticelli as a “beat up car covered with bondo racing
(against) Formula racers.” A great deal has been done to make the school look nice on
the outside but it continues to lose 200 -300 high performing students each year to other
schools in the district. The school wide one-to-one laptop program initiated at Botticelli
two years ago in order to make the school attractive has been retracted to a 70-student
pilot program. The principal indicated, “Technology in the hands of kids adds
complexity and must be highly intentional.” The principal anticipates that scaling back to
develop capacity will result in much stronger and meaningful implementation for more
students in the future.
The laptop story illustrates the principal’s “business model” for school
improvement and leadership. Overt measures and external validation determine success
of programs and strategies. Neither improved test scores nor retention of high
performing students resulted from the laptop program, so the program is being retooled
and reconsidered.
Through regular data analysis school leaders and teachers have recognized that
they have been teaching below grade level. The staff has committed to school wide
instructional strategies, development and use of benchmark assessments, and continual
goal setting for effective instruction, student learning, and a college-going culture at
Botticelli. Collective goal setting and instructional commitments demonstrate the
school’s move toward a professional, collaborative culture.
329
Creation and Maintenance of a Professional, Collaborative School Culture
A group of lead teachers and counselors alongside administrators attended a two-
year School Leadership Team training program to firmly develop a collaborative school
culture grounded in student achievement goals. Elements of that culture include
quarterly benchmark assessments in all core content areas followed by departmental
discussions and plans to modify instruction. The school’s professional and collaborative
culture is supported by a weekly early release schedule in which students complete the
school day 55 minutes early every Wednesday, permitting teachers to conduct data-
driven departmental and grade-level meetings. The leadership team has taught all faculty
members to use a 30-minute meeting protocol in which collaborative groups focus on
assessment, curriculum, or instruction as it pertains to a current student-learning goal.
Meeting participants rotate facilitation and other roles, and create a document in which
they identify their action plan to address their objective. Leadership guiding professional
collaboration and other school improvement initiatives has emanated from the principal’s
ongoing review of current school improvement literature. The district supports
Botticelli’s school-driven efforts as described in the following section.
District Leadership and Support
Instructional improvement efforts at Botticelli JHS are school-centered with
district support. The superintendent and the principal have discussed the tension between
trying to make the school look good through such efforts as the one-to-one laptop
program and making the school truly effective. The principal continually promotes the
school and efforts being undertaken to improve it. In return the superintendent has
330
supported the school’s plans, including the plan to retreat from the school wide one-to-
one laptop program in order to improve it to the point that it can be connected to student
achievement gains. The principal has emphasized the critical nature of capacity
development and the superintendent supports this focus, encouraging the school to “go as
fast as it can,” but not comparing Botticelli to other schools in the district.
The district provides support through stipends for teacher leaders who assume
such additional responsibilities as Beginning Teacher Support and Assessment coaches
and Peer Assistance and Review mentors. Other district-level coordinators provide on-
call support ranging from technology to gifted and talented education. In addition to
district support providers, Botticelli avails itself to a variety of sources to bring
professional knowledge into the school.
Bringing Professional Knowledge Into the School
Botticelli’s curriculum leadership team connects multiple academic and student
discipline measures with external mandates to develop school goals and identify areas of
professional knowledge that might be of greatest benefit to staff and students.
The school utilizes its “inside experts” as well as outside consultants and professional
development experiences to address professional development needs. The principal
references Accountability in Action by Doug Reeves in describing professional
development activities connected to the learning results noted in “90-90-90” schools (i.e.,
schools with 90% minority students, 90% identified as economically disadvantaged, and
90% scoring proficient or advanced on state assessments). These activities include
modeling of effective teaching and assessment strategies, collaboration, and closely
331
following student progress. According to the principal, Reeves’ research informs the
school’s approach to professional development.
Current professional development opportunities include a school wide review
with Action Learning Systems in which the school intends to identify its most effective
practices and institutionalize them. Additionally, the mathematics department continues
to work with consultants from Choice of Action to hone instructional skills and
collaboration skills. As new instructional programs are purchased (e.g., Language!)
appropriate teachers receive implementation training as required for Program
Improvement schools.
All teachers participate in evaluating student-writing benchmarks. Additionally,
departments collaboratively develop formative assessments to add to Data Director. The
result is site-based professional learning grounded in common expectations of student
achievement. While teachers may request to attend specific workshops and seminars,
much of the current professional development efforts are centered on long-term
departmental explorations of instructional improvement through professional learning
communities.
This chapter of the Botticelli school improvement story centered on the school’s
current activities. The following chapter identifies research-based school-improvement
strategies that provide points of comparison for Botticelli’s efforts.
Features of Improving Schools
An extensive study of school resource use to improve student learning in
Washington resulted in a set of eight key features associated with improving schools
332
(Fermanich, Mangan, Picus, Odden, Gross, & Rudo, 2006). These features include (a) a
focus on educating all students, (b) data-driven decision making, (c) use of a rigorous
curriculum that aligns with state standards, (d) effective professional development to
support instructional improvement, (e) a restructured learning environment, and (6)
extended learning opportunities for struggling students. These six success features are
facilitated through (f) instructional leadership and (g) professional learning communities.
These eight findings provide a framework for summarizing the story of this school’s
improvement efforts.
Focus on Educating All Students
The principal noted the increased efficacy of teachers following successful
implementation of CHAMPS. Management of student behavior has freed teachers to
consider the achievement of all students. According to the principal, effects of this
newfound efficacy include academic interventions and support classes, current efforts to
retool special education programming, and use of student achievement data and school
wide assessments to create a sense that every teacher is responsible for every student.
Examination of the program for resource (mild-to-moderate special education
classification) students has led the principal to establish a new focus with the department.
Goals include (a) development of a progression by which students may exit special
education, and (b) a focus on academic needs for course placement as opposed to special-
needs identification. Instead of placing resource students in special education language
arts and mathematics classes, Botticelli is experimenting with a blended services model
333
entailing placement of resource students in mainstream classes. Here, a resource teacher
with content expertise teams with the content teacher to monitor and support all students.
Botticelli strives to meet No Child Left Behind performance targets each year. In
2007-2008 the school did not meet federal proficiency levels in language arts and
mathematics with key student groups: English learner, economically disadvantaged,
Hispanic/Latino, and special education. For 2008-2009 Botticelli has announced its goals
that 50% of all students will score proficient or advanced on both district benchmark
exams and California Standards Tests. These goals require achievement gains upward of
45% proficient or advanced in some student groups and content areas. Use of data is
critical in making each of the school’s instructional decisions.
Data-Driven Decision Making
Teachers regularly use data to reflect on student achievement and instruction.
Data Director, with which all teachers are familiar, is available for ongoing formative
assessment and data review as well. Quarterly benchmark data provide a springboard for
reflection and consideration of new instructional approaches and student interventions.
The history-social science department has led the school in developing formative
assessments and programs such as Accelerated Reader provide data that inform both
teachers and students. Data use supports the school’s efforts to provide a standards-
aligned curricular program to all students.
Use of a Rigorous Curriculum that Aligns with State Standards
Botticelli’s efforts in the area of rigorous standards-aligned curriculum include
use of state-approved instructional materials, institutionalization of effective instructional
334
practices school wide, and reworking its technology-based learning program. In addition
to California State Board of Education-approved and district-adopted instruction
materials for all core content areas, Botticelli’s principal and curriculum leaders have
researched and selected a number of state-approved intervention programs to support
reading development and basic mathematics concepts. Teachers in all subject areas
provide instruction in effective writing to support language arts and discipline-specific
studies simultaneously. As indicated previously, Botticelli teachers and administrators
investigated and are now implementing several new programs to meet performance goals.
These research-based programs include Read 180, Algebra Readiness, and Reading
Counts. They also employ Cornell Notetaking, an evidence-based strategy associated
with increased student achievement.
And, in recognizing that “technology in the hands of kids adds complexity and
must be highly intentional,” the principal and staff members have scaled back the
school’s flagship laptop program to ratchet up its instructional rigor and potential for
student learning. Using a business model, the principal uses data to monitor the
effectiveness of programs as well as implementation of them. Similarly, the principal
emphasizes professional development as a critical component of providing a rigorous
curricular program.
Effective Professional Development to Support Instructional Improvement
Professional development at Botticelli addresses school goals for academic
achievement and institutionalization of commonly implemented strategies toward that
end. To improve student achievement in mathematics a consulting firm provides ongoing
335
support of the math department while Action Learning Systems is involved with
identifying and institutionalizing practices that seem to make a difference at the school.
Training has supported institutionalization of Step Up to Writing for all staff members
and staff experts have taught colleagues Cornell Notetaking as a school wide strategy to
support effective notetaking and summarizing.
The curriculum leadership team and academic departments identify areas of need
for professional learning, which are then incorporated into the school plan.
Reorganization of the school schedule created increased learning opportunities for both
staff members and students.
A Restructured Learning Environment
Both students and teachers experience a restructured environment to enhance
student learning and professional collaboration. The return to a traditional school day has
resulted in more time for academic interventions and support. Additionally, a mandatory
eighth period class for struggling students further increases instructional time as needed.
By increasing the school day by a few minutes four days each week, Botticelli’s teachers
gain an hour of collaborative meeting time, made efficient and productive through use of
a timed collaborative meeting protocol with an action plan as a product at the end of each
meeting. Students in need receive support beyond the regular school day through
extended learning opportunities.
Extended Learning Opportunities for Struggling Students
Students with failing scores on district benchmark exams enroll in the Academic
Empowerment Club, an eighth period class that meets three times weekly. Mandatory
336
participation continues until grades and scores meet levels deemed proficient.
Approximately 500 students participate in summer school for remediation in language
arts and mathematics. Robust support and improvement mechanisms such as the
extended school day rely on instructional leadership. Revisited below in short.
Instructional Leadership
Words that summarize instructional leadership at Botticelli include responsive,
data- and research- driven, distributed, and goal-oriented. Botticelli’s principal
responded to staff concerns regarding student behavior and worked with the school’s
leadership team to explore behavior programs with positive results with similar student
populations. In guiding the efforts of the school’s leadership team, the principal turned to
school improvement literature. A school wide focus on non-fiction reading and writing
strategies emanates from the principal’s reviews of research by Reeves, Fisher and Frey,
Schmoker, and Marzano. The school leadership team utilizes the research-based
strategies introduced during the two-year leadership team development program in which
it participated. In summary, key leadership elements have included distribution across
staff members, a research base, and working toward a vision for student achievement by
prioritizing actions.
Leadership actions include quarterly recognition of student achievement through
special student assemblies for students with high and improving grades. Every morning
at Botticelli includes an announcement from the principal to remind students of the steps
they are expected follow on the pathway to success. Leadership additionally includes an
337
expectation of and consistent support for professional learning through collaborative
experiences.
Professional Learning Communities
As described previously, early-release Wednesdays provide protected time for
collaborative meetings focused on department-determined goals aligned with school
achievement targets. Meetings emphasize DuFour’s questions: What do we want
students to learn? How will we know they have learned it? What will we do when
students struggle to learn? What will we do to enhance learning when students succeed?
As departments analyze student data, collaboratively develop assessments and
instructional plans, members share ideas, successes, and challenges with each other.
They identify areas in which they need assistance in improving their content knowledge
and craft expertise. Different departments have embraced professional learning
communities more completely as a way of conducting the business of education than
have others. The history department and leadership team serve as models of professional
learning communities for the rest of the school.
Botticelli’s story reflects evidence of each of the eight elements associated with
improving schools. These evidence-based elements have been justified further by nearly
identical findings related to schools that have doubled student performance (Odden,
2007). While this chapter of Botticelli’s story has centered on a narrative of mostly
unquantified school improvement strategies in place at Botticelli, the next chapter
continues the story with a quantitative component. Data collected from Botticelli J.H.S.
338
provide a point of comparison between current school resource allocation levels and
those suggested by the Evidence-Based Model.
Essential Features and Doubling School Performance
The essential features associated with improving schools reflect the ten steps to
double school performance as determined by Odden (2007). Schools experiencing
dramatic improvement (a) conduct a needs assessment, (b) set higher goals, (c) adopt a
new curricular program, (d) implement data-based decision making, (e) invest in long-
term, extensive professional development, (f) use school time more efficiently, (g)
employ multiple extra-help strategies for struggling students, (h) create professional
learning communities dedicated to boosting student performance, (i) support instructional
improvement at all levels of leadership, and (j) utilize external professional expertise.
The side-by-side comparison in Table 1 illustrates the similarities between observed
elements of school improvement in Washington and those identified in studies of school
improvement across many states. These features appear in differing degrees across
schools included in this series of case studies.
Table 1. “Washington Learns” and Ten Steps to Double Student Performance – Critical
Elements of School Improvement
Washington Learns Critical Elements of School
Improvement
Ten Steps to Double Student Performance
Data-Driven Decision Making Needs Assessment
Focus on Educating All Students Higher Goals
Rigorous Curriculum Aligned to State Standards New Curricular Program
Data-Driven Decision Making Data-Driven Decision Making
Professional Development Long-Term, Extensive Professional Development
339
Table 1 continued
Washington Learns Critical Elements of School
Improvement
Ten Steps to Double Student Performance
Extended Learning Opportunities for Struggling
Students
Multiple Extra-Help strategies for Struggling
Students
Restructured Learning Environment Effective Use of School Time
Professional Learning Communities Professional Learning Communities
Instructional Leadership Support for Instructional Improvement at All Levels
of Leadership
Professional Development Utilization of External Professional Expertise
In addition to these tightly aligned critical school improvement features, the
Evidence-Based Model proposed by Odden and Picus (2008) identifies quantifiable
research-based elements associated with improving and successful schools. Odden and
Picus have translated extensive research findings into formulas for establishing
appropriate levels of human and material resources to support adequate educational
outcomes for all students. Data collected from Botticelli Junior High School provide a
point of comparison between current school resource levels and those suggested by the
Evidence-Based Model.
The Evidence-Based Model for Educational Adequacy and Resource Allocation
Levels at Botticelli Junior High School
The Evidence-Based Model proposed by Odden and Picus (2008) identifies
quantifiable research-based elements associated with improving and successful schools.
Odden and Picus have translated extensive research findings into formulas for
establishing appropriate levels of human and material resources to support adequate
educational outcomes for all students. The Evidence-Based Model provides a yardstick
340
against which current resource allocations at Botticelli Junior High School may be
compared. Extensive school improvement research informs the calculations and
suggestions contained within the model. Botticelli’s resource allocation diverges from
most elements contained within the Evidence-Based Model.
Key points of contrast between resource allocations levels at Botticelli J.H.S. and
that recommended by the Evidence-Based Model include school and class size, number
of certificated staff members, course assignments, coaches for teachers, and professional
development time.
Research findings indicate a connection between school size and student
achievement. At 927 students, Botticelli is roughly two times larger than the size
recommended by research and suggested by the Evidence-Based Model. With an
average class size of 38 students, classes are approximately 1.5 times larger than those
associated with student performance gains. In addition to adding approximately 10
teachers to reduce class size, the Evidence-Based Model suggests reducing the number of
elective courses from 40% of course offerings to 20%. If resourced in accordance with
the model, 25 certificated teachers would provide wide-ranging support in the form of
coaching, tutoring, providing extended learning opportunities beyond the regular school
day and year, and supporting mild-to-moderate special education and English learner
students. For a school of Botticelli’s size, model recommendations further include more
guidance counselors and staff members, librarians, clerical staff, and professional
development time.
341
The area in which resource allocation at Botticelli exceeds model
recommendations include the number of electives teachers, instructional aides, and a
quasi-administrative position (i.e., Title I Coordinator) that does not exist in the model. If
resourced based on the model, Botticelli’s staff would grow by nearly 40 members.
