University of Southern California Dissertations and Theses

Academic, environmental and social integration variables that maximize transfer preparedness for Latino community college students: An application of academic success models to the study of Tran...

(USC Thesis Other)

Academic, environmental and social integration variables that maximize transfer preparedness for Latino community college students: An application of academic success models to the study of Tran...

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content ACADEMIC, ENVIRONMENTAL AND SOCIAL INTEGRATION VARIABLES
THAT MAXIMIZE TRANSFER PREPAREDNESS FOR LATINO COMMUNITY
COLLEGE STUDENTS: AN APPLICATION OF ACADEMIC SUCCESS
MODELS TO THE STUDY OF TRANSFER AND RETENTION OF URBAN
COMMUNITY COLLEGE STUDENTS (TRUCCS)
by
Rita M. Cepeda
A Dissertation Presented to the
FACULTY OF THE ROSSER SCHOOL OF EDUCATION
UNIVERSITY OF SOUTHERN CALEORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF EDUCATION
May 2003
Copyright 2003 Rita M. Cepeda
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
UMI N um ber: 3 1 0 3 8 6 6
Copyright 2003 by
Cepeda, Rita Maria
All rights reserved.
®
UMI
UMI Microform 3103866
Copyright 2003 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
University of Southern California
Rossier School of Education
Los Angeles, California 90089-0031
This dissertation written by
R ita M. Cepeda
under the discretion of h e r D issertation C om m ittee,
and approved by all mem bers of the C om m ittee, has
been presented to and accepted by the Faculty of the
Rossier School of Education in partial fulfillm ent of the
requirem ents for the degree of
Doctor of Education
May 16, 2003
Dare
Deim
v /
Dissertation C om m ittee
C hairperson
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
ii
DEDICATION
This dissertation is dedicated to my soul mate and husband Juan Tobias
Cepeda, to my daughters Mariel and Rocio Cepeda, to my parents Juan Manuel and
Maria Dolores Aviles, Nicaraguan immigrants whose pursuit of the American Dream
of an education for their children made my own achievements possible. And to my
sisters and brother, Dolores Olson, Juan Aviles, Martha Aviles, and Sylvia Aviles
who shared in the belief that our children would become second generation college
graduates.. .and so they have. Finally, this dissertation is dedicated to the students,
faculty and staff at Santa Ana College where I am privileged to serve as president; a
community college where students, in their daily pursuit of educational access, defy
the limits of any and all theories about aspirations and motivation.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
iii
ACKNOWLEDGEMENTS
I would like to express my appreciation to those individuals who have been
most helpful in the completion of this study.
My deep gratitude to Dr. Linda Serra Hagedom, chair of my dissertation
committee and the individual most responsible for pulling me out of ABD purgatory
wherein in I languished much too long.
To Dr. Melora Sundt and Dr. Lawrence Picus also members of my
dissertation committee for their invaluable advice and patience.
To the Transfer and Retention of Urban Community College Students
(TRUCCS) research team including from USC; Linda Serra Hagedom, Ph.D.;
William Maxwell, Ph.D.; Reynaldo Baca, Ed.D.; Estela Mara Bensimon, Ed.D.; Hye
Sun Moon, Ph.D.; Phil Brocato, Project Coordinator; Scott Cypers; Research
Assistant; Nurgul Kinderbaeava, Intern Student.
Finally, I would like to acknowledge the support of Dr. Juan Francisco Lara,
Assistant Vice Chancellor, University of California, Irvine for providing me with a
much needed safe haven where I could complete my writing and Scott Simpson,
Research Analyst, University of California, Irvine for his foundational support in
helping me to develop the research design for this study and to tackle my nemesis,
“descriptive statistics.”