Botticelli’s principal indicates that the school has enough financial resources to
continue and expand the school’s improvement efforts. A focused, singular effort to
make the school appear safe, comfortable, and organized has been and continues to be a
key element connected to the school’s effort to regain the confidence of neighborhood
families that have elected to place their children in other district junior high schools.
Table 2 provides a comparison of the actual resource use in Botticelli Junior High
School to that suggested by the Evidence-Based Model. The total amount of state and
categorical funds allocated to the school for the 2008-2009 school year is $360,000.
Table 2. Resource Comparison: Botticelli Junior High School and Evidence-Based
Model (EBM) for Adequate School Resources
School Element EBM Prototypical
Middle Schools
Botticelli JHS –
Current Resource
Status
Botticelli JHS – EBM
Comparison & EBM
Resource Suggestions
School
Characteristics
Configuration Grades 6-8 Grades 7 & 8 2 grade levels v. 3
School Size 450 927 2.1 x larger than EBM
Core Class Size 25 44.4 1.5x higher than EBM
Free & Reduced Price
Lunch Count
(used below as a
calculator for school
resourcing)
690
Number of ELL
Students
(used below as a
calculator for school
resourcing)
362
342
Table 2 continued
School Element EBM Prototypical
Middle Schools
Botticelli JHS –
Current Resource
Status
Botticelli JHS – EBM
Comparison & EBM
Resource Suggestions
Personnel Resources
(measured in FTEs)
1. Core Teachers 18 20.9 37.1
2. Specialist/
Elective Teachers
20% of Core Teachers 40% of Core Teachers
(13.3)
7.4
Total Teachers 21.6 34.2 44.5
3. Instructional
Coaches
1 per 200 Students 0 4.6
4. Tutors 1 per 100 At Risk
Students (Free &
Reduced Price Lunch)
0 6.9
5. Teachers for ELL
Students
1 per 100 ELL
Students
1.3 3.6
6. Extended Day 1 per 15 Eligible
Students (Staffing
based on 50% of Free
& Reduced Price
Lunch Students)
180 Students
13 Teachers
3 Days/Week
1 Hour/Session
≈1 FTE Teacher Paid
at Hourly Rate
5.8
7. Summer School
1 per 15 Students
(Staffing based on
50% of Free and
Reduced Price Lunch
Students)
500 Students
20 Teachers
4 Weeks
4 Hours Daily
Average Class Size:
25
≈2 FTE Teachers Paid
at Hourly Rate
5.8
Personnel Resources
(measured in FTEs)
8. Alternative
Schools
NA
NA
NA
9a. Learning- and
Mildly Disabled
Students
3 Teachers per 450
Students
3 Resource Teachers
& 2 FTE Aides for 84
Students
6.2 Teachers
343
Table 2 continued
School Element EBM Prototypical
Middle Schools
Botticelli JHS –
Current Resource
Status
Botticelli JHS – EBM
Comparison & EBM
Resource Suggestions
9b. Severely Disabled
Students
100% State
Reimbursement for
Top 1% Minus
Federal Funds
3 Self-Contained
Special Education
Teachers (Special
Day and Severely
Handicapped) – One
Teacher Teaches an
Extra Section and 4.5
FTE Aides for 41
Students
Based on Need
10. Services for
Gifted Students
$25 per Student 0 $23,175
11. Career/ Technical
Education
.3 Weight per CTE
Student to Maintain
Low Class Sizes
$7000 per CTE
Teacher Contract
NA NA
12. Substitutes 10 Days per Teacher
@ $100 per Day
0
Daily Rate = $105
10 Days per Teacher @ $100
per Day
13. Pupil Support
Staff
1 Guidance Counselor
per 250 Students plus
1 Pupil Support Staff
Position per 100 Free
& Reduced Price
Lunch Students
2 Guidance
Counselors,
1 Teacher on Special
Assignment to
Counsel “Far Below
Basic” Students and
their Parents, &
1 “Healthy Start”
Coordinator
3.7 Guidance Counselors plus
6.9 Pupil Support Staff
Members
14. Non-Instructional
Aides
2 per 400-500
Students
0 4
15 Instructional Aides 0 7 (4.5 FTE) Special
Education Self-
Contained Aides, 4
(2.5 FTE) Resource
Aides
0 Instructional Aides
16. Librarians/Media
Specialists
1 0.5 Library/Media
Assistant
2.1 Librarians
344
Table 2 continued
School Element EBM Prototypical
Middle Schools
Botticelli JHS –
Current Resource
Status
Botticelli JHS – EBM
Comparison & EBM
Resource Suggestions
17. Administrator 1 Principal plus .5
Assistant Principal
1 Principal,
1 Assistant Principal,
.5 Title I Coordinator
1 Principal plus 1.6 Assistant
Principals
18. Secretary 2 School Site
Secretaries
1 Secretary
(11 months) &
2.5 Clerks
(10 months)
4.1 Secretaries
Non-Personnel
Resources
19. Professional
Development
10 Days Intensive
Professional
Development in
Summer Included in
Teacher Contract
Instructional Coaches
(included in Personnel
Resources above)
Planning & Prep time
$100 per Student
3 Professional
Development Days: 1
Prior to Beginning of
the School Year, 1 to
Review Data in
September, & 1 in
October
Conferences: AVID
& Others as
Determined by Need
and Teacher Request
Weekly Collaborative
Meetings
(1 hour/week ≈ 5
days’ time)
Contracted
Consultants as per
School and
Department Goals
BTSA & PAR
$14,000
7 Additional Professional
Development Days Added to
Current 3 – Group All Days
During Summer
Instructional Coaches to
Provide Ongoing Modeling,
Coaching, & Instructional
Support
Additional Trainings &
Conferences as Needed
$92.700
345
Table 2 continued
School Element EBM Prototypical
Middle Schools
Botticelli JHS –
Current Resource
Status
Botticelli JHS – EBM
Comparison & EBM
Resource Suggestions
Dollar/Pupil
Resources
20. Technology Calculation: 1
Computer per 3
Students.
Distribution: 1
Computer per 4
Students & 1:1 Ratio
for Administrators,
Teachers, &
Secretaries
$250/Pupil
$231,750 school-wide
$7.50 Per Student:
Data Director
.5 FTE Technology
Assistant
$250K Annually to
Update & Repair
Laptops & Related
Hardware/ Software -
$170K from District;
$80K from School
Total Technology:
≈ $289/Pupil
With its 1:1 student laptop ratio,
Botticelli is a unique school and
exceeds the EBM suggestion
for both number of and total
expenditures. However, the
school’s expenditures for what
it maintains are nearly 67%
lower than what the model
suggests for the number of
units.
21. Instructional
Materials
Formative
Assessments
Library Materials
$120/Pupil
$25/Pupil
$20/Pupil
$10,000 Reading
Intervention Programs
(Language! &
REWARDS)
District-Developed
Benchmark
Assessments & Site-
Developed Formative
Assessments
Comparison: Instructional
Materials Funding Realignment
Program (IMFRP), Categorical
Funds, & Library Funds at
District and Site Levels provide
resources comparable to EBM
suggestions.
22. Student Activities $200/Pupil ASB – 1 Class
Section for Select
Students
Yearbook – 1 Class
Section for Select
Students
After School Athletics
Pathways Assemblies
(Quarterly
Recognition for
Students and
Teachers)
Comparison: EBM suggests
equitable distribution across all
students. Botticelli appears to
direct greatest allocation of
resources toward students with
the highest proficiency levels –
making them eligible for special
class offerings.
346
Future Considerations
A number of implications emerge through the data analysis process.
Considerations for the future include the significance of behavior management, teacher
efficacy, collaboration, use of research-based strategies, use of technology, course
offerings, alignment of professional development offerings, and staffing decisions on
student achievement.
Botticelli Junior High School struggles with a reputation for being the “bad”
school in the district. The school’s highest-performing students have been “filleted” off,
as have its best teachers, to the district’s high performing schools. Underlying the
principal’s leadership is a goal of returning high-performing students to the school.
Before implementation of CHAMPS, teachers felt incapable of teaching their
students. Efficacy levels have improved demonstrably, evidenced by creativity and
perseverance in planning, instructing, and assessing for student achievement. A great
deal of research exists in the area of efficacy. Ongoing examination of efficacy, factors
influencing it, manifestations of it, and possible connections between it and gains in
student achievement may provide worthwhile information for the future.
Now that behavior issues appear to be under control at Botticelli Junior High
School, the 2008-2009 school year may prove to be an ideal laboratory to explore the
influence of ongoing teacher collaboration, school wide use of research-based
instructional strategies, extended school days for struggling students, and use of data to
inform instruction. Although the data may be biased due to self-selection, the 70
“Anytime Anywhere” one-to-one laptop program students and their teachers also provide
347
potentially valuable end-of-year data points: Will these students score significantly higher
than their peers on the California Standards Tests (CST)? How do their instructional
experiences differ from those of their non-laptop program peers?
Outside special education nearly 30% of courses offered are non-core classes.
Physical education is included as a core course in this calculation as it is mandated per
state legislation. If physical education is included as an elective, the percentage of non-
core courses increases to 40%. Notable is the lack of such academic courses as foreign
language. These data leave the observer with questions regarding the effectiveness of
and possibilities for current strategies to support CST proficiency levels through use of
school-wide strategies in elective classes.
The principal noted the complexity that a one-to-one laptop program adds to
instruction. Botticelli’s initial implementation did not include a technology coordinator.
Reinstating the program school wide would benefit from a full-time technology
“integrater,” someone with the capacity to manage and troubleshoot complex technology,
expertise in its instructional possibilities, and ability to provide professional development
and coaching to teachers for effective implementation. In the meantime, a cadre of
teachers committed to incorporating relevant 21
st
century learning opportunities
collectively is attempting to develop its own capacity in these technology integrater
functions. How the school addresses these issues of providing the human resources for
effective technology implementation following its 2008-2009 Anytime Anywhere pilot
may serve an illustrative purpose for other schools attempting similar programs.
348
The Guggenheim School District provides district-level services for beginning
and struggling teachers as well as in the use of educational technology. Teachers on
special assignment hold district-office positions to support teachers of English learner,
special education, and gifted and talented students. Affected teachers at Botticelli attend
various district-sponsored workshops. As much as the district supports Botticelli’s
school improvement efforts, district offerings and the school’s professional development
agenda do not appear to be related. It may be of interest to see if such alignment exists in
other districts. And, if it does exist, does it connect with improved student learning
results?
As with a number of other case study schools, Botticelli retains a Title I
coordinator and a part-time technology support staff member. If this trend reflects
staffing allocations in improving schools, these positions warrant further study as to their
unique contribution to student learning. With its research-based improvement efforts
underway, Botticelli holds potential for subsequent analysis by future researchers.
349
APPENDIX J
Chagall Junior High School: A Case Study of Instructional
Improvement and Resource Use
Chagall Junior High School is a grade seven and eight school with 1,360 students
in an urban district in Orange County, California. The district serves approximately
26,000 students in grades seven through twelve. Chagall Junior High School is one of
eight middle schools in the district. Data released by the California Department of
Education in September 2008 indicate that Chagall (along with 75% of California Title I
middle schools) did not attain the minimum percentages of students with scores at or
above “proficient” in language arts (35.2%) and mathematics (37%) to meet federal
Adequate Yearly Progress goals. As a result, this Title I-assisted school has been
identified as a Year Three Program Improvement school for the 2008-2009 school year.
A number of state and federal classifications assist in sketching an image of the
population served at Chagall Junior High School. Many students fit multiple descriptors.
For example, many English learners are also Latino as well as free and reduced price
lunch participants.
Roughly 89% of the students at Chagall Junior High School are classified as
“minority.” Comprising 71% of Chagall’s students, the predominant student subgroup
considered numerically significant for state and federal reporting purposes consists of
Hispanic/Latino students. Student race and ethnicity proportions at Chagall diverge
350
somewhat from the district overall. Chagall’s Hispanic/Latino population is 10% higher
than that of the district while its White population is 7% lower than seen across the
district.
Approximately 67% of Chagall Junior High School students are eligible for free
and reduced price lunch, as compared with 56% of students district-wide. In line with the
district, nine percent of Chagall’s students receive special education services. The school
serves as a central service point for visually and orthopedically impaired students.
Students identified as English learners comprise 62% of Chagall Junior High School’s
student population, a rate that is 15% higher than the district’s overall English learner
proportions.
Chagall’s Academic Performance Index has risen 144 points, from 571 in 1999 to
715 in 2008. Its greatest API increase occurred between 2007 and 2008 when this score
increased by 36 points. Significant gaps persist across the school’s student subgroups:
White, economically disadvantaged, special education, English learner, and
Hispanic/Latino. Yet, Hispanic/Latino students have experienced greater API increases
during the past 10 years than their White counterparts. Between 1999 and 2008, White
students experienced 119-point API growth while Hispanic/Latino scores increased by
216 points. Economically disadvantaged student scores increased by 147 points between
2001 and 2008. Between 2006 and 2008, English Learner API scores have increased by
22 points and special education student scores have grown by 19 points. Because the
state added various student groups to its API reporting system in different years, finding
an average rate of increase helps in painting a picture of student academic achievement
351
gains. Figure 1 highlights average API increases for each significant student group at
Chagall Junior High School. As illustrated, English learner and special education
students have experienced the slowest average growth in API scores. These student
groups also are identified with the lowest API results at the school.
Figure 1: Average (Mean) annual API growth across significant student groups.
Eighth-grade Algebra instruction and assessment point both to the achievement
gap and efforts to bridge it at Chagall. Changes in the size and demographics of the
school mask the story told by numbers reflecting the rate at which students take Algebra.
Percentages describe these rates more effectively. During the past six years the
percentage of Hispanic/Latino eighth-graders tested in Algebra has increased from 1.8 in
2003 to 21.1 in 2008. At the same time, the number of economically disadvantaged
students enrolled in Algebra has increased from four percent of the school population to
352
21.5. English learner participation has increased from one percent (7 students) to 4.9%
(36 students). The single special education student that took Algebra at Chagall JHS was
the first to take the Algebra exam since 2003. Figure 2 elaborates upon this achievement
gap narrative by comparing Algebra participation in 2008 for significant student groups
with the representation of that student demographic at the school.
The California Department of Education’s Standardized Testing and Reporting
(STAR) Web site includes nuanced scores such as those for economically disadvantaged
Hispanic/Latino and non-economically disadvantaged Hispanic/Latino students. While
both of these Hispanic/Latino student populations are aggregated into one in the graph
below, it is of interest to note that seven percent of the non-economically disadvantaged
Hispanic/Latino students enrolled in Algebra. At 14.2%, just over twice as many
economically disadvantaged Hispanic/Latino students participated in Algebra.
Furthermore, 53% of this group’s students attained proficient or advanced Algebra results
compared with 37% of its non-economically disadvantaged Hispanic/Latino peers.
353
Figure 2. Algebra participation and student group representation among eighth grade
students at Chagall JHS.
Figure 3 provides a final illustration of the achievement differential across student
groups at Chagall Junior High School. Students are tested in the areas of language arts,
Algebra, general mathematics, social studies, and science at the eighth grade, providing
an opportunity to explore achievement in all core content areas. It is significant to note
that at the eighth grade students currently do not have to take Algebra in order to attain
mathematics proficiency. However, the California State Board of Education decision to
test all eighth graders in Algebra by 2010, as well as competitive college admissions
expectations highlights the significance of this as a gatekeeper course. White students
take the Algebra exam at a higher rate than their Hispanic/Latino, economically
disadvantaged, English learner, and special education peers.