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
iv
TABLE OF CONTENTS
DEDICATION .................................................................................. ii
ACKNOWLEDGEMENTS ..... iii
LIST OF TABLES ................... vi
LIST OF FIGURES............................ ..vii
ABSTRACT.............. vii
CHAPTER I: THE PROBLEM AND ITS UNDERLYING THEORETICAL
FRAMEWORK......................... 1
Introduction ....... 1
Background of the Problem .................................................. 2
Statement of the Problem.......................................................... .5
Purpose of the Study ..... 6
Significance of the Study .............................................................................. 8
Research Questions..............................................................................................8
Definition of Terms.......................................................................... 8
Methodology .................................................................................. 11
Assumptions..................... 11
Limitations .................................................................... 12
Delimitations.................................................. 12
Organization of the Study ........................................................ 12
CHAPTER II: REVIEW OF THE LITERATURE ...................... 14
Introduction............................................... 14
Synthesis of the Literature .................. 15
Conclusions .............................................................................................. 49
Implications................................................... 51
Recommendations for Future Research................................................... 53
CHAPTER III: METHODOLOGY ....... 55
Introduction ....................................................................... 55
Research Question.............................................. 55
Community College Model for Student Life and Course Completion .......... 57
Methodology ....................................................... 59
Data Analysis.......................................................................................................63
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
V
CHAPTER IV: RESULTS....................................................................................... 65
Introduction........................................................................................................65
Construct Validity and Reliability ................................................. 66
Statistical Analysis..................................................... 69
Model Summary.................................................................................................89
CHAPTER V: SUMMARY CONCLUSIONS AND RECOMMENDATIONS ... 94
Introduction........................................................................................................94
The Purpose of the Study....................................................................... 94
Summary of Findings................................................................ 95
Discussion.................................................................. 100
Implications.................................................................................................... 100
Conclusions................................................................................ 102
Recommendations for Future Research.............................. 104
REFERENCES ........................................................ 106
APPENDIX A:
Community College Student Life Model Factorial Validation Tables. ............... 113
APPENDIX B:
Data Analysis Descriptive Statistical Output Tables........................................... 125
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
vi
LIST OF TABLES
Table Page
1. California Community College Full Year Transfers to UC, by 32
Ethnicity
2. California Community College Full Year Transfer to CSU, by 33
Ethnicity
3. Summary and Comparison of Statewide Transfer Rates for All 37
Students vs. Students with “Transfer Intent”
4. Transfer Rates in the Year 2000 by Gender and Ethnicity Fall 1994 38
Cohort of Students with “Transfer Intent”
5. Annual Counts of Transfer Directed, Transfer Prepared, and Transfer 39
Ready 1997-2001
6. Pattern Matrix of Scales Selected 67
7. Los Angeles Community College District Student Headcount by 70
Ethnicity by College for 2002 Spring Semester
8. Demographic Descriptors/Establishing Representational Value (RV) 71
9. LACCD 2001 Full-Time Faculty Demographics 72
10. LACCD 2001 Part-Time Faculty Demographics 73
11. Mean Academic Performance Variables for Colleges with High, 75
Moderate and Low Latino Student Representation
12. Latino Students Remedial and Non-remedial Enrollment Patterns in 77
Gate Keeping Courses
13. Transfer Level English and Mathematics Course Enrollment for 81
Native and Non-Native Speakers of English by RV Level
14. Patterns of Remedial Course Enrollment Variables for Native and 85
Non-Native Speakers of English by RV Level
15. Block Entry Analysis of the Impact of Institutional, Individual and 91
Demographic Variables on Latino Student Success: Regression with
Scale Values
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
LIST OF FIGURES
Figure_____________________________________________________________ Page
1. Ethnic Representation of California’s Adult Population in 1993 20
Compared with 2010.
2. Los Angeles Unified School District Graduates Eligible for UC and 29
csu.
3. Enrollment by Ethnicity in California Four-Year Colleges and 31
Universities
4. Community College Model for Student Life and Course Completion 58
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
ABSTRACT
This study documents the significance of community colleges as the primary
access point to postsecondary education for Latino students, the largest growing
demographic group with the greatest gap in educational achievement nationwide.
The investigation is a secondary analysis of the Transfer and Retention of Urban
Community College Students (TRUCCS) survey data and is designed to examine the
causal relationship between academic success and such factors as student and faculty
demographics, as well as those academic, environmental and social integration
variables supported in the database. In particular the study uses measures of
academic success to ascertain the role that community colleges play in preparing
Latino students for transfer.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
1
CHAPTER I
THE PROBLEM AND ITS UNDERLYING THEORETICAL FRAMEWORK
Introduction
Equal access to educational opportunity has been established as one of the
primary factors for sustaining the strong fabric of American society. Education is
indisputably acknowledged as the vehicle necessary to sustain a strong economy, the
exercise of civic and political involvement in a democratic society and ultimately the
source for future development and innovation. The American education system, and
in particular its promise of equity and access for all, has been the envy of the rest of
the world that ultimately recognizes education as the cornerstone of a free,
democratic society. Most students of the American system of education are often in
awe of the unparalleled ability of this nation to offer educational opportunity to
students that are widely diverse ethnically and linguistically and who are often
unevenly prepared. Specifically, international scholars have come to recognize the
pivotal role played by community colleges, an unique American invention, regarding
it as the system whose open door access and comprehensive curriculum make it
possible to extend educational opportunity to a very diverse native and immigrant
population. However, despite the fact that the educational construct for open access
is well established, warning signs of its erosion were sounded with increasing
urgency in the second half of the 20t h century and into the dawn of the of the 21s t
century, especially in those states where population and demography are far more
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
2
dense and diverse than those of the rest of the nation. Enrollment projections
that are integrally related to discussions of equity and access to higher education
project a looming crisis of major proportions which has already resulted in denial of
access to an increased cohort of school and college age students bom to the “baby
boomer” generation. This phenomenon is described in vivid terms as Tidal Wave II
borrowing on the Tidal Wave terminology used frequently to describe the
extraordinary number of individuals, primarily World War II veterans, who sought
enrollment in postsecondary institutions following World War II. Proponents of
Tidal Wave II indicate with alarm that in states such as California, access is already
denied by an educational infrastructure that has far exceeded its capacity (The
California Center for Higher Education, September 1995). Within this context, this
study focuses on Latino students as a demographic subgroup with the lowest rates of
participation in higher education exacerbated by a growing gap in educational equity
that continues to widen since 1964 (United States Office of Civil Rights, School
Civil Rights Compliance Reports. 1988-1994). This study focuses on the
achievement gap for Latino students and in particular, the role which community
colleges play as the primary point of access to postsecondary education for Latinos
nationwide.
Background of the Problem
The achievement gap for Latinos at all levels of education juxtaposed with
their significant growth in the population continues to pose serious questions not
only about the future of this particular group, but also about the future of the country
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
3
and chiefly because of their numbers in states like California. Educational
attainment for Hispanics is significantly lower for Hispanics than non-Hispanic
Whites with 57 percent and 88.4 percent respectively graduating from high school.
It is of critical importance that institutional, social and academic integration
strategies, that show promise to stem the growing educational gap for Latinos be
identified. To provide background for this study several bodies of literature
concerning Latinos in higher education were reviewed beginning with demographics
and the condition of education for Latinos in the United States and California. In
addition, the review of the literature extends to: studies of equity and access, the
community college transfer function and an overview of interactionist theories of
student success.
Demographics
Based on data collected by the Census Bureau in March 2000 Current
Population Survey (CPS) approximately one in eight people in the United States is of
Hispanic origin. In 2000, 32.9 million Latinos resided in the United States,
representing 12 percent of the total U.S. population.
This same report presents very critical demographic trends which note that
Hispanics are more likely than non-Hispanic Whites to be concentrated in key
geographic areas and to live inside central cities of metropolitan areas. In addition,
Hispanics are more likely to be unemployed at a rate of 6.8 percent compared 3.4
percent of non-Hispanic Whites. It follows that Hispanics are more likely to live in
poverty at a rate of 22.8 percent compared with 7.7 percent for non-Hispanic Whites
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
4
and 12 percent of the total population. Poverty rates are exacerbated by the fact
that 30.6 percent of Hispanic families consist of five or more people in contrast to
11.8 percent non-Hispanic Whites (U.S. Census Bureau, March 2000). In addition,
Hispanics are the most youthful population in the U.S. with 35.7 percent of the
population under 18 years of age.
Hispanics currently comprise 14.5 percent (3.6 million) of the total traditional
college age population between the ages of 18 and 24 and are projected to comprise
22 percent by the year 2020 (White House Initiative for Educational Excellence for
Hispanic Americans, 2000). Of particular importance to this study is the
juxtaposition of this telling demographic data with the continued and sustained
increased gap in academic achievement for Hispanics nationwide. Latinos drop out
of high school at the rate of 30 percent, which constitutes the highest dropout rates of
any ethnic group as compared with 12.1 percent for African Americans and 8.6
percent for white students (Gandara, 1999; NCES, 1997b).
The community colleges are the primary point of access for Latinos into
higher education not only in the state of California but also at the national level.
Fully 29 percent of all Latinos enrolled in higher education in the United States
attend California Community Colleges. And even if private institutions are included
74 percent of Latinos enrolled in a California college or university are enrolled in
community colleges (California Community Colleges, Pocket Profile 2002,
Community College League of California). It is against this background that the role
of the community colleges acquires great significance not only as the primary point
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
5
of access but also as the critical vehicle for transfer to four-year colleges and
universities.