354
*Note: Percentages listed next to student group titles in the legend represent the
percentage of students from each group that took the Algebra standards test in 2008.
Figure 3. Chagall Junior High School 2008 California Standards Test proficient and
advanced scores for significant student groups as defined by No Child Left Behind.
As mentioned above, Chagall JHS has experienced an increase in its annual API
results. Incongruously, standards test results indicate a dramatic drop in language arts
scores between 2006 and 2007 as well as a decline between 2005 and 2006 in Algebra
scores. However, between 2003 and 2008, the school increased the number of students
taking the Algebra exam from 43 to 274. Between 2005 and 2006, the number of
Algebra students more than doubled, increasing from 61 to 147. At the same time, the
percentage of proficient and advanced students declined from 92 to 61%. By 2008, the
proficiency percentage had nudged above its all time low in 2007 of 56% while providing
Algebra instruction to the largest percentage of students during the period measured.
355
The school continues to experience a discrepancy across student groups receiving
instruction in this area. White student scores exceed those of all other student groups in
all content areas. Science scores have increased rapidly since inclusion of this exam in
2006. Figure 4 depicts Chagall’s overall student achievement and growth in the core
content areas tested at the eighth grade between 2003 and 2008: language arts, Algebra,
general mathematics, social studies, and science.
Figure 4. Chagall Junior High School California Standards Test proficient and advanced
scores in core content areas from 2003 to 2008.
Since attaining Program Improvement status in 2006, Chagall Junior High School
has concentrated school improvement efforts on student engagement, mentoring, use of
benchmark assessments and resulting data, and effective academic and social
interventions. The purpose of this case study is to tell the story of efforts to improve
356
student learning at Chagall Junior High School and to identify the resources allocated
toward those efforts.
Instructional Improvement Strategies
The Evidence-Based Model informs the structured interview protocol that guided
data collection and frames this case study. Program Improvement middle school
principals responded to a series of quantitative and qualitative questions regarding their
2008-2009 school improvement plans and strategies. Principals’ responses provided
quantitative resource allocation data for comparison with the Evidence-Based Model and
told the story of the school’s use of resources toward instructional and academic
improvement. The “what, how, and why” part of the story follows.
Impetus for Change
The principal described Chagall as a hidden treasure. The school has a “bad”
reputation associated with low academic achievement and student violence. Its Program
Improvement status provides parents with an option to send their children to other
schools in the district. Some leave. However, according to the principal, those that opt to
remain at Chagall “love” the school and the academic, social, and enrichment
opportunities available to its students. Attaining Program Improvement status resulted in
the school seeking means to identify struggling students and finding ways to support their
achievement. Purposeful selection of instructional materials to meet student-learning
needs represents an initial step toward supporting student achievement.
357
Adopting New Curricula to Achieve Goals
In addition to using the district-adopted instructional materials in math, science,
language arts, and history-social science, Chagall invested in several research-based
intervention programs. These include Read 180 (reading intervention program), Step Up
to Writing (developmental writing program), and Measure Up (mathematics). The school
additionally has adopted use of Cornell Notetaking and Socratic seminars (widely used
Advancement via Individual Determination [AVID] strategies). Curricular augmentation
has been an important piece of Chagall’s school improvement story.
Implementing the Research-based Curricula
Some elements of the curricular augmentation occur in all classes while others are
part of special intervention classes. For example, Read 180 is offered as an intervention
class for students reading more than two years below grade level. A language arts
teacher teaches Read 180 sections to non-special education, English-speaking students.
A resource specialist teaches Read 180 sections to resource (mild-to-moderate special
education) students. And a special day class (SDC) teacher instructs self-contained
special education students with the Read 180 curriculum as well.
The Proficient Paragraph program is a research-based, structured instructional
program to teach students to construct paragraphs with topic and concluding sentences,
connected body sentences, and appropriate transitions. Teachers in language arts, social
studies, and science provide structured instruction in and require that students write
paragraphs meeting specific criteria. Five benchmark assessments throughout the school
year measure student improvement and proficiency with paragraph writing.
358
Chagall participates in the national AVID program and has AVID classes at both
the seventh and eighth grades. A key curricular component of the AVID program is use
of Cornell Notes. While AVID students use this dialectical notetaking system that
includes a summary of learning each day, at Chagall all students are taught and
reinforced in the use of Cornell Notetaking in all content classes. Teachers participate in
both formal and informal professional development to enhance their knowledge and skills
in instruction and leadership to support both student learning and teacher professional
growth.
Support Teachers with Professional Development
Professional development opportunities consist primarily of departmental
collaboration time, case-by-case attendance at conferences and trainings, and
participation in School Leadership Training (SLT) with the county office of education.
At the conclusion of each quarter teachers meet by subject area to examine
student benchmark data. Preparation for the meetings entails use of Data Director, a sub-
scription-based online data organization system commonly used in schools. Teachers
complete a meeting preparation form in which they identify areas of success and of
challenge. The meetings are intended to serve as opportunities to reflect, support
colleagues, and consider ideas to reteach essential concepts.
Teachers also may request collaboration days. Members of departments working
together on data examination and instructional plans reserve substitute teachers with the
principal’s permission on an ad hoc basis. This year, science teachers have utilized two
collaboration days. No other teachers have requested to do so as of October 2008.
359
No Child Left Behind Program Improvement school requirements include
mandatory use of 10% of the school’s Title I funds for professional development.
Chagall uses some of these funds to send a group of teacher leaders and administrators to
School Leadership Team training. This two-year program emphasizes distributed
leadership based on a common vision for student achievement, effective use of data, and
goal setting for student achievement. In addition to substitutes for teachers, the program
fee is $10,000 each year.
Use of a rigorous standards-based curriculum, formative assessments and teacher
collaboration, academic interventions, and parent/community involvement are the staff-
determined cornerstones of Chagall Junior High School’s instructional improvement
plan. Teachers request approval to attend seminars, workshops, conferences, and
trainings on an individual basis. Requests must relate to the school’s improvement
cornerstones and/or to school-designated target student populations: students with low
CST scores, English Learners, and Special Education. Upon return, participating
teachers are required to share what they learned with their departments. In addition to
professional growth activities, Chagall Junior High School recognizes the potential for
the school schedule to promote student achievement and a positive school climate.
Restructured School Day
Chagall employed a block schedule until the 2008-2009 school year. Recent
budget cuts resulted in a return to a traditional schedule. The principal commented on a
surge in fights among seventh-grade students and speculated that there may be a
connection between the increased passing periods and the fights. Unsupervised time in
360
the halls and the anonymity associated with rotating among seven classes daily may be
connected to the after school fights, the most recent leaving a student unconscious.
A group of eighth-grade teachers representing U.S. history, physical science, jazz,
and advanced band has created an interdisciplinary team to connect the disciplines. This
team shares a common group of students similar to a school-within-a-school. At the
conclusion of the first semester the team will host a star-watching night and a concert of
music related to the sky and stars. In the spring, the music production will relate to U.S.
history topics being studied (e.g., Civil War). In addition to efforts to structure school
time to maximize student-learning experiences, teachers at Chagall intervene to support
struggling and special-needs students.
Coordinated Help for Struggling Students
Chagall employs various intervention methods to support struggling students.
Examples include a “Reteach” period for resource students, a quarter-length Response to
Intervention (RTI) class for struggling students, a three-week RTI Learning Skills class
for students falling behind in Algebra, pre-Algebra, or language arts, and an extra class
period daily of language arts support for English learners in transition from English
Language Development classes to mainstream English/language arts classes.
Reteach consists of a daily class period of mathematics and/or language arts
support provided by resource teachers for students on their caseloads. These teachers
have two conference periods daily with the expectation that they use some of this time to
meet with their students’ math and language arts teachers, stay abreast of topics and
361
assignment expectations, and provide relevant reteaching and support during designated
class periods with their resource students.
A special education teacher provides instruction to “regular” education students in
a quarter-length language arts course. Teachers whose efforts do not lead to adequate
levels of student achievement recommend students for the RTI class. The class follows a
RTI curriculum in which the teacher received training. The purpose is twofold: provide
remediation in language arts and utilize the special education teacher’s diagnostic skills
and experience to consider a recommendation for special education testing.
Chagall’s principal and teacher leaders investigated and implemented “mandatorials,” a
practice employed by other middle schools in the area. Algebra teachers may refer
struggling students to three-week mandatorial sessions following poor test performance.
Students are released from their elective classes and attend a focused, small-group class
to catch up on the material they appear to not understand. Mandatorials also are available
for language arts, pre-Algebra, and Algebra readiness students after school. These
sessions run 10 to 12 hours spread over a three-week period.
Upon completion of a mandatorial session, math students retake the test on which
they previously attained a poor score and keep the new score. In language arts, students
produce a new writing sample and take a skills-based test to demonstrate mastery.
Throughout the mandatorial series, students continue to take the math or language arts
class to which they are assigned. All mathematics and language arts teachers agreed to
permit students to be reassessed and to replace the previous grade with the new one.
362
To support English learners when they are redesignated as English-language
proficient, Chagall provides an additional class period of language arts support.
Additionally, English learners may attend a Beginning English Language Development
class after school to learn about American culture.
Ongoing Parent Communication and Education
Chagall JHS hosts a comprehensive web site that is updated frequently.
Additionally, the school has an active PTSA and parents are interested in volunteering on
campus. Currently, the district is unable to support the school’s efforts to test parents for
tuberculosis so they may be cleared to volunteer in classrooms. The principal is
searching for other ways to meet legal compliance issues.
Parents will be invited to attend a star gazing night toward the end of the first
semester and a special concert in the spring. Additionally, the school’s multiple
performing arts electives include performances to which family members are invited.
This year Chagall will host a fitness night in which parents participate in the state fitness
test activities required of their children.
Chagall hosts four English Learner Advisory Committee (ELAC) meetings
annually, providing transportation, childcare and food. Other parent meetings include
school site council, parent information nights, and counseling meetings with parents of
students with grade point averages below 2.0, California Standards Test scores in the
Basic and lower proficiency ranges, and behavioral issues (as needed).
Parent involvement also includes the Parent Institute for Quality Education
(PIQE, pronounced “P.K.”). Using Title I and EL funds the school pays counselors to
363
conduct workshops for parents. Parents learn the same information counselors share with
students in planning for high school and beyond. The principal anticipates scaling back
from 10 sessions during the 2007-2008 school year to three or four during 2008-2009.
Parent and community involvement as well as all other school improvement efforts rely
on site-level leadership commitments from the principal.
Independent Reading
Chagall Junior High School implements a reading for pleasure program. All
students participate by reading “chapter” books and students reading the most receive
recognition and small prizes. This program reflects the school’s mission to motivate
students to become independent learners.
Instructional Leadership as a Foundation for Change
Leadership and guidance for instructional improvement occurs primarily and
predominantly at the school site level. Staff members determined the school’s four
cornerstones for improvement efforts based on standards-test, California Healthy Kids
Survey, discipline, and anecdotal data. In addition to instruction-focused improvement
efforts, Chagall Junior High School also focuses on mentoring and engaging students.
Leadership includes attempts to develop a professional, collaborative culture school-
wide.
Creation and Maintenance of a Professional, Collaborative School Culture
Efforts to facilitate a professional, collaborative school culture at Chagall include
master schedule modifications, data-driven conversations, and departmental
collaboration. Last year, Chagall began a multiyear program of modifying the master
364
schedule to include common conference periods for teachers of the same subjects. After
successfully providing English Language Development teachers with common
conference periods, the school had to revert from a block to a traditional (seven period)
schedule. Language arts teachers anticipated common conference periods this year but it
has not happened. The status of this modification remains in limbo due to the challenges
of master schedule changes within a seven-period school day.
Expectations for collaboration include quarterly benchmark data dialogues, ad hoc
collaborative days as requested by departments, and after-the-event reports of
professional development experiences to departments by their members. Some
components of school collaboration (i.e., data-driven dialogues) benefit from district-
level support.
District Leadership and Support
District leadership and support exists in the form of periodic evaluations of the
principal and technology to manage student data. The principal receives an evaluation
every other year. The district requires information regarding the school’s Program
Improvement plan annually. Additionally, the district underwrites technology used at
each school site: Data Director (assessment data organization service), Zangle (student
information database system), Nambience (software used by counselors), and Integrade
(combines attendance, behavior/discipline referrals, and grades to identify at-risk
students). In general, school improvement efforts are site-developed and site-driven.
These efforts include bringing outside knowledge into the school and are tempered by the
district’s philosophy of using inside expertise.
365
Bringing Professional Knowledge Into the School
Professional knowledge primarily enters Chagall through teachers who share with
colleagues about off-site experiences. Following participation at trainings, seminars, or
workshops Chagall teachers provide a report of what they learned to colleagues. The
district was the recipient of a federal Teaching American History Grant as well as an
educational technology grant. Several Chagall teachers participated in each. These
provided content knowledge and pedagogy to U.S. history teachers as well as a focus on
use of technology both to teach students and for students to demonstrate their knowledge.
The district traditionally discourages use of outside professionals, preferring to tap into
internal knowledge bases.
Chagall Junior High School’s support for struggling students, curricular choices,
and use of data reflect the school’s ongoing efforts to improve instruction and student
achievement. To varying extents, the school’s efforts also reflect research-based findings
associating staff member actions with increases in student learning. The next section
contains a review of case study findings framed by what current research indicates as
features connected to school improvement.
Features of Improving Schools
An extensive study of school resource use to improve student learning in
Washington resulted in a set of eight key findings. Features associated with improving
schools include (a) a focus on educating all students, (b) data-driven decision making, (c)
use of a rigorous curriculum that aligns with state standards, (d) effective professional
development to support instructional improvement, (e) a restructured learning
366
environment, and (f) extended learning opportunities for struggling students. These six
success features are facilitated through (g) instructional leadership and (h) professional
learning communities. These eight findings provide a framework for summarizing the
story of this school’s improvement efforts.
Focus on Educating All Students
Chagall strives to meet No Child Left Behind performance targets each year.
Hispanic/Latino, special education, economically disadvantaged, and English learners
continue to lag behind White peers who comprise 11% of the school’s student population.
Schoolwide, “Mandatorials,” mentoring, after school learning opportunities, and electives
reflect the school’s commitment to social, developmental, and academic achievement.
Chagall’s science teachers in particular reflect a core belief that all students can learn.
Forty-nine percent of eighth grade students at Chagall in 2008 earned California
Standards Test scores in science at the proficient and advanced levels. Science
department members emphasize the importance of agreement on essential knowledge and
skills, of frequent and formative assessment, and of preparing all students for high school
success. They, along with the interdisciplinary school-within-a-school, serve as a model
for the rest of Chagall. These departments demonstrate the goal orientation, use of data,
and instructional innovation the principal hopes to see in every department and
classroom.
Data-Driven Decision Making
Teachers use data to reflect on student achievement and instruction. Data
Director, with which all teachers are familiar and in which the district makes an annual
367
investment of $7.50 per student, is available for ongoing formative assessment and data
review as well. Quarterly benchmark data provide a springboard for reflection and
consideration of new instructional approaches. It is unclear to what extent teachers
modify instruction during the course of the school year as a result of data or analysis of
such. However, assessment data are used to determine student placement in courses (e.g.,
honors, “sheltered” – for English learners) and interventions (e.g., RTI, Mandatorials).
Use of a Rigorous Standards-Aligned Curriculum
Standards-aligned instructional materials in all core content areas that have met
California adoption criteria support Chagall’s standards-aligned curriculum. As indicated
previously, Chagall teachers and administrators investigated and are now implementing
several new programs to meet performance goals. These research-based programs
include the Proficient Paragraph, Read 180, and Step Up to Writing. They also employ
Cornell Notetaking, Power Words, and Socratic Seminars, evidence-based strategies
associated with increased student achievement. The school selected from the list of state-
approved, standards-aligned texts in the four state-recognized content areas of language
arts, mathematics, science, and history-social science. The science department stands out
in terms of its teachers collectively determining student learning objectives that guide
instruction and use of materials.