Statement of the Problem
In 1960, the enactment of the California Master Plan for Higher Education
was hailed as a public policy masterpiece because it guaranteed access to all students
that met eligibility requirements. The Master Plan delineated the role and function
of the University of California, California State University and Community Colleges
by developing an ongoing plan for the operation of an educationally and
economically sound, vigorous, innovative and coordinated system. The Master Plan
was to be revisited every ten years but the first review did not come until 1973 and
then again in 1989 with the most recent review completed in September 2002.
However, the conditions facing higher education as we enter the 21st century
threaten the promise of the Master Plan. Enrollment is expected to grow by 20
percent, or 450,000 students, over the next six to eight years (CPEC, 1998).
Constitutional and legal constraints commit a significant portion of the state budget
to prescribed services. As a result, the state does not have the available resources
needed to meet its Educational Master Plan commitment.
Against this context the promise of educational equity is reduced for all
students and withheld altogether for Latinos, the largest growing and most under
prepared population group. A study of projected enrollment of individuals 18 to 24
in postsecondary institutions in California resulted in an almost unimaginable
outcome. Assuming natural increases in participation given the current participation
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
6
rates for whites, African-American and Latinos, equity in participation rates
would not be reached until 2062 and 2122 for these two groups respectively. This
equity status would not be reached until the white participation stabilized at 100
percent; in other words, the time when the rate of participation of 18-24 year olds in
the white population is the same as their rate of participation in postsecondary
education (Prather 1998). Latinos therefore, must wait another 120 years before they
can achieve educational equity in the state of California. This begs the question of
economic sustainability and income equity for this group.
The Purpose of the Study
The education gap for Latinos, the largest growing minority in the nation
continues to widen at alarming rates (Gandara, 1999; Lundquist, 2002). In
California, the state with the largest number of Latinos in the nation, the coming of
“Tidal Wave II” has been prognosticated by agencies including the California
Postsecondary Education Commission (CPEC), the California Center for Public
Policy and Higher Education, the RAND Corporation, and the California Citizens
Commission on Higher Education. According to CPEC, by 2003, the number of
qualified students seeking an education in community colleges, State University, the
University of California, and private colleges will exceed capacity (CPEC, 2001). It
is projected that in the next ten years 700,000 students will be unable to find a place
in public higher education, an ironic turn of events for Latinos in particular coming
at a time when outreach efforts to increase access for this group result in a drastic
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
7
reduction of capacity. A case of expectations raised and dashed at the most
critical moment (Weiner and Wolf, 2002; Lundquist, 2002).
Increasingly, educators and policy makers are turning to community colleges
as the primary point of access for Latinos and in particular at the critical role that
these institutions play in preparing students for transfer to four year colleges and
universities. The purpose of this study is to determine which set of variables
constitutes the most effective combination of factors to increase academic success
for Latino community college students in preparation for transfer to four-year
colleges and universities. Further, the study examines the impact of several set of
variables on academic success including: student background variables, college
variables and demographic variables. In particular, demographics will be examined
to assess the dynamics of Latinos in colleges where they represent the majority of the
student population in a given campus. Institutions characterized by this change in
demographics are also known as “minority/majority” colleges, a factor of great
interest to researchers who question whether gaps in student achievement would be
impacted by the new acquired majority status of students of color; or whether under
achievement is more a function of institutionalized racism and non-responsive
academic programs and policies (Conoley and Parrisher, 1978). Linked to this view
of demographics, the study will also examine the relationship of other critical
demographic factors namely, the ethnic representation of faculty and staff.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
8
Significance of the Study
The significance of the study is based on the premise that Latino students
have significantly low levels of participation in postsecondary education and
community colleges represent their single greatest point of access. Identification of
those factors that maximize academic success to make these students transfer-ready
will enhance the ability of institutions to make programmatic and policy decisions to
stem the growing gap in access to higher education for Latino students.
Research Questions
QUESTION 1: What academic, environmental and social integration
variables maximize academic success leading to transfer readiness for Latino
community college students?
QUESTION 2: What is the relation between the level of representation of
Latino students on campus to overall academic success?
QUESTION 3: What is the relation between the level of ethnic
representation of faculty and staff on campus to overall academic success of Latino
students?
Definition of Terms
Academic Success - the degree to which students consistently attain their
academic goals as measured most directly by the successful completion of course
work.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
9
Babv Boom Echo - terminology used to describe the phenomenon in the
subsequent set of increased births to the large number of “Baby Boomers” now with
children of college age.
Baby Boomer - euphemism applied to describe the children bom to the large
number of veterans returning to the United States after World War II.
Campus Climate - the sum total of interpersonal and group dynamics that
comprise the experience of participants in a collegiate setting.