Effective Professional Development to Support Instructional Improvement
Professional development opportunities at Chagall include three contract days
each year for site-based, quarterly, data-based collaboration sessions. Additional
professional development takes the form of teacher-to-teacher sharing, limited grant-
368
funded professional development participation, and specific program-based trainings.
The school’s AVID teachers participate in an annual institute with day-long follow-up
sessions during the school year. One special education teacher underwent training to
implement the school’s RTI curriculum. Quarterly data-based collaboration sessions
provide a common experience for all Chagall teachers. These sessions and the two-year
School Leadership Team training encompass the school’s comprehensive professional
development experiences.
Restructured Learning Environment
Recent changes in the school’s master schedule have reduced collaboration and
instructional time. However, the school continues to provide additional support through
intervention classes during the school day, ultimately adding a class period for students
requiring such support in language arts and mathematics and providing additional
relatively short-term booster shots of support in mathematics (either one quarter or three
weeks). The school day is extended for English learners interested in learning about
American culture.
Extra Learning Opportunities for Struggling Students
Use of student assessment data and teacher referrals determine student placement
in extra help classes both during and after the regular school day. Student course
placement includes language proficiency status as well as course preparedness
information. For example, seventh grade students typically take some form of pre-
Algebra: regular pre-Algebra, sheltered pre-Algebra (for English learners), or honors pre-
Algebra. In addition, students requiring additional pre-Algebra support take Math
369
Interventions class, resulting in 90 minutes of pre-Algebra instruction daily. Counselors
assist with course placement.
Acknowledge Successes at the School Level
Chagall initiated a pleasure reading program during the 2007-2008 school year.
Students were and continue to be encouraged to read for fun. The principal provides
rewards for students who read the most chapter books. Rewards include school pencils
and other small trinkets. High volume readers are recognized in school announcements.
A visit to the school’s Web site includes an announcement of the school’s current API.
Instructional Leadership and Professional Learning Communities
As part of the school’s Program Improvement corrective action plan, a team of
Chagall teachers and administrators participates in School Leadership Team training.
School teams learn the skills necessary for distributed leadership. Additionally, Chagall
staff members collectively use multiple data sources to determine focal points for
improvement efforts. Distributed leadership links with professional learning
communities in providing a foundation for school improvement.
Frequent meetings of staff members committed to considering key questions and
data related to student academic achievement are not evident at Chagall. However,
nascent forms of professional collaboration that also reflect shared decision-making exist
at this school. The principal and staff members used multiple data sources to identify
target student groups for extra support. These groups include students with low
California Standards Test scores, English learners, and special education students.
370
Accountability in terms of producing quarterly data analysis documents compels
teachers and departments to use Data Director and to develop improvement goals in
terms of student achievement. Data also have informed selection of instructional
strategies implemented across core content areas school-wide (e.g., Cornell Notes &
Proficient Paragraph).
The school’s leadership efforts are tied to its attempts to develop a culture of
professional collaboration. The science department and the newly formed history-
science-music collaborative serve as distinct cases of professional collaborative
communities at Chagall, both of which are supported by the principal. Additionally, the
principal supports common preparation periods for members of each department to
facilitate professional collaboration during the regular school day.
Essential Features and Doubling Student Performance
The essential features associated with improving schools reflect the ten steps to
double school performance as determined by Odden (2007). Schools experiencing
dramatic improvement (a) conduct a needs assessment, (b) set higher goals, (c) adopt a
new curricular program, (d) implement data-based decision making, (e) invest in long-
term, extensive professional development, (f) use school time more efficiently, (g)
employ multiple extra-help strategies for struggling students, (h) create professional
learning communities dedicated to boosting student performance, (i) support instructional
improvement at all levels of leadership, and (j) utilize external professional expertise.
The side-by-side comparison in Table 1 illustrates the similarities between observed
elements of school improvement in Washington and those identified in studies of school
371
improvement across many states. These features appear in differing degrees across
schools included in this series of case studies.
Table 1. “Washington Learns” and Ten Steps to Double Student Performance – Critical
Elements of School Improvement
Washington Learns Critical Elements of School
Improvement
Ten Steps to Double Student Performance
Data-Driven Decision Making Needs Assessment
Focus on Educating All Students Higher Goals
Rigorous Curriculum Aligned to State Standards New Curricular Program
Data-Driven Decision Making Data-Driven Decision Making
Professional Development Long-Term, Extensive Professional Development
Extended Learning Opportunities for Struggling
Students
Multiple Extra-Help strategies for Struggling
Students
Restructured Learning Environment Effective Use of School Time
Professional Learning Communities Professional Learning Communities
Instructional Leadership Support for Instructional Improvement at All Levels
of Leadership
Professional Development Utilization of External Professional Expertise
In addition to these tightly aligned critical school improvement features, the
Evidence-Based Model proposed by Odden and Picus (2008) identifies quantifiable
research-based elements associated with improving and successful schools. Odden and
Picus have translated extensive research findings into formulas for establishing
appropriate levels of human and material resources to support adequate educational
outcomes for all students. Data collected from Chagall Junior High School provide a
point of comparison between current school resource levels and those suggested by the
Evidence-Based Model.
372
The Evidence-Based Model for Educational Adequacy and
Resource Levels at Chagall Junior High School
The Evidence-Based Model provides a yardstick against which current resource
allocations at Chagall Junior High School may be compared. Extensive school
improvement research informs the calculations and suggestions contained within the
model. Chagall’s resource allocation diverges from most elements contained within the
Evidence-Based Model.
In addition to a qualitative narrative description the school’s improvement efforts
by comparing with the key findings from the review of improving and successful school
districts in Washington (Fermanich, Mangan, Picus, Odden, Gross, & Rudo, 2006), this
study also includes a quantitative comparison of case-study schools with the Evidence-
Based Model for adequately resourcing schools. Striking differences between current
resource allocations and those suggested by the model include the number of core content
and electives teachers, teacher tutors, and professional development.
If resourced in accordance with the model, Chagall JHS would employ roughly 34
additional full time equivalent teachers for core instruction and extra support.
Counseling and pupil support staff along with non-instructional aides and librarians
would increase while the number of elective/specialist teachers and instructional aides
would diminish. In addition to adding 7 days of focused professional development to the
school’s existing three contracted days, the Evidence-Based Model suggests adding
nearly seven full time equivalent instructional facilitator/coach positions to support
ongoing instructional improvement.
373
School improvement research findings point to a connection between school size
and student achievement. Smaller schools are linked to higher achievement results.
Chagall JHS is double the size suggested by the Evidence-Based Model. Its class size is
1.7 times the size suggested by school improvement findings. The summary above
highlights notable findings. Table 2 comprehensively details the quantitative findings of
this case study.
Table 2. Resource Comparison Chagall Junior High School and Evidence-Based Model
for Adequate School Resources
School Element Evidence-Based
Model Prototypical
Middle Schools
Chagall JHS
Current Resource
Status
Chagall – EBM Comparison
& EBM Resource Suggestions
School
Characteristics
Configuration Grades 6-8 Grades 7 & 8
School Size 450 1359 3x larger than EBM
recommendation
Core Class Size 25 44.8 1.8x larger than EBM
recommendation
Free & Reduced Price
Lunch Count
(used below as a
calculator for school
resourcing)
911
Number of ELL
Students
(used below as a
calculator for school
resourcing)
843
Personnel Resources
1. Core Teachers 18 30.3 54.4
2. Specialist/
Elective Teachers
20% of Core Teachers 57% of Core Teachers
(17.4 – Includes 2
FTE Spanish
Teachers)
10.9
Total Teachers 21.6 47.4 65.3
3. Instructional
Coaches
1 per 200 Students 0 6.8
4. Tutors 1 per 100 At Risk
Students (Free &
Reduced Price Lunch)
0 9.1
5. Teachers for ELL
Students
1 per 100 ELL
Students
0 8.4
374
Table 2 continued
School Element Evidence-Based
Model Prototypical
Middle Schools
Chagall JHS
Current Resource
Status
Chagall – EBM Comparison
& EBM Resource Suggestions
6. Extended Day 1 per 15 Eligible
Students (Staffing
based on 50% of Free
& Reduced Price
Lunch Students)
1.3
(Extended Day and
Homework Club)
7.6
7. Summer School 1 per 15 Students
(Staffing based on
50% of Free and
Reduced Price Lunch
Students)
0
District cancelled
summer school for
junior high school
students due to budget
cuts.
7.6
8. Alternative
Schools
NA NA NA
9a. Learning- and
Mildly Disabled
Students
3 Teachers per 450
Students
3 9.1
9b. Severely Disabled
Students
100% State
Reimbursement for
Top 1% Minus
Federal Funds
5
10. Services for
Gifted Students
$25 per Student $0
(Funds not released
for 08-09. In 07-08
received $7840)
$33,975
11. Career/ Technical
Education
.3 Weight per CTE
Student to Maintain
Low Class Sizes
$7000 per Course for
Materials
2 FTEs
310 Students
2.6 FTEs
$7,000 for Materials
12. Substitutes 10 Days per Teacher
@ $100 per Day
Daily Rate = $105 per
Day
10 Days per Teacher @ $100
per Day
13. Pupil Support
Staff
1 Guidance Counselor
per 250 Students plus
1 Pupil Support Staff
Position per 100 Free
& Reduced Price
Lunch Students
3 Counselors, 2
Full-Time Attendance
Clerks,
1 Ed. Psychologist,
.6 Speech Therapist,
1 Sp. Ed. Program
Specialist
5.4 Guidance Counselors plus
9.1 Pupil Support Staff
Positions
14. Non-Instructional
Aides
2 per 400-500
Students
0 6
15 Instructional Aides 0 9 0
16. Librarians/Media
Specialists
1 0.3 Librarian plus
1 Media Assistant
3
375
Table 2 continued
School Element Evidence-Based
Model Prototypical
Middle Schools
Chagall JHS
Current Resource
Status
Chagall – EBM Comparison
& EBM Resource Suggestions
17. Administrator 1 Principal plus 0.5
Assistant Principal
1 Principal plus
1 Assistant Principal,
plus
.2 Title I Coordinator
1 Principal plus
2.5 Assistant Principals
18. Secretary 2 School Site
Secretaries
1 Registrar
1 Admin. Asst.
1 Gen. Office Clerk
0.5 Counseling Clerk
6
Non-Personnel
Resources
19. Professional
Development
10 Days Intensive
Professional
Development in
Summer Included in
Teacher Contract
Instructional Coaches
(included in personnel
resources)
Planning & Prep time
$100 per Student to
Fund
4 Days Professional
Development in
Teacher Contract
Used for Quarterly
Data Dialogues - Use
3 of the 4 Days –
Opted to Use the 4
th
Day to Teach
AVID Conference (2
teachers)
One Prep Period Per
Teacher Daily.
SpEd Teachers have 2
Prep Periods to
Provide Support to
Caseload Students in
Content Classes &
Meet with Content
Teachers
5 Days per Participant
(≈ 8 Participants) for
School Leadership
Team Training
Departmental
Collaboration Time
(as approved by
principal)
$5050 Budgeted
10 Days Intensive Professional
Development in Summer
Instructional Coaches as
Delineated Above
Additional Conferences and
Time as Needed
$135,900 to Fund
376
Table 2 continued
School Element Evidence-Based
Model Prototypical
Middle Schools
Chagall JHS
Current Resource
Status
Chagall – EBM Comparison
& EBM Resource Suggestions
Dollar/Pupil
Resources
20. Technology Calculation: 1
Computer per 3
Students.
Distribution: 1
Computer per 4
Students & 1:1 Ratio
for Administrators,
Teachers, &
Secretaries
$250/Pupil
District provides
Software: Data
Director (student test
data management),
Nambience
(counseling data),
Zangle (student
information database),
& Integrade (at-risk
criteria management)
Internet and Cable
Access School-wide
453 Computers
$339,750 Annually for
Maintenance and Replacement
21. Instructional
Materials
Formative
Assessments
Library Materials
$120/Pupil
$25/Pupil
$20/Pupil
State-approved
Materials in Each
Core Content Area
and for Academic
Intervention
4 Student-Free Days
to Evaluate
Assessment Data and
As-Requested Time to
Develop New Tests –
Utilized Most by
Algebra Teachers
$1200 School Site
Council Funds,
Some EL and Title I
Funds
For Library Materials
Required instructional materials
purchases align with EBM
instructional materials
recommendations.
$163,080 Recommended
Budget for Instructional
Materials
Each department develops
formative assessments.
$33,975 Recommended Budget
for Formative Assessments
$27,180 Recommended Budget
for Library Materials
377
Table 2 continued
School Element Evidence-Based
Model Prototypical
Middle Schools
Chagall JHS
Current Resource
Status
Chagall – EBM Comparison
& EBM Resource Suggestions
22. Student Activities $200/Pupil Computer Lab (1 hour
daily)
Homework Club (1
hour daily)
Piano and Guitar
Instruction
Spelling Bee
History Day
Clubs
Athletics
Community Learning
Center
Non-profit After-
school Program
ASB
Music and Drama
Performances
Transportation
≈ $180,000
$271,800
Current expenditures to staff
student activities include hourly
pay and stipends for clubs and
activities offered outside the
contract day. ASB is offered as
a class.
Total expenditures are
approximately $50,000.
Future Considerations
Challenges for Chagall Junior High School center on staffing, extended use of the
school facility that derail the leadership focus, and meeting the needs of students whose
assessment results have not attained proficiency targets. These principal-identified
challenges, along with other areas that emerged in the interview process, hint at areas of
future interest along Chagall’s school improvement journey.
In spite of state and district budget woes, the principal at Chagall feels
comfortable with the amount of funding available through Title I, II, and III sources.
However, those funds cannot be translated into the type of staff members that could free
378
up administrative staff members from the “massive” amounts of clerical work to which
they must personally attend. Freeing up administrators through additional clerical
support would, in the principal’s opinion, provide increased opportunities to be in
classrooms as “true instructional leaders.”
The district’s afternoon and evening adult school program held at Chagall creates
ongoing challenges as adult school teachers and students interact with Chagall staff
members and students. The principal spends an estimated five hours weekly dealing with
adult school problems. Removing the adult school from Chagall would give the principal
significant time that could be spent in classrooms.
Chagall experienced a 36-point increase in its school-wide Academic
Performance Index in 2008, aligning the school with the district average. However, even
in science, Chagall’s strongest curricular area as per CST scores, far fewer English
learners and special education students achieved proficiency than other students on
campus. Although the school has identified low performing students and subgroups
achieving at lower-than-acceptable levels to target instructional improvement efforts,
Hispanic/Latino, English learner, economically disadvantaged, and special education
students continue to trail their White peers. A dramatic example is reflected in student
achievement in social studies. While 49% of White eighth graders were determined to be
proficient and advanced on the standards test in social studies, 27% overall attained
similar status. Three percent of special education students were declared proficient or
better as were 5% of English learners. The targeted efforts as currently implemented do
379
not appear to have the desired effect and it is unclear to what extent instruction is
changing as a result.
District-level budget challenges have directly affected Chagall’s capacity to
support professional collaboration and an extended school year for students in need of
credit recovery and remediation. These cuts follow a year in which the school has found
pride in its achievement gains. Chagall may present an ideal laboratory for exploration of
student achievement following the removal of these evidence-based student achievement
components.
The personal-interest approach to professional development may benefit from
some form of analysis to gain a sense of its effectiveness. Additional areas of potential
ongoing analysis lie in the current practice of segregating different types of learners as
well as the effectiveness of the various intervention classes.
As the School Leadership Team continues its training, it has the potential to bring
“SMART” (i.e., Strategic, Measurable, Attainable, Research-based, Timebound) goal
planning to Chagall, perhaps making the large task of whole-school improvement more
manageable. The school is committed to the social and emotional success of students.
Additional resources and goal setting in this area may align with the school’s mission.
Actual implementation of agreed-upon instructional strategies school-wide (e.g., Cornell
Notetaking) as well as actually modifying instruction based on formative assessment data
are subject to individual teacher discretion. This may be an area for additional
consideration.
380
Two areas to watch closely during the 2008-2009 school year are science and the
history-music-science collaborative. Not only do scores in science reflect the greatest
percentage of student proficiency on campus, but science teachers also utilize
collaboration days at a greater rate than colleagues in other departments. Collaborative
efforts necessary to link multiple disciplines may require more joint planning than in
other departments. Additionally, the possibility of identifying meaningful connections
across disciplines may serve as a starting point for action research at the school.
Levels of resource allocation fall significantly below those recommended by the
Evidence-Based Model. Additionally, implementation of evidence-based strategies for
school improvement occur at varying levels of consistency and robustness. For example,
professional development takes place at Chagall JHS, but not with the time allocation,
intensity, and follow-up support recommended by school improvement research findings.
Budget shortfalls have compelled the school to move away from a schedule and extended
assistance that seemed to student improvement. The principal has already noticed a
change in the school’s climate, including a surge in student violence, following the return
to a standard schedule. In the face of budget challenges, populations within the school
continue to approach student achievement and instructional improvement creatively. It
may be of special interest to examine the programs and departments associated with
student achievement at the end of this school year.
381
APPENDIX K
Degas Intermediate School: A Case Study of Instructional
Improvement and Resource Use
Degas Intermediate School is a grades six-through-eight school with 1,173
students in an urban district in Orange County, California. The district serves
approximately 42,000 kindergarten-through-grade-twelve students. Degas Intermediate
School is one of nine middle schools in the district. Three of the nine are labeled
“fundamental” schools and provide no services to English learners or special needs
students. Data released by the California Department of Education in September 2008
indicate that Degas Intermediate School (along with 75% of California Title I middle
schools) did not attain the minimum percentages of students with scores at or above
“proficient” in language arts (35.2%) and mathematics (37%) to meet federal Adequate
Yearly Progress goals. As a result, this Title I-assisted school has been identified as a
Year Five Program Improvement school for the 2008-2009 school year.
A number of state and federal classifications assist in sketching an image of the
population served at Degas Intermediate School. Although proportions attached to
various classifications are reported individually in California, many students fit multiple
descriptors. At Degas many English learners are also Latino as well as free and reduced
price lunch participants.
Almost 100% of the students at Degas are classified as “minority.”
Hispanic/Latino students comprise the single numerically significant ethnic subgroup and
382
represent over 99% of the student population at Degas. This ethnicity proportion is
slightly higher than the district’s 94.4% Hispanic/Latino population.
Approximately 96% of Degas Intermediate School students are eligible for free
and reduced price lunch, as compared with 88% of students district-wide. Sixty eight
percent of the school’s students actually participate in the free and reduced price meal
program at Degas. Approximately 11% of Degas students receive special education
services compared with 9% throughout the district. Students identified as English
learners comprise 87% of Degas’ student population, diverging from the district’s
English learner percentage (44%) at the middle school level.
During the time that Degas has grounded its student-learning goals in federal
accountability requirements, the Whitney Unified School District has attained district-
level federal Program Improvement status. Now in Year Three of federal Program
Improvement, Whitney USD also has been named a state “District Assistance and
Intervention Team” (DAIT) district. Both DAIT and Year Three Program Improvement
entail corrective action. The district currently employs cabinet-level support from the
county office of education as its external support and monitoring agency.
Of the nine intermediate schools in the district, six are in Year Five of Program
Improvement and one is in Year Two. All of the traditional middle schools have attained
Year Five designations and one fundamental middle school is in Year Two or Program
Improvement. The two intermediate schools not identified as “failing” by federal
standards are fundamental schools. While the graph below reflects the state Academic
Performance Index (API), its calculations heavily weight federal measures for language
383
arts and mathematics. It provides a succinct comparison of student achievement across
the district’s intermediate schools. As illustrated below, Degas student achievement test
results trail those of the other five “traditional” intermediate schools. Along with four
other intermediate schools, Degas has not broken the 700 API ceiling.
Figure 1. 2008 APIs - Whitney USD intermediate schools.
During the four years that the state has calculated school-, district-, and state-
level Academic Performance Indexes, Degas consistently has attained lower numbers
annually than those of the district overall. While California Department of Education
district-level reports do not calculate an average middle school API for school districts,
state-level reports provide an opportunity to compare middle school results and growth.
In 2008, Degas missed the state middle school average API by 122 points. The school
trailed the Whitney USD API (grades 2-11) by 86 points. During the 10 year period that
APIs have been calculated by the California Department of Education, Degas’ API has
grown 162 points: 441 in 1999 to 603 in 2008. The school, district, state middle school,
and state overall APIs grew at roughly the same rate between 2007 and 2008 (i.e.,
between 1.02 and 1.05%). However, the school to date has been unable to eclipse its
highest API, a score of 621 attained in 2005 at the end of the current principal’s first year
serving in this capacity at the school. As federal proficiency targets increase by
*Note: Fundamental school.
384
approximately 11% annually between 2009 and 2014, Degas’ growth must accelerate
dramatically in order to exit Program Improvement status. Figure 2 illustrates the
comparison between Degas, Whitney USD, and California APIs.
Figure 2. School-district-state API comparison 2005-2008.
Another look at Degas IS – Whitney USD achievement data highlights the
persistent gap between the school and the district for key Degas student groups. The
graph below depicts 2008 API results for significant student groups across the district as
well as for those groups at Degas. Not only does the district fall behind the state API
average, Degas also trails the district for each of the school’s significant student groups.
Academic achievement test results at Degas also illustrate the homogeneity of the
school’s student population. Nearly all of the school’s students are identified as
Hispanic/Latino, and Socio-economically Disadvantaged. Additionally, 87% of Degas’
students are identified as English Language Learners, virtually all of which come from
Spanish-speaking homes. The school’s enrollment of African-American, Asian, White
and other students is in single digits and is not recognized as significant for NCLB
385
reporting purposes. The school recognizes a persistent achievement gap between its
students and others throughout the district and state. While the number of Special
Education students in 2007-2008 resulted in this group being deemed insignificant for
reporting purposes, low test results among Hispanic/Latino, and English Language
Learner students prevented Degas from meeting its federal Annual Yearly Progress
objectives. Student achievement in language arts proved problematic for all student
groups.
Based on California Standards Test results, mathematics achievement is the high
point in calculating Degas’ school success. Twenty-four percent of Degas’ eighth-grade
students enrolled in Algebra during the 2007-2008 school year. Roughly 43% of those
Algebra students attained proficient or advanced scores on the California Standards Test
in this subject. The majority of eighth-grade students (74%) were enrolled in a pre-
Algebra course. They were tested on sixth and seventh grade California mathematics
standards. Twenty-one percent (74 students) attained proficient or advanced scores on
this assessment. The recent State Board of Education decision to test all eighth graders’
mathematics proficiency using an Algebra test by 2010 adds urgency to the school’s
efforts to provide effective Algebra instruction to all students. Degas teachers continue to
Figure 3. Degas IS – Whitney USD API Comparison Across Significant Student Groups,
2008.
386
consider the learning needs of Special Education and English Language Learner students
in this area. Figure 4 illustrates student achievement results across the four content areas
tested at the eighth grade including Algebra.
Note: Numbers appearing in the legend reflect percentages of each group’s representation in Algebra
alongside the percentage from that group with proficient or advanced scores on the California Standards
Test in Algebra. All other students took a test in General Mathematics, measuring grades 6 & 7 standards
to assess proficiency in this subject area.
Figure 4 Percentage of students attaining academic achievement scores at proficient or
advanced levels across student groups and subject areas – 2008.
The six-year period from 2003 to 2008 includes both the time that the school has
held Program Improvement status and the state has implemented criterion-referenced,
standards-based academic achievement tests annually (previously, norm referenced tests
were included in the battery). Additionally, 2003 marked the first year of a history-social
science exam administered in the eighth grade. This was followed in 2006 with an eighth
grade test in science. A multi-year, multi-subject view affords a longer-term perspective
on student achievement efforts. The graph below follows the high-stakes test results in
the four state-identified core content areas tested at the eighth grade during the 2003-
2008 period. Scores in science appear to be moving consistently and steadily upward
while the trajectory of language arts and social studies appear similar and relatively flat.
387
Achievement scores in Algebra dipped in 2004 and 2005 when the school
included as many as 88% of the eighth-grade student population in the course. Scaling
back the number of participating students, then gradually increasing it has coincided with
achievement score increases. Degas’ API has moved from 441 in 1990 to 603 in 2008 at
a lurching pace including a 31 point decline in 2006 from which the school is recovering.
Figure 5. Degas Intermediate School eighth grade California Standards Test proficient
and advanced scores in core content areas from 2003 to 2008.
During the past five years, Degas Intermediate School has concentrated school
improvement efforts on language arts, mathematics, science, and student behavior. The
school has worked to organize its faculty and schedule to support consistent collaboration
and increased instructional time in mathematics and language arts. The purpose of this
case study is to tell the story of efforts to improve student learning at Degas Intermediate
School and to identify the resources allocated toward those efforts.
388
Efforts to Improve Instruction at Degas Intermediate School
The structured interview protocol that guided data collection also frames the case
study. Program Improvement middle school principals responded to a series of
quantitative and qualitative questions regarding their 2008-2009 school improvement
plans and strategies. Principals’ responses provided quantitative resource data for
comparison with the Evidence-Based Model and told the story of how and why the
school utilizes its resources toward instructional and academic improvement. The “how
and why” part of the story follows.
Impetus for Change
When the principal joined Degas Intermediate School in 2003, the school had
entered Year Three of Program Improvement, the first year of “corrective action” under
the federal plan. Following several years of continual administrator changes the
professional staff had reached a point of low morale. The entire staff met repeatedly with
county office of education and other school improvement consultants to determine focal
points for school improvement efforts. Data analysis coupled with grant-funded
initiatives led the staff and leadership team to target language arts, mathematics, physical
education, and science instruction while additionally establishing effective systems for
supporting appropriate student behaviors.
Adopting New Curricula to Achieve Goals
District-adopted instructional materials in mathematics, science, and history-
social science in addition to a similarly adopted program for language arts support the
standards-based curriculum in place at Degas Intermediate School. The school also has
389
invested in Highpoint, a combination language arts intervention and English language
development program. Degas currently has no math support program in place for
struggling students. With its Advancement Via Individual Determination (AVID)
teachers, Degas also has begun to infuse research-based instructional methods such as
Cornell Notetaking into the school experience for some of its seventh- and eighth-grade
students. Step Up to Writing, a research-based writing development program has been
added to language arts classes throughout the school during third period. The school
additionally has purchased Mind Institute mathematics and algebra software for regular
classroom and remediation use.
Implementing the Research Based Curriculum
Time, course prioritization, and teacher training frame implementation of Degas’
curriculum. Each Degas student enrolls in two periods of language arts and mathematics
daily. History and science each have been reduced from full-year to one-semester
courses to create additional time for language arts and math instruction. Electives such as
art have been eliminated to maintain a focus on meeting federal Adequate Yearly
Progress objectives. All language arts and mathematics teachers have participated in a
40-hour, legislatively funded, and State Board of Education-approved training so they
may effectively implement district-adopted instructional programs. In addition to this
technical training, Degas maintains an ongoing contract with a language arts intervention
program specialist to coach and train teachers to implement Highpoint as the publisher
intended.
390
In addition to training specific to instructional programs, Degas teachers now
receive ongoing training and support in effective instructional methods. They turn in
lesson plans every Monday indicating instructional objectives, materials to be used, and
various assessments to measure student learning.
Support Teachers with Professional Development
Professional development is built into the teacher contract, the school calendar,
the school day, and district requirements. While funds to attend conferences have been
frozen by the district for the 2008-2009 school year, the district underwrites SB 472
training in language arts and language arts intervention program implementation to
ensure that all language arts teachers have the technical capacity to use their language arts
program as per the publisher’s intent. In addition to this 40-hour summer training with an
80-hour school-year practicum for language arts educators, all teachers at Degas
participate in 18 hours of professional development each year. The first 12 hours are
district-mandated and district-organized over the course of two days prior to the
beginning of the school year. The remaining six hours are met through a combination of
teacher-selected seminars and workshops as well as those selected and provided by the
school and district. Examples of ongoing professional development include departmental
common assessment and grading scheme development, unit design, technology training,
and grant-funded workshops (for the math department).
Degas holds a contract with UCLA’s School Management Program. A consultant
has trained administrators and department leaders to conduct “classroom walkthroughs.”
Every three weeks a team of observers visits classrooms, collects data based on the
391
classroom walkthrough protocol, then meets briefly with each observed teacher to share
findings.
The current teacher contract and school schedule permits weekly gatherings for
meetings and professional development during the contract day. Table 1 illustrates the
schedule of meetings that take place Wednesdays from 1:45 – 2:45 PM. On months with
five weeks staff and department meetings occur on an as-needed basis.
Table 1. Degas Intermediate School Monthly Meeting/Professional Development
Schedule
Week One & Week Three Week Two & Week Four
Staff meeting (agenda determined in advance by
leadership team)
Department meeting – focus on collaboration and
goal setting based on data as well as on specific
topics (i.e., Academic Program Survey, California
Standards, Blueprint Prep, & Categorical Program
Monitoring review)
Restructured School Day
Degas Intermediate School operates on an eight-period school day. However, the
first period class for seventh- and eighth-grade students consists of a short homeroom to
provide students with informational announcements and to create a time buffer for tardy
students to minimize any lost time in academic classes. As mentioned above all seventh-
and eighth-grade students receive two class periods of mathematics and language arts
instruction daily. Sixth graders do not have a homeroom period. Instead, they receive
approximately 10 minutes additional language arts instruction daily. One period each of
science, social studies, and physical education completes students’ schedules. As the
392
recipient of a California Quality Education Investment Act grant, Degas has been able to
reduce class size to approximately 30 students per class.
Coordinated Help for Struggling Students
Coordinated help for struggling students includes common assessments to
diagnose student need, a language arts intervention program, and after school learning
support. The district abandoned its benchmark assessment program due to its lack of
reliability, validity, or ability to predict subsequent student achievement on the California
Standards Tests. Degas teachers have begun to develop their own standards-based,
course-alike common assessments, which are used both for student grading and
identification of student need for additional academic support.
After school academic support and activities are funded through a district
partnership with a non-profit community-based organization. This program is staffed by
“Think Together” employees, creating challenges in terms of connecting directly to the
school’s curriculum. However, the school uses categorical funds to underwrite an after
school homework club.
Ongoing Parent Communication and Education
Parent communication and education occur through formal and informal channels
at Degas IS Degas’ office and administrative staff are bilingual, supporting informal
school-parent communication and ease of interactions with school personnel. The school
maintains a “Parent Room,” for meetings and to access school information. Additionally,
parents may enroll in a physical fitness class using the school fitness center alongside
their children. Students serve as official translators, dramatically increasing the number
393
of parents attending events such as Back to School Night and Open House each year.
Organized parent communication and involvement occurs through general parent
meetings and special information sessions, newsletters, English Language Advisory
Committee and Parent Teacher Organization meetings, and attendance at student
performances. A foundation supports parent education programs at Degas.