Latino - because ethnic classifications vary in response to socio-political and
historical circumstances, for purposes of this study, Latino is used interchangeably
with other labels used to describe individuals from Latin-American descent and may
include such terms as: Chicano (a), Hispanic, Latin-American, Mexican-American,
Central American, South American and Spanish-American.
Mai oritv/Minoritv - a phenomenon used to describe a circumscribed
demographic unit of analysis, which is applied when the number of ethnic and
language minority individuals within that group makes up 50 percent plus one of the
entire population. A maj ority/minority campus within the context of this study is
used when the ethnic and language minority students in a given campus make up 50
percent plus one of the entire campus enrollment.
Persistence - the rate at which students complete a course successfully and
enroll in the subsequent course the following term.
Representational Value (RV) - in relationship to this study, this term stands
for the percentage of Latino students as a part of the overall college population,
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
10
namely; the degree to which these students are proportionally represented on
campus.
Student Life - the set of co-curricular activities present in a college campus as
part of student affairs including all activities that make up the climate of a campus.
Successful Course Completion - completion of a course with an A, B, or C
grade.
Tidal Wave I -imagery used by educational policy analyst in discussions of
enrollment growth projections. In particular reference to the large population “wave”
of students that hit higher education beginning in the late 1950’s following World
War II.
Tidal Wave II - imagery connected to the 1950’s phenomenon receding in
subsequent years and now expected to swell once again as a large number of 18 to 24
year old students increases in 1997 and to continues until 2010.
Transfer Directed - students annually crossing the threshold of having
completed both transfer level math and English (regardless of units earned).
Transfer Prepared - the number of students annually crossing the threshold of
having completed 56 CSU or UC transferable units (regardless of whether they have
completed transfer level math or English).
Transfer Ready - the combination of transfer directed and transfer prepared;
the number of students annually crossing the threshold of having completed 56 CSU
or UC transferable units and both transfer level math and English. A student
becomes “transfer ready” through the process of enrollment and completion of the
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
11
full array of lower division pattern of transfer courses required by the receiving
four-year colleges and universities.
Transfer - the primary mission of California Community Colleges as
established in the Master Plan for Higher Education for the provision of the first two
years of undergraduate study leading to transfer to a four-year college or university.
In general this requires the completion of 60-64 units in prescribed general education
curriculum.
Methodology
This study uses a quasi-experimental design to test the possible relationship
between the level of representation of Latinos as an underrepresented group and
academic success when analyzed in concert with other variables such as academic,
environmental and social integration. The basic premise of the study is that Latino
students have significantly low levels of participation in postsecondary education
and community colleges represent their single greatest point of access. Identification
of those factors that maximize academic success to make these students transfer
ready will enhance the ability of institutions to make programmatic and policy
decisions to stem the growing gap in access to higher education for Latino students.
Assumptions
For this study, the following assumptions are made:
1. The measures were reliable and valid indicators of the constructs to be
studied.
2. The data was accurately recorded and analyzed.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
12
3. The purposes, processes, and elements of the framework studied have a
degree of applicability and generalized ability to community colleges
throughout the country.
4. The research, data gathering, and findings and conclusions of the study
represent “good research.”
Limitations
1. This study is limited to subjects who participated in the TRUCCS study.
2. It is limited to the number of subjects surveyed and the amount of time
available to conduct this study.
3. Validity of this study is limited to the reliability of the instruments used.
Delimitations
The study confined itself to a secondary analysis of the data collected by the
TRUCCS Project (Hagedom, 1998 USC), which surveyed 5,000 students across the
nine campuses of the Los Angeles Community College District (LACCD).
The study focused on the faculty on the academic, environmental, social
integration variables and demographic factors that can be gleaned from the database
which maximize transfer for Latino community college students.
Organization of the Study
Chapter 1 of the study presents the introduction, the statement of the
problem, the purpose of the study, the questions to be answered, the significance of
the study, the definitions of terms, the assumptions, limitations, delimitations, and
the organization of the study.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Chapter 2 is a review of recent literature.
Chapter 3 presents the methodology used in the study, including a description
and rationale of the sample, the data collection procedures, a description of
instrument development and the methods of analysis of the data.
Chapter 4 presents the findings of the study
Chapter 5 summarizes the findings, draws conclusions, and makes
recommendations. References and appendixes conclude the study.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
14
CHAPTER II
REVIEW OF THE LITERATURE
Introduction
Four major topics are selected to anchor the theoretical basis of the study: (a)
the condition of Latinos in higher education; (b) the unique role and importance of
the transfer function for Latinos; (c) implications of reduced access and equity; and
(d) interactionist theories of student success.