Instructional Leadership as a Foundation for Change
Degas IS continues to work on developing its teacher leaders. Administrative
turnover, lack of policies and systems for managing student discipline, and resistance to
change among teachers have colored Degas’ leadership picture. The administrative
offices at Degas Intermediate School have been subject to frequent changes in staffing.
The current principal has been in place since 2003 following four principals in as many
years. Assistant principals frequently move to other schools in the district or leave the
district altogether. Currently, one assistant principal also teaches part of the day and the
school is working to hire an assistant principal to fill a vacancy. The principal indicated a
concern with the constant administrative turnover and with a lack of self-directed
behavior on the part of current assistant principals.
Staff members initially resisted instructional improvement efforts at Degas IS
They identified a laundry list of reasons unrelated to instruction that resulted in low
student achievement rates: lack of student motivation, misbehavior, gang orientation, low
levels of English language competence, and lack of parent involvement. School
improvement efforts have included and continue to feature student discipline.
394
Use of instructional time represents leadership priorities associated with
improvement at Degas IS. While the school considered a block schedule to ensure
adequate instructional time in language arts and mathematics the staff elected to retain its
traditional schedule and double up on language arts and mathematics sections. Each
student takes two periods of each of these courses. In order to do this the school
eliminated one semester from its social studies and science programs. Its leaders also
removed art and most electives.
Degas has relied on external agencies to direct its school improvement efforts.
The local county office of education began to work with the school in 2004 and continues
to provide instructional and systems improvement support in the 2008-2009 school year.
UCLA’s School Management Program consultants also have worked with the school for
a number of years focusing on observing classroom instruction and using resulting data to
determine areas for improving lesson design. Whitney USD has hired Dataworks to film
professional development sessions, which emphasize differentiated instruction, data
analysis, and instructional delivery. Additionally, a team of school leaders has begun
Positive Behavioral Intervention Systems (PBIS) training in order to develop systems for
creating a positive school culture in which students become increasingly self-managing.
Trainings and consulting agreements requiring ongoing follow-up have proven
instrumental in developing a leadership team and in underscoring instruction in the
school’s improvement efforts. By maintaining contracts with a handful of consultants
who return year after year Degas’ leaders have created continuity and focus for
instructional improvement efforts.
395
Creation and Maintenance of a Professional, Collaborative School Culture
Modified Wednesdays, a professional learning community grant in mathematics
and development of common course assessments support an increasingly professional
and collaborative school culture. As mentioned previously, faculty and administrators
meet each Wednesday afternoon using contracted employment time to conduct meetings
regarding schoolwide initiatives, departmental objectives, and assessment development.
The math department participates in a federal grant in which instructional
improvement is predicated upon learning from mathematics experts and developing as a
professional learning community to continually visit a set of critical questions: What do
we want our students to learn? How will we know they have learned it? What will we
do in the times that they do not learn it? What will we do to enrich their learning when
they have learned it? The multi-year grant funds after-hours meetings, coaching,
seminars, and institutes.
District Leadership and Support
Degas’ improvement efforts during the past five years have been site-driven.
However, as the district has been identified for corrective action itself, it has become
somewhat involved with site-level improvement goals. The district has mandated each
school to conduct an Academic Program Survey in which staff members assess the level
of implementation nine Essential Program Components deemed critical to school
improvement by the state. As per NCLB requirements, Degas provided the district and
school board with a presentation of its Year Five Program Improvement plan for school
improvement.
396
As mentioned above, the district has underwritten ($24,000) the filming of
professional development involving Degas teachers and conducted by Dataworks. Also,
the district annually orchestrates and funds SB 472 training for language arts and
mathematics teachers so they may implement state-approved and district-adopted
research-based instructional programs as intended by the publishers. District efforts to
create partnerships to bring resources (i.e., “Think Together”) to schools include well-
intended ones with limited potential.
Bringing Professional Knowledge Into the School
Outside consultants have been hired to increase faculty and system capacity in a
variety of areas including lesson design, delivery of instruction, program implementation,
student discipline and school culture, and school leadership. Each member of the staff
engages with one or more consultants, ranging from individual coaching for
administrators and teachers to trainings for the entire staff. Some efforts require
participating staff members to extend their learning to colleagues. For example, a team
of teachers and administrators participates in PBIS training with one teacher serving as
the team lead. This teacher attends additional training sessions and leads team members
in developing school plans for creating a positive school culture in which students receive
instruction in appropriate behavior. This model of outward-moving concentric circles of
capacity development results in consultants meeting over time with a smaller group of
staff members who then influence the rest of the school.
397
Features of Improving Schools
An extensive study of school resource use in Washington to improve student
learning resulted in a set of eight key features associated with improving schools
(Fermanich, Mangan, Picus, Odden, Gross, & Rudo, 2006). These features include (a) a
focus on educating all students, (b) data-driven decision making, (c) use of a rigorous
curriculum that aligns with state standards, (d) effective professional development to
support instructional improvement, (e) a restructured learning environment, (f) extended
learning opportunities for struggling students. These six success features are facilitated
through (g) instructional leadership and (h) professional learning communities. These
eight findings provide a framework for summarizing the story of this school’s
improvement efforts.
Focus on Educating All Students
Degas’ professional development focusing on instructional improvement, creation
of a “Corporate Academy,” inclusion of Advancement Via Individual Determination,
prioritization of essential content areas, and foci on student health and positive school
culture reflect a commitment to educating all students. School leaders made difficult
decisions to eliminate courses in order to focus on basic academic skills while also
preparing students for college preparatory courses when they move on to high school.
The school has gradually increased the number of students participating in Algebra each
year as teachers develop instructional expertise to meet student learning needs.
Degas’ physical education department has emphasized health and student well
being. Through partnerships and grants the school has been able to equip a fitness center,
398
which is used both during and after the school day. Physical education teachers include
mathematics and health connections in their curricula.
Special learning opportunities including the Corporate Academy are limited to
students with proficient or advanced California Standards Test scores. This, along with
AVID is intended to encourage career and college aspirations as well as to support the
learning of students ready for acceleration.
Data-Driven Decision Making
The principal hopes to develop a professional culture in which data are used to
inform instructional practices. Data Director, a web-based subscription data service
houses Degas’ common assessment and high-stakes test results. Data may be sorted in
any number of fashions including by class, content standard addressed, and proficiency
levels. This year the principal and Instructional Leadership Team decided to create
scatterplots based on common assessment data to identify the classes, teachers, and
students with scores reflecting especially high achievement levels. The leadership team
hopes to uncover instructional practices being used within the school that appear to yield
outstanding results, then increase their use across the school.
In October 2008 all staff members attended a short workshop to refresh their
technical understanding of how to use Data Director. The program has been in place for
three years and most teachers do not use it. Instead, school administrators create data
reports that are shared and discussed at staff meetings.
Teachers participate in short feedback sessions every three weeks following a
group observation of their class. Using UCLA’s school walkthrough protocol, a team of
399
observers spends approximately 10 minutes in a class, completes observation checklists,
then meets immediately with the observed teacher to share what they observed. This
feedback provides an opportunity for teachers to consider modifications to their
instruction and class routines.
The principal hopes to move the faculty forward in terms of them independently
analyzing data to inform and modify instruction to meet student learning needs. The
above-mentioned workshop and monthly “data dialogue” meetings scheduled throughout
the year reflect an effort to institutionalize data-driven decision making.
Use of a Rigorous Standards-Aligned Curriculum
Degas teachers use state-approved, standards-based curricular materials selected
in accordance with their proven potential to support student learning. The district
provides mandated professional development in its state-approved language arts and
mathematics programs.
This is the first year that the principal has required teachers to submit weekly
lesson plans. Examination of these plans reveals a range of understanding among
teachers of what they want students to learn, the depth to which they want students to
learn it, and conceptual understanding of the difference between learning objectives and
activities. Effective instruction continues to be a focal point at the school.
Effective Professional Development to Support Instructional Improvement
Professional development is limited to 18 hours annually, with 12 hours
determined by the district. However, regularly scheduled collaborative meetings,
400
judiciously selected conferences, and ongoing coaching and consultation from outside
agencies support ongoing teacher learning.
Restructured Learning Environment
The school’s restructured learning environment includes multiple class periods
dedicated to language arts and math. At present, sixth-grade students meet with two
teachers daily as opposed to seventh and eighth graders who meet with up to six different
teachers each day. The leadership team is strongly considering a self-contained
classroom model similar to that in sixth grade for all grade levels. Modifications may
include complimentary course teaming (e.g., language arts and social studies taught by
one teacher) as course demands require increasingly greater content-area expertise. .
These features will minimize student transitions to different teachers while maximizing
differentiation to meet student instructional needs.
Extended Learning Opportunities for Struggling Students
Extended learning opportunities include an after school homework club, summer
school, and participation in “Think Together,” an after school recreation and academics
program provided at the school to all interested students at no fee. This program is
operated by a non-profit organization in partnership with the district. Neither the school
nor the district are involved in selection or implementation of program objectives or
activities.
Instructional Leadership
The Instructional Leadership Team at Degas consists of its administrators,
program coordinators, and department leaders. The principal continues to work on
401
leadership development among all staff members, including those on the leadership team.
For example, a teacher serves as Degas’ Positive Behavior Intervention and Support
(PBIS) team leader. This teacher attends additional PBIS training sessions in order to
facilitate orderly implementation of key elements associated with a positive school
culture. The principal recognizes staff member dependence on de facto leaders and on
the principal in particular as the nominal leader of the school. In developing teacher
leaders and more distributed leadership the principal notes the tension inherent in sharing
decision making and taking care of urgent affairs of the school.
Professional Learning Communities
Modified Wednesdays include scheduled departmental collaboration meetings as
well as some “free” afternoons in which departments may meet as needed. Departments
may determine additional collaboration time necessary for data analysis, common
assessment development, and planning. Substitute teachers provide release time for
teachers to meet based on departmental needs.
The mathematics department participates in ongoing professional development
grounded in professional learning communities. National experts provide both content
expertise in mathematics concepts and pedagogy as well as in how to function as a
community of professionals learning together. By adding a language arts coach to
Degas’ staff the principal hopes to institutionalize professional learning community
features within that department as well.
402
Essential Features and Doubling School Performance
Essential features associated with improving schools reflect the ten steps to
double school performance as determined by Odden (2007). Schools experiencing
dramatic improvement (a) conduct a needs assessment, (b) set higher goals, (c) adopt a
new curricular program, (d) implement data-based decision making, (e) invest in long-
term, extensive professional development, (f) use school time more efficiently, (g)
employ multiple extra-help strategies for struggling students, (h) create professional
learning communities dedicated to boosting student performance, (i) support instructional
improvement at all levels of leadership, and (j) utilize external professional expertise.
The side-by-side comparison in Table 2 illustrates the similarities between observed
elements of school improvement in Washington and those identified in studies of school
improvement across many states. These features appear in differing degrees across
schools included in this series of case studies.
Table 2. “Washington Learns” and Ten Steps to Double Student Performance – Critical
Elements of School Improvement
Washington Learns Critical Elements of School
Improvement
Ten Steps to Double Student Performance
Data-Driven Decision Making Needs Assessment
Focus on Educating All Students Higher Goals
Rigorous Curriculum Aligned to State Standards New Curricular Program
Data-Driven Decision Making Data-Driven Decision Making
Professional Development Long-Term, Extensive Professional Development
Extended Learning Opportunities for Struggling
Students
Multiple Extra-Help strategies for Struggling
Students
Restructured Learning Environment Effective Use of School Time
Professional Learning Communities Professional Learning Communities
Instructional Leadership Support for Instructional Improvement at All Levels
of Leadership
Professional Development Utilization of External Professional Expertise
403
In addition to these tightly aligned critical school improvement features, the
Evidence-Based Model proposed by Odden and Picus (2008) identifies quantifiable
research-based elements associated with improving and successful schools. Odden and
Picus have translated extensive research findings into formulas for establishing
appropriate levels of human and material resources to support adequate educational
outcomes for all students. Data collected from Degas Intermediate School provide a
point of comparison between current school resource levels and those suggested by the
Evidence-Based Model.
The Evidence-Based Model for Educational Adequacy and
Resource Levels at Degas Intermediate School
The Evidence-Based Model provides a yardstick against which current resource
allocations at Degas Intermediate School may be compared. Extensive school
improvement research informs the calculations and suggestions contained within the
model. Degas’ resource allocation diverges from most elements contained within the
Evidence-Based Model.
Key points of contrast between Degas’ configuration and resource allocation and
that suggested by the Evidence-Based Model include school size, number of specialist
teachers and tutors, quantity of professional development and instructional coaches as
support to teachers, number of secretarial and support staff members, number of guidance
counselors, and resources for a robust student activities program. If resourced as per the
model both extended day increase and summer school support would increase
substantially.
404
Degas is the recipient of a Quality Education Investment Act grant. It has used
these funds to decrease class size with the result of core class sizes averaging
approximately 28 students compared with 25 as suggested by the Evidence-Based Model.
Electives consist primarily of AVID classes and along with physical education represent
an addition of approximately 13% of teachers to the core teacher staff. The model
suggests additional core and specialist teachers as well as over 10 teachers to support the
needs of English Language Learners. As per the model struggling students at Degas
would receive one-to-one and small-group reinforcement during the school day from a
cadre of eight certificated teachers who serve as full-time tutors. Tutoring support
currently takes the form of two college students, each hired to work eight hours weekly to
work with AVID students.
If resourced in accordance with the Evidence-Based Model non-instructional
support for students (e.g., non-instructional aides, guidance counselors, & pupil-support
staff) would increase considerably. The school would invest in student activities and the
school library would increase its professional library staff from one-half to 2.6 full time
equivalent positions.
Professional development provides a final significant area of difference between
the Evidence-Based Model and current practice at Degas Intermediate School. While the
model suggests 20 days of focused professional development annually, Degas teachers
participate in up to six days of staff training with additional one-hour meetings weekly.
While this summary serves to highlight notable findings, Table 4 comprehensively details
the quantitative findings of this case study.