The review of the literature begins by providing an explicit account of the
most recent statistics and outcome measures associated with educational attainment
of Latinos at the national, state, county and city levels. This section is designed to
establish the critical levels of underachievement and the growing equity gap
supporting a compelling rationale for the subsequent identification of strategies to
increase academic success for Latinos. Against this context of under-participation,
the critical role of the community colleges and in particular the transfer function is
examined as the most promising vehicle to increase access to postsecondary
education for Latinos. A review of interactionist theories of student success follows
with a special focus on the correlation of key factors including student background
variables, college variables, faculty, staff and student demographics. The review
concludes with a synthesis of applicable findings in the literature that support this
study and implications for future research.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
15
Synthesis of the Literature
Condition of Latinos in Higher Education
The literature confirms that Latinos are significantly impacted by inequities
in the educational system. They are over-represented in high poverty urban schools,
and as a result, are being educated in lower quality, more overcrowded facilities with
less access to technology (de los Santos & Rigual, 1994). They are more likely to be
taught by a teacher who is not fully credentialed, and much less likely to be taught by
someone who shares their ethnic and cultural background than their rural or
suburban counterparts (Delpit, 1995). About 75 percent of all students enrolled in
limited English proficient programs (LEP) are Hispanic, indicating the significance
of the challenge of mastering two languages in order to succeed in school. Latino
students enter school with lower levels of readiness for kindergarten than the
national average, by age nine lag behind their non-Hispanic peers in reading, math,
and science proficiency, and do not catch up through high school (NCES, The
Condition of Education, 1999; Indicators 1-2,4, 6, 18,46, Elementary and
Secondary School Civil Rights Compliance Report, 1988-1994).
The data also show that Latinos students are less than half as likely to be
enrolled in high school level intermediate Algebra than non-Hispanic White students,
8.3 percent compared with 15.1 percent and for Chemistry, 7 percent as compared to
11.8 percent. Even more significantly, schools appear to have different expectations
of students’ academic potential based on ethnicity, as demonstrated by course
placement practices in California high schools. In research conducted by the
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
16
Achievement Council in relation to a California school district, students scoring
at or above the 75t h percentile on algebra readiness tests were placed into college
preparatory algebra at significantly different rates, with 87 percent of White students
above the 75t h percentile being placed in college preparatory algebra whereas only 42
percent of Latino students in that same percentile being similarly placed (CBEDS,
1997-98, Chemistry Tables, www.cde.ca.gov/demographi/files).
A look at a few selected indicators of proficiency as related to key academic
skills further demonstrate the status of educational inequity for Latinos. In 1998,
approximately 11 percent of Latino 8t h graders were considered to be proficient
readers in comparison to 36 percent of non-Hispanic White students (National Center
for Education Statistics, NAEP, 1998 Reading Assessment, Table 5.8, National
Center for Education Statistics, NAEP, 1998 Reading Assessment, Table 5.8,
www.nces.ed.gov). The percent of 8t h grade students considered math proficient also
presented a dismal portrait of achievement with 5 percent of Latino students
achieving proficiency as compared to 28 percent for non-Hispanic White students
(National Center for Education Statistics, NAEP, 1996 Math Assessment, Table B.6,
www.nces.ed. gov). Latinos have the highest dropout rates of any ethnic group, and
leave high school at a rate of 30 percent nationally, as compared to 12.1 percent for
African-Americans and 8.6 percent for White students (Gandara, 1999, NCES,
1997b). In addition, Hispanics are still the least likely group to attend college and
least likely to attain their degrees compared to both White and Black students
(NCES, Digest of Educational Statistics, 1998k Table 207).
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
17
The Context of Education for Latinos in California
California Demographics
The County Population Estimates and Components o f Change, report issued
by the State of California, Department of Finance, Demographic Research Unit
reported that California’s population increased by just over 4 million from July 1990
to July 1999. About 61 percent of this growth can be attributed to the growth in the
Hispanic population, while the Asian population accounts for another 27 percent.
The Black, White, and Native American groups were responsible for the remaining
12 percent of the population growth (State of California, Department of Finance,
Demographic Research Unit, 2003).
This same report confirmed that the race/ethnic distribution in the state
shifted during the 1990s, with the White population’s share of the total decreasing,
while the Hispanic and Asian/Pacific Islander populations’ increased notably.