405
Table 3. Resource Comparison - Degas Intermediate School and Evidence-Based Model
(EBM) for Adequate School Resources
School Element Evidence-Based
Model Prototypical
Middle Schools
Degas Intermediate
School – Current
Resource Status
Degas IS – EBM
Comparison & EBM
Resource Suggestions
School
Characteristics
Configuration Grades 6-8 Grades 6 – 8 Same
School Size 450 1,173 2.6x larger than EBM
Core Class Size 25 27 1.1x larger than EBM
Free & Reduced Price
Lunch Count
(used below as a
calculator for school
resourcing)
798
Number of ELL
Students
(used below as a
calculator for school
resourcing)
1,021
Personnel Resources
(measured as FTEs)
1. Core Teachers 18 43.3 46.9
2. Specialist/
Elective Teachers
20% of Core Teachers 14.0% of Core
Teachers
(6.1)
9.4
Total Teachers 21.6 49.4 56.3
3. Instructional
Coaches
1 per 200 Students .4 5.9
4. Tutors 1 per 100 At Risk
Students (Free &
Reduced Price Lunch)
0 8
5. Teachers for ELL
Students
1 per 100 ELL
Students
.3 10.2
6. Extended Day 1 per 15 Eligible
Students (Staffing
based on 50% of Free
& Reduced Price
Lunch Students)
.12
Voluntary After
School Homework
Club 1 Hour 4x/Week
6.7
7. Summer School 1 per 15 Students
(Staffing based on
50% of Free and
Reduced Price Lunch
Students)
2.3
(15 Teachers 4.5
Hours Daily for 6
Weeks)
6.7
8. Alternative
Schools
NA NA NA
406
Table 3 continued
School Element Evidence-Based
Model Prototypical
Middle Schools
Degas Intermediate
School – Current
Resource Status
Degas IS – EBM
Comparison & EBM
Resource Suggestions
9a. Learning- and
Mildly Disabled
Students
3 Teachers per 450
Students
8.2
7.8
9b. Severely Disabled
Students
100% State
Reimbursement for
Top 1% Minus
Federal Funds
1
1 FTE Aide
Based on Need
10. Services for
Gifted Students
$25 per Student 0 $29,325
11. Career/ Technical
Education
.3 Weight per CTE
Student to Maintain
Low Class Sizes
$7000 per CTE
Teacher Contract
0
(Corporate Academy,
but No Specific
Career or Technical
Classes)
100 x 1.3 to Maintain Low
Corporate Academy Class
Sizes
$7000 per Corporate
Academy Teacher Contract
12. Substitutes 10 Days per Teacher
@ $100 per Day
Encumbered funds for
160 Days ($20,000)
of Substitute Teachers
(≈ 2.8 days per
teacher)
10 Days per Teacher at $100
per Day
13. Pupil Support
Staff
1 Guidance Counselor
per 250 Students plus
1 Pupil Support Staff
Position per 100 Free
& Reduced Price
Lunch Students
3 Counselors 4.7 Counselors
8 Pupil Support Staff
Members
14. Non-Instructional
Aides
2 per 400-500
Students
0.5 5.2
15 Instructional Aides 0 2 0
16. Librarians/ Media
Specialists
1 0.5 2.6
17. Administrator 1 Principal plus .5
Assistant Principal
1 Principal;
2.5 Assistant
Principals,
1 Title I Coordinator,
&
.5 Bilingual
Coordinator
1 Principal plus 2.1 Assistant
Principals
407
Table 3 continued
School Element Evidence-Based
Model Prototypical
Middle Schools
Degas Intermediate
School – Current
Resource Status
Degas IS – EBM
Comparison & EBM
Resource Suggestions
18. Secretary 2 School Site
Secretaries
3.3:
1 11-Month Office
Manager
1 11-Month Registrar
1 10-Month Office
Assistant
1 Part-Time Clerical
Support Assistant
5.2
Non-Personnel
Resources
19. Professional
Development
10 Days Intensive
Professional
Development in
Summer Included in
Teacher Contract
Instructional Coaches
Planning & Prep time
$100 per Student to
Fund
18 Hours
(3 Days) Professional
Development
Required by Teacher
Contract, 12 Hours at
End of Summer,
Remaining 6 Hours
During School Year.
SB 472 Language
Arts and Mathematics
Training in Summer
(40 Hours) with 80
Hour Practicum
during School Year
.4 FTE Mathematics
Instructional Coach
36 1-Hour
Staff/Collaboration
Meetings
$30,000 + .4 Coach +
0.2 Administrator
budgeted
7 Additional Professional
Development Days Added to
Current 3 – Group All Days
During Summer
Increase and Diversify
Instructional Coaches to
Provide Ongoing Modeling,
Coaching, & Instructional
Support
Additional Trainings &
Conferences as Needed
$117,300 to Fund
408
Table 3 continued
School Element Evidence-Based
Model Prototypical
Middle Schools
Degas Intermediate
School – Current
Resource Status
Degas IS – EBM
Comparison & EBM
Resource Suggestions
Dollar/Pupil
Resources
20. Technology Calculation: 1
Computer per 3
Students.
Distribution: 1
Computer per 4
Students & 1:1 Ratio
for Administrators,
Teachers, &
Secretaries
$250/Pupil
.5 FTE Technology
Coordinator
$7.50 Per Student
Annually for Data
Director Subscription
($8797.50)
Action Learning
Systems and Mind
Institute Software for
Self-Directed Student
Remediation in
Mathematics
1 Computer Lab
789 Computers
391 Computers
$293, 250 to Maintain,
Replace, & Update
Equipment and Software
21. Instructional
Materials
Formative
Assessments
Library Materials
$120/Pupil
$25/Pupil
$20/Pupil
Current State-
Approved & District-
Adopted Instructional
Materials in Each
Core Content Area,
Step Up to Writing,
Mind Institute
Software, Action
Learning Systems
Software
Developed by
Departments During
Collaboration
Sessions and Release
Days
State Library Grant
Funds
EBM Suggests $193,545
($140,760 Instructional
Materials, $29,325 Formative
Assessments, & $23,460
Library Materials)
Comparison: Instructional
Materials Funding
Realignment Program
(IMFRP), Categorical Funds,
Staff Time, & Library Funds
at District and Site Levels
provide resources
comparable to EBM
suggestions.
409
Table 3 continued
School Element Evidence-Based
Model Prototypical
Middle Schools
Degas Intermediate
School – Current
Resource Status
Degas IS – EBM
Comparison & EBM
Resource Suggestions
22. Student Activities $200/Pupil Lunchtime Soccer
League
1 Section (1/6 FTE)
Student Leadership
Class
After school
Activities and
Academics Program
Provided by Partner
Organization
After school Fitness
Club
After school Art Club
EBM Suggests $234,000
Allocated to Student
Activities
Comparison: The school
expends approximately
$50,000 to staff student
activities positions and
purchase materials.
Implications
While student achievement did not meet federal No Child Left Behind standards
for Adequate Yearly Progress, growth in academic achievement scores surpassed that of
middle schools across the state in 2008. Degas’ Academic Performance Index (API)
grew at a rate of 1.5% as compared with 1.2 for middle schools statewide. Yet its API
continues to lag behind that of all other traditional middle schools in the district and most
across the state. Resource allocation does not meet that suggested by the Evidence-Based
Model as reflected by both the quantitative comparison and features of improving
schools.
An infusion of grant funding has permitted the school to increase its number of
teachers in order to reduce class size. This offers the school an opportunity to assess the
research-suggested influence of smaller classes on student achievement. Efforts to
410
improve instruction and develop instructional leaders have resulted in a consistent
professional development theme, use of data to inform instruction, and use of outside
consultants to support distribution of leadership. Tracking instructional and leadership
change may inform both the school’s efforts and those of others also striving to increase
staff capacity, efficacy, and professionalism.
Modifying the school day has come at the expense of core courses. Students
receive one semester each year of science and social studies instruction. The high stakes
exams administered in the eighth grade include both these subjects. The eighth grade
history-social science test also has the distinction of being the only three-year cumulative
test included in the battery of standards exams. Degas has elected to focus its
instructional time on language arts and mathematics due to their significance for NCLB
reporting purposes. Implications include effects of curtailed instruction on student
learning and test results. More globally, implications include the possibility of reducing
core subjects or reframing student learning objectives to include skills and content
simultaneously.
While virtually all of the students at Degas are Spanish speakers, none has the
opportunity to develop academic bilingual skills that could be further developed in high
school foreign language courses. Elimination of foreign language from the school
curriculum restricts cultural validation and deeper understanding of linguistic structures
that could be helpful in English-language arts.
Two Degas IS staff positions are not contained within the Evidence-Based Model.
The school maintains a half-time Bilingual Education Coordinator and a Title I
411
Coordinator, both of which are quasi-administrative positions requiring expertise in
managing federal and state categorical compliance requirements. Relinquishing these
specialized positions holds psychic and paradigm implications for schools long familiar
with this staffing design component.
The Evidence-Based Model supports investing in a robust program of student
activities. Degas IS students participate in intramural sports during lunch and after
school. School-sponsored after school activities include art, fitness, and homework
clubs. Select students participate in school leadership and music classes. Through a
partnership with an endowment-funded non-profit organization Degas students also play
after school sports and engage in academic enrichment programs. The school values this
partnership while recognizing the limitations of an outside agency providing these
services. Challenges include a disconnect between academic and behavioral expectations
expounded during the school day by Degas staff members and after-school program
choices and expectations supported by agency staff members. Exploration of strong
partnerships might inform effective collaborations between educators and non-education
agencies.
During the current principal’s tenure, Degas Intermediate School has focused its
improvement efforts on effective instruction, school climate, and high levels of student
engagement. Through mandatory trainings such as those for language arts and
mathematics, teachers have been introduced to their instructional materials and the school
has met compliance requirements. The principal hopes to move beyond compliance to
expertise. The district affirms but is not actively involved in most of the school’s
412
improvement efforts. Implications stemming from the district’s current improvement
status include steps toward aligning school and district initiatives. Handled effectively,
the school’s staff could feel supported and their efforts validated. Handled poorly, the
school’s staff could feel demoralized. Leadership becomes critical at this juncture.
Several resources currently unavailable to the school, which in the principal’s
opinion would result in dramatic improvements in student achievement include funds to
hire instructional coaches and substantially increase professional development. Low
achievement levels among English Language Learners and Special Education students
might be areas addressed through concentrated professional development and coaching
efforts. Each of these items is supported by the research driving the model and reflects a
need for increased resource levels to accelerate improvement. This case study
underscores the significance of leadership in developing professionalism among staff
members and highlights the possibility of rapidly raising student achievement levels by
providing students with adequate levels instruction and support from highly competent
staff members
413
APPENDIX L
Picasso Middle School: A Case Study of Instructional
Improvement and Resource Use
Picasso Middle School serves 963 students in grades six through eight in an urban
district in Orange County, California. The Hermitage School District serves
approximately 6,400 students in grades kindergarten through eight. Picasso Middle
School is one of two middle schools in the district. Data released by the California
Department of Education in September 2008 indicate that Picasso (along with 75% of
California Title I middle schools) did not attain the minimum percentages of students
with scores at or above “proficient” in language arts (35.2%) and mathematics (37%) to
meet federal Adequate Yearly Progress goals. As a result, this Title I-assisted school has
been identified as a Year Three Program Improvement school for the 2008-2009 school
year.
A number of state and federal classifications assist in sketching an image of the
population served at Picasso Middle School. Although proportions attached to various
classifications are reported individually in California, many students fit multiple
descriptors. For example, many English learners are also Latino as well as free and
reduced price lunch participants.
Student demographics at Picasso are nearly identical with those of the district
overall. Roughly 82% of Picasso Middle School’s students are classified as “minority,”
with 77% listed as Hispanic/Latino. The next largest minority group consists of Asian
students and comprises 2.6% of the school’s enrollment. Sixty-two percent of Picasso’s
414
students are considered economically disadvantaged and 53% are identified as English
learners. In contrast with the district, 8.3% (80 students) of Picasso’s students receive
special education services, whereas 11.3% are identified for special eduction district-
wide. A final indication of demographic characteristics closely resembling those across
the district includes 8.3 % of Picasso’s and the district’s enrollment representing
youngsters identified for special education services. Student groups numerically
significant for federal achievement data reporting purposes include those identified as
Hispanic/Latino, White, English learner, special education, and economically
disadvantaged.
Picasso’s student body has undergone demographic changes between 2003 and
2008. The percentage of White students has declined from 35% of the student population
to 17.7%. A corresponding increase in Hispanic/Latino students is reflected in the
increase in this group’s representation from 61% to 76.2%. The percentage of English
learners has not increased in such a notable fashion, moving upward from 33% in 2003 to
35.5% in 2008. This implies that the increased percentage of Hispanic/Latino students
consists primarily of English speakers. Finally, the percentage of students identified as
economically disadvantaged has increased from 50% of the student population to 61.2%.
While the school has experienced recent demographic changes, perennially low
achievement test results in two stable student groups, English learner and special
education, led to Picasso attaining Program Improvement status in 2007.
Both Picasso and the district’s other middle school share Year Three Program
Improvement status. The Hermitage School District also wears the badge of Year Three
415
Program Improvement. The district attained this distinction as a result of its two middle
schools failing to meet Annual Measurable Objectives in terms of proficiency targets in
language arts and mathematics with English learners and special education students.
A glance at student language arts achievement rates during the six years that the
California Standards Tests have measured student academic achievement in a criterion-
referenced format highlights several trends. Even with a recent decline, White students
outscore other subgroups by as much as 40%. In both 2005 and 2007, no eighth-grade
special education student attained a proficient or higher language arts test result. Of the
24 special education eighth-grade students tested in language arts in 2008, two attained
scores deemed proficient. Eight of Picasso’s 128 eighth-grade English learners attained
proficient or higher scores in language arts.
While improvement rates for special education and English learner students
appear relatively flat and that of White students seems to be declining, improvement for
students identified as economically disadvantaged or Hispanic/Latino has roughly
doubled between 2003 and 2008. Picasso strives to overcome a persistent performance
gap between White and all other student groups combined with lower-than-federally-
acceptable achievement levels among all non-White student groups. Figure 1 illustrates
this six-year trend.
416
Figure 1. 2003-2008 language arts achievement rates among eighth grade students at
Picasso Middle School.
In addition to improving performance to meet federal achievement targets for all
student groups in language arts, Picasso Middle School also must improve performance
in mathematics. Again focusing on eighth grade as a sample of Picasso’s student
population, test results in both general mathematics and Algebra reflect challenges in
meeting NCLB achievement goals. Additionally, a currently contested State Board of
Education decision to mandate Algebra testing for all eighth-grade students in 2010 adds
pressure to dramatically increase the number of students prepared for and taking Algebra
in the eighth grade. In 2008, just over 15% of eighth graders at Picasso participated in
Algebra. Of the 361 eighth-grade students participating in the California Standards Tests
in spring 2008, 55 took the Algebra test. Three English learners and no special education
417
students were among the Algebra test takers. Figure 2 illustrates the relatively flat
participation rates in Algebra during the past six years.
Figure 2. 2003-2008 Algebra participation rates among eighth-grade students at Picasso
Middle School.
In addition to low participation rates, eighth-grade student achievement in
Algebra has fluctuated wildly both during the six-year period between 2003 and 2008,
and across student groups within a school year. Notable are 2005 when no White
students attained proficient results, and 2006 when the same held true for Picasso’s
economically disadvantaged students. Gaps appear in the picture for English learners and
special education in years when the California Department of Education recorded no data
regarding their participation. However, in the years for which data are provided, no
418
students from either group have attained proficient or advanced results in Algebra.
Figure 3 illustrates the complexity of Picasso’s school improvement challenge.
Fig
ure 3. 2003-2008 Algebra proficiency rates among eighth-grade students at Picasso
Middle School.
A multi-year examination of student achievement reveals history and trends.
Looking at one year across disciplines identifies current areas of strength as well as what
is happening within one cohort of students. As mentioned above, federal proficiency
targets in 2008 for language arts and mathematics were 35.2% and 37% respectively.
The White student group was the only one to attain an adequate percentage of students at
and above proficient to meet NCLB requirements. No student group’s combined general
math and Algebra scores attained the 37% of proficient scores necessary to make
Adequate Yearly Progress. Each of these targets increases by approximately 11% in
419
2009. Assuming last year’s eighth-grade class was relatively similar to this one, Figure 4
provides a sense of the distance Picasso staff and students must cross to meet upcoming
federal benchmarks.
Figure 4. 2008 California Standards Test achievement rates among eighth graders.
During the past two years since acquiring Program Improvement status, Picasso
Middle School has focused its improvement efforts on effective instruction and support
for struggling students. This has entailed changes to the student and staff schedules,
professional development, and additions to the professional staff roster. The purpose of
this case study is to tell the story of efforts to improve student learning at Picasso Middle
School and to identify the resources allocated toward those efforts.
420
Efforts to Improve Instruction at Picasso Middle School
The Evidence-Based Model Informs the structured interview protocol that guided
data collection and frames this case study. Program Improvement middle school
principals responded to a series of quantitative and qualitative questions regarding their
2008-2009 school improvement plans and strategies. Principals’ responses provided
quantitative resource allocation data for comparison with the Evidence-Based Model and
shared the strategies they employed to improve instruction and academic achievement.
The “what, why, and how” part of the story follows.
Impetus for Change
Program Improvement status highlighting a long-standing struggle to provide
English learner and special education students with educational experiences that will lead
to proficient or advanced results on the state’s annual tests provided Picasso Middle
School with a strong impetus to change. The principal and staff recognized that
instruction as provided was not resulting in learning, leading to a focus on effective
teaching.
Setting Goals
Picasso’s staff uses NCLB proficiency targets for student achievement goals.
Additionally, the school has established goals for teachers with regard to improving
instruction. Goals include identification of learning goals and provision of differentiated
instruction to meet student-learning needs. These goals provide the focus for
professional development at Picasso with the expectation that improved instruction will
result in improved student performance.
421
Adopting New Curricula to Achieve Goals
In addition to state-approved instructional materials in each core content area,
Picasso adopted Language! for reading intervention, Guided Language Acquisition
Design (GLAD) to teach academic language skills while teaching content, and
Advancement Via Individual Determination (AVID) to provide a college-preparatory
experience for students who may not otherwise consider college an option. Even more
significant to the principal than the curricular choices is effective implementation of the
selected curriculum.
Implementing the Research Based Curriculum
Effective delivery of instruction at Picasso involves detailed lesson planning,
appropriate student placement, strategic scheduling, and adequate staffing, and use of
data to measure student learning. In working with Linda Moore from Targeted
Leadership, an education consulting firm contracted by the district, Picasso has
developed a new lesson format. “Think, Do, Prove” is a modification of the lesson plan
designed by Michael Scott, a principal in Oakland, California. Picasso’s protocol
includes a differentiation component for instructional activities. The essentials of the
format include (1) what the students will learn, (2) what the students will do, (3) what
the teachers will do, and (4) how the teachers will prove that their students have attained
the learning objective(s). Teacher accountability includes turning in lesson plans weekly
to the principal. Data from the assessment component contribute to identification of
students in need of additional assistance.
422
Multiple measures including previous year’s California Standards Test, California
English Language Development Test, and grades assist in placing students in appropriate
language arts and mathematics classes. Based on placement results, over half of
Picasso’s seventh- and eighth-grade students take at least two periods of language arts
instruction daily. All sixth graders receive two periods of language arts as well. Using
the “Excel” model from Hesperia, students requiring additional support in language arts
all receive instruction in the core program. This is different from schools in which
students more than two years behind are placed in an intensive intervention program to
quickly build the skills that will permit them to rejoin their peers in the core program.
Additional-assistance students are placed in smaller classes. In addition to a period of
instruction in the core program, these students also receive a period of reading
intervention or English language development. In order to accommodate this additional
language arts class, these students forego an elective class. The school uses a mental
model of 20 to 22 students per class to support struggling learners. To staff these
courses, the principal made a special request to the district and received funding to hire
one additional teacher.
2008 marks the first year that Picasso has subscribed to Data Director, an online,
standards-aligned student assessment data management system. Data Director permits
teachers to analyze student formative and benchmark assessment results to inform
instructional needs of students. This system serves the goal of teachers identifying in
their lesson plans how they will prove that students have learned the objectives of a given
lesson, and further, how teachers will use assessment data to modify future instruction.
423
As a result of subscribing to Data Director, the principal hopes that ongoing formative
assessment will become a key element of instructional improvement at Picasso Middle
School.
Restructuring the School Day
Picasso recently switched from a seven period day to one with six classes daily.
This change resulted in fewer daily instructional minutes for students. However, it
allows for longer individual class periods and for daily common meeting, professional
development, and planning time for all teachers. The current schedule begins with
teacher prep and meeting time between eight o’clock and 8:50. The school day begins at
8:58 AM and ends at three o’clock.
The restructured school day presents challenges for students in need of
mathematics support. The school offers no additional mathematics intervention courses
or support classes during the school day the way it does with language arts. However, as
the principal and teachers prepare to purchase new mathematics materials, they are
considering a schedule adjustment in which math intervention and science both might be
offered as trimester courses. Struggling students would then receive the additional
instructional time in mathematics suggested by the state and take less science (a lower-
stakes subject).
Restructuring the school day has led to challenges with the teaching staff as well.
In conjunction with creating a common meeting, professional development, and planning
time daily, the principal required attendance at a weekly professional development
session linked to the school’s instructional improvement goal. Additionally, the principal
424
required participation in weekly professional collaborative meetings in which
departments examined student data, set student-learning goals, and considered how to
effectively teach to those goals. As a result of union concern with the loss of teacher prep
time, the principal has rescinded the collaboration day but maintains the weekly
professional development meeting.
As all Picasso teachers have the same prep time, the principal hopes that
departments have continued to meet at least weekly. Preparing lesson plans as per the
required protocol provides a ready opportunity for course-alike teachers to meet and plan.
Weekly professional development sessions includes collaborative development of
systems to help struggling students.
Coordinated Help to Struggling Students
Addressing the needs of struggling learners is a centerpiece of Picasso Middle
School’s improvement efforts. Both the district office and the school take active roles.
Identification, intervention, and instruction form the foundation of the combined district
and school efforts.
The district uses standards test scores, student grades, and office referrals to create
its own at-risk list to determine summer intervention attendees. Two district-funded
employees, a special education teacher and a counselor, split their time between Picasso
and the district’s other middle school to support the schools’ at-risk students under the
auspices of special education support.
As its first effort to meet the needs of all learners, including those who struggle in
various classes, Picasso teachers plan their lessons using a staff-developed lesson plan
425
protocol in which teachers identify how they will differentiate instruction for learners
with language, readiness, or other needs. Additionally, Picasso’s staff has added school
climate and student behavior as part of its instructional plan.
This year, a team of Picasso teachers and administrators is participating in
Positive Behavioral Interventions and Supports (PBIS). In addition to increasing support
for students academically, the school is focusing on teaching students appropriate school
behaviors and attitudes. As the year progresses, the team will conduct planning sessions
with the staff to develop a school-wide system of expectations, mechanisms to teach
those expectations, and consequences for both attainment and breech of expectations.
Use of multiple data sources supports placement of students in appropriate
classes. Students whose previous results indicate a need for more intensive instruction in
English/language arts are enrolled in an extra period of reading or English language
development. Their classes are limited to no more than 22 students and each class
includes an instructional aide. Picasso somewhat blurs the lines between special and
regular education by using resource specialist (mild-to-moderate special education)
teachers to teach reading intervention classes. Both regular education and special
education students are assigned to these courses based on performance data. Previous
performance likewise determines placement in one of three levels of mathematics:
remedial, grade-level, or accelerated.
Special education students identified as needing intensive reading comprehension
and vocabulary support participate in the Lindamood Bell program after school as per
their Individual Education Plans. Student behavioral and academic contracts, follow-up
426
with a staff specialist, parent meetings, parenting classes, and mandatory summer school
for students who fall behind describe the essentials of the school/district-controlled
response to students who struggle.
Additional support is available to all students via voluntary homework and
tutoring after school, as well as through the California Prop 49-funded After School
Education and Safety Program (ASES), which underwrites a partnership in which the
local Boys and Girls Club provides recreational and academic activities each afternoon
until six o’clock.
At Picasso, resource classroom aides collect data regarding student work
completion and academic progress. These data, along with rough information from the
district and results from grade-level team meetings regarding student achievement, serve
as the focal points for weekly meetings attended by the principal, assistant principal,
school psychologist, Title I staff specialist, special education support specialist, and
special education counselor. The principal notes the significance of these meetings.
Students who otherwise might go unnoticed emerge during the data-driven conversations.
In addition to considering various intervention possibilities, the school and district
consider how to most effectively separate students who together form academically and
socially counterproductive combinations.
When school-level interventions do not result in desired student achievement
gains, parents attend formal meetings, which often include the special education support
specialist and counselor. Here, staff specialists set specific goals with students and
suggest parenting classes as needed for parents.
427
Picasso’s principal noted concern regarding adequate support in mathematics.
The new state-approved materials and protocols for math instruction suggest specific
numbers of instructional minutes in order to adequately teach students at different levels
of preparation and readiness. As a Program Improvement school, Picasso feels the
pressure to meet these suggestions. This will require finding a way to squeeze additional
math time into the school day for many students. One strategy under consideration is to
teach science and math intervention as trimester classes. This will reduce instructional
time in science, an area considered conceptually easier to teach at the middle school
level.
Figure 5 illustrates Picasso’s current response to students whose performance
indicates the need for additional support.
428
Figure 5. Response to Intervention at Picasso Middle School.
The principal explained that instruction at Picasso must improve if student
achievement results are to increase. Use of data to determine areas of student need and
development of teachers’ skills in differentiating instruction to meet those needs are
critical elements of the school’s plan to support teacher and student growth
simultaneously.
Supporting Teachers with Professional Development
Both the school and the district support Picasso teachers with professional
development. In addition to ongoing state- and categorical-funds professional
Tier One – All Students
• “Learn, Do, Prove” Lesson Plan Format
• 6
th
Grade: Two periods of language arts, and alternating days of electives from an
exploratory “wheel” and physical education
• 7
th
& 8
th
Grade: Six-period class day including language arts, math, science, social studies,
physical education, and an elective
• After-school clubs and sports Monday – Thursday (10 activities daily ranging from Magic
Club to Track & Field)
• After-school homework help and tutorials
• After-school recreation and academic program operated in partnership with the local Boys
and Girls Club
• Zero Period Band: permits all interested students to participate regardless of impact of
intervention classes on electives choices during the school day
Tier Two – Students Demonstrating Low Standards &/or
Benchmark Test Results, Failing Grades, &/or
Behavioral Challenges
• “Excel” Model for Language Arts Instruction: 1 period of core + 1 period of
reading intervention or English language development in smaller classes with
instructional aides.
• Academic/Behavioral Counseling and Goal Setting
• Parent-Student-Staff Meetings
• Parenting Classes
• District identification for summer intervention
429
development programs provided through the district, the Hermitage School District has
hired Targeted Leadership, an education consulting firm to provide professional
development to each Program Improvement school. Picasso Middle School’s teacher
schedule includes weekly professional development time, and teachers are encouraged to
attend at least one conference or seminar annually.
Examples of district-supported professional development include Beginning
Teacher Support and Assessment (BTSA), a state-funded, two-year induction program
for teachers new to the profession and Peer Assistance and Review, a categorical program
for struggling teachers. Picasso Middle School contributes to district and regional, which
in turn provide professional support and training in the areas of special education, health
and nutrition, and Guided Language Acquisition Design (GLAD). The majority of
professional development sessions are offered through after-school workshops or release
time during the school day.
Picasso Middle School’s current school schedule includes a common period for
planning, preparation, and professional development each morning before students arrive
for instruction. The principal has arranged for each Monday morning to be dedicated to
professional development. This year’s focus is on differentiated instruction based on
work by Tomlinson in meeting the needs of students with varying levels of academic
preparation. As mentioned above, the school has worked with Targeted Leadership to
develop a lesson plan format, which emphasizes learning goals, methods of teaching to
various student needs, and assessing student learning. The principal provides staff
members with readings and articles for discussion at these Monday meetings. With its
430
subscription to Data Director, Picasso’s on-site professional development Monday’s
include lessons in use of the online program so teachers may effectively diagnose
students’ learning challenges and address them.
The principal maintains a “pledge to teachers” that they may attend any
conference or seminar of interest during the school year. Popular choices each year are
the California League of Middle Schools conference and seminars featuring Kate
Kinsella, an expert in academic literacy and language development in culturally diverse
classrooms.
The teacher contract includes three days of professional development. As
identified above, these days are augmented with district-supported and site-level learning
opportunities for teachers. In line with the district, Picasso focuses on professional
development to improve instruction and utilizes the services of an education consulting
firm toward that end.
Bringing Professional Knowledge Into the School
Picasso turns to the outside to inform the efforts that take place within its walls.
Use of a consulting firm and visits to other schools help the school consider focal points
for improving instruction.
As mentioned above, Picasso Middle School works with Linda Moore of Targeted
Leadership Consulting. Student performance data influenced the school’s decision to
focus on improving instruction. Together with the consultant, Picasso’s staff
collaboratively developed a lesson plan format to which each teacher adheres and for
which each teacher is accountable.
431
Various members of the teaching and administrative staff have visited other
schools to glean ideas. Trips include one to Vancouver, Canada to examine systems in
place in a school experiencing dramatic improvement.
The principal provides faculty with professional readings weekly as part of their
staff development. The intent is that discussion about instructional improvement articles
will cement the school’s focus and reinforce professional development activities. Weekly
readings reflect the principal’s efforts to create a professional and collaborative school
culture.
Creating and Maintaining a Professional, Collaborative School Culture
Energies expended to create a professional and collaborative school culture have
led to mixed results. One significant rationale for the daily common prep period was to
provide teachers with maximum opportunity to meet within departments and grade levels,
as well as to create ad hoc collaborations. In addition to the weekly professional
development day mentioned above, a second day each week included professional
learning community meetings and collaborative planning time. However, the union
voiced concern about teachers losing control of their prep time. Following a joint union-
principal-superintendent meeting, the school retained the weekly professional
development session. Departments utilize the collaboration time on an “as needed” basis.
As with making changes to the student schedule, adjusting the staff schedule has
resulted in challenges to its improvement efforts. An example of the principal’s
leadership in the face of this setback has been the use of professional development
Mondays, time with Targeted Leadership, and now use of Data Director in such a fashion
432
that teachers may be inclined to meet with each other for support and to work “smarter,
not harder.”
Using Instructional Leadership as a Foundation for Change
Finding resources, using data, identifying key focal points, creating time for
regular professional development and collaboration, and developing areas of
accountability reflect leadership directed toward instructional improvement at Picasso
Middle School. The principal has served in this capacity at this school for over 20 years
and has extensive knowledge of district systems and resources. As a result, the principal
knows about and has tapped into Title IIB professional development funds previously
hoarded by the district office. The principal successfully lobbied for an additional FTE to
support the school’s plan for improving language arts instruction. Participation in Title I
school regional network meetings has provided the principal and leadership team with
additional ideas for improving student achievement at Picasso Middle School
Although weekly organized professional collaborative time did not materialize,
effective use of Data Director, creation of a school-wide lesson plan format, development
and implementation of a behavioral system school-wide, lesson plan accountability, and
providing students with additional supports all necessitate increased collaboration.
Whether intentional, intuitive, or accidental, school-and district-level leadership efforts
have resulted in task-based collaborative efforts.
433
Benefiting from District Leadership and Support
The Hermitage School District provides resources and direction to Picasso Middle
School. District support reinforces the school’s realization that increased student learning
and achievement gains depend upon improving classroom teaching.
Resources include an additional language arts teacher and student support
specialists. Professional development from the district includes state-funded beginning
and veteran teacher support services as well as in-services in GLAD, a district-adopted
strategy for teaching language skills and content simultaneously.
In terms of school improvement, the district requires that Picasso use the services
of Targeted Leadership but leaves it to the school and the firm to determine what the
services will be. In addition to resources and direction for the school, the Hermitage
School District provides on-line access to district and school information via the district
Web site. This supports the school’s efforts to communicate with and educate parents.
Communicating with and Educating Parents
Communicating with, educating, and involving parents at Picasso Middle School
includes both formal and informal events and opportunities. Parents are introduced to
middle school by attending one of nine orientation sessions offered in August before
instruction begins. Additional communication and education experiences take place
throughout the year.
Traditional communication mechanisms employed at Picasso include Back-to
School night each fall and Parent-Teacher Conferences in late-fall and spring. Each
parent receives a Parent Handbook in Spanish or English describing need-to-k