Whites were 57 percent of the population in 1990, and were 51 percent of the
population in 1999. In that same period, Hispanics increased from 26 percent in
1990 to 30 percent of the population in 1999. The Asian population increased from
9 to 11 percent and the shares of the Black and Native American populations
remained constant over the course of the decade, at 7 and 1 percent, respectively.
For purposes of this study it is important to note that the Hispanic population
increased at an average of 275,000 persons per year from 1990 to 1999. Growth in
the Hispanic population has mostly been driven by natural increase, with net
migration accounting for about 424,000 (17 percent) of the 2.5 million persons
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
18
increase over the period. The Hispanic population is estimated to have grown
in virtually all counties over the period. Los Angeles, Orange, and San Diego had
the largest increases in the Hispanic population, increasing 975,000, 241,000 and
186,000 respectively. In comparison, the White population has been nearly static
over the period, increasing by only 1 percent. This is the result of two opposing
factors. Natural increase (the difference between births and deaths) was just over
376,000, but this was largely offset by net negative migration of about 190,000.
The White growth rate was roughly only 0.1 percent per year. The total
White population grew in 44 counties over the period. San Benito and Placer had the
fastest growing White populations, growing at an annual rate of 3.5 and 3.0 percent
per year, respectively. Riverside and San Diego had the largest gains in White
population, gaining about 142,000 and 94,000 Whites each. Los Angeles County’s
White population declined by about 472,000 over the period, while Alameda, San
Francisco, and Santa Clara each lost roughly 25,000 of their White population.
Three counties—Nevada, Sierra, and Calaveras—each have more than 90 percent of
their population as White. A total of 21 counties have more than 80 percent of their
population as White. San Francisco, Los Angeles, and Imperial had the lowest
proportions, with 40,32, and 23 percent White, respectively.
In 2001, Whites constituted the majority of the population in 48 counties,
Hispanics the majority in one county. This is a change from 1990, when Whites were
a majority in 55 counties, Hispanics a majority in one county, and only two counties
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
19
had no single race/ethnic group majority. Continuation of current trends will
see an increase in the number of counties with no single race/ethnic majority.
The statistics reflected for California clearly confirm that the demographic
trends in this state reflect a future in which no one group can claim majority
representation. In Figure 1 below, the current demographic trends are projected into
the future showing how the four major ethnic groups are expected to change in a
period between 1993 and 2010. Hispanic-Latinos and Asian and Pacific Islanders
will continue to be the fastest growing racial-ethnic groups. The Asian and Pacific
Islander population is projected to increase by 4.0 percent by 2010 while the
Hispanic-Latino group is expected to increase by 7.0 percent over the same period.
Looking further ahead, the Hispanic-Latino group will comprise nearly 48 percent of
California’s population by 2040.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 0
Figure 1
Ethnic Representation of California’s Adult Population in 1993 Compared with 2010
so-r
□ 1993
African Am Latino
Racial/ Ethnic Group
Source: State of California, Department of Finance, Demographic Research Unit
Los Angeles County Demographics
As the nation braces for the educational impact of projected enrollment
growth and the critical problems of educating a student population in an increasingly
knowledge based economy, it is very useful to examine the demographic and
educational factors and characteristics of those areas where the future has already
arrived namely, Los Angeles County.
Los Angeles County is comprised of 88 cities, each with its own unique
business, lifestyle and recreational environment. Based on 2000 census data, 17
million persons live within the county’s 60-mile region. It may be said that Los
Angeles County is the place where the future is now in terms of the projected
demographic trends that will describe the rest of the nation by the middle of the 21s t
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
century. It is for this reason that Los Angeles County represents an ideal
laboratory to study and anticipate the challenges that will be facing the entire
country. The county has been characterized by ethnic transition from its early days
and continues to the present so that now 70 percent of the residents are people of
color and no single ethnic group is a majority of the population. Latinos are the
county’s largest group projected to represent 48 percent of the population by the end
of 2003. In that same year, the rest of the population will be made up of 30 percent
Whites, 9 percent African American, 13 percent Asian and less that 1 percent
American Indian (www.unitedwavla.org). State of the County Report: Los Angeles
1998-99, March 1999).
City of Los Angeles Demographics
Los Angeles is the largest of the 88 cities that comprise the County of Los
Angeles with a population of 9.5 million persons. Latinos are the largest ethnic
group in the city with 44.6 percent of the total population followed by Whites (non
Latinos) at 31.1 percent, Asians 11.9 percent, African Americans 9.8 percent,
American Indians 0.8 percent and Native Hawaiians or other Pacific Islanders at 0.3
percent (http://www.losangelesalmanac.com/topics/Population/po211ahtm 0/7/2002
8:17 p.m.).
The California Master Plan
The 1960 Master Plan for Higher Education recognized that critical to the
success of the State's tripartite system of public higher education was a coordinated
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 2
strategy responsible for ensuring the future of higher education for all
Californians. The following are among the principal features of the original Master
Plan:
• Every qualified and interested student is guaranteed access to higher
education.
• Higher education would be tuition-free for California residents. State
grants would be increased and could support students’ access to
independent institutions.
• A distinct governing board was established for the CSU as a statewide
system.
• All local community college districts were to be governed by their own
boards; district boundaries were modified to cover every geographic
region of the state.
Primary missions of the segments were delineated as follows:
• The Community Colleges (CC) would offer lower-division academic
instruction for transfer as well as general courses and vocational
education.
• The California State University (CSU) would provide instruction in
liberal arts and sciences, “applied” professions, and teacher education.
Research at the CSU would be consistent with its instructional function.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 3
• The University of California (UC) would provide undergraduate,
graduate, and professional instruction, and would have sole authority to
grant doctoral and professional degrees.
• The UC also was designated as the primary provider of state-supported
research.
• The eligibility pool of the UC was determined as the top 1/8 of high
school graduates, and of the CSU as the top 1/3 (down from the top 15
percent and 50 percent, respectively)
• All high school graduates and others over 18 years of age would be
eligible for admission to the community colleges.
• Standards were established for transfer of community college students to
the UC and CSU, and a 40:60 ratio of lower to upper division students
was established for these segments to ensure a healthy transfer function.
• New four-year campus construction was prohibited until sufficient
community college campuses could be developed.
The Master Plan committed the state to fully fund enrollment expansion. At
the same time, the confluence of eligibility and transfer recommendations had the
intended, practical effect of modifying UC and state college enrollment patterns to
redirect 50,000 students to the community colleges, thereby significantly reducing
costs to the state. The Master Plan was not just a framework of principles; it also
specifically planned for the location and type of campuses to be built in the coming
years.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 4
The Legislature undertook another review of the Master Plan in the
1970’s and 1980’s. In 1973 the legislative review looking into the Master Plan
recommended the creation of an oversight commission, the California Postsecondary
Education Commission (CPEQ).
The primary statutory purposes of the CPEC are to:
• Develop an ongoing statewide plan for the operation of an educationally and
economically sound, vigorous, innovative and coordinated system of
postsecondary education;
• Identify and recommend policies to meet the educational, research and public
service needs of the State of California; and
• Advise the Governor and Legislature on policy and budget priorities that best
preserve broad access to high quality postsecondary education opportunities
(Center for Studies in Higher Education, 2002).
The 1989 report also led to a significantly strengthened transfer system to
assure successful community college students a place in the university systems.
Numerous reforms of community college governance and function emerged in the
two years subsequent to recommendations on those issues. Additionally, as a result
of the committee’s work, the missions of all three systems were modified to include
public service. The 1989 report recommended a significant increase in student
financial aid; this began to be realized only after the recession of the early 1990’s
ended.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 5
In 2001 the Legislature undertook its most recent review of the Master
Plan. There are several characteristics of this last review that are worth noting. The
title of the plan itself has changed from, The California Master Plan for Higher
Education to The California Education Master Plan. This new title represents a
change of philosophy by policy makers as they have chosen to expand the context of
the review to span the entire educational pipeline K-University. This approach
made it possible to ensure consonance between and among the various
recommendations that emerged.
The final report contained 56 recommendations and a series of sub
recommendations on the future of education in California. Those recommendations
most pertinent to this study are listed below:
Recommendation 11.4 urges adult schools and noncredit college personnel to
collaborate to strengthen articulation of adult education with community
college coursework and to articulate career technical courses at either level
with postsecondary coursework.
Recommendation 12 continues language currently in the Master Plan for
Higher Education that UC will admit students in the top 1/8 , and CSU will
admit students in the top 1/3, of the high school graduating class; community
colleges will continue to be open access institutions ‘for all high school
graduates and adults who can benefit from postsecondary instruction.”
Recommendation 20.3 states that colleges should be encouraged to develop
end-of-course assessments to measure student mastery at each grade or
course level and readiness for learning at the next grade/course/level. It also
recommends that assessment of 11t h graders should be aligned with entrance
or placement examinations of the State’s college and university systems.
Recommendation 34 would authorize community colleges to provide
instruction at the upper-division level jointly with CSU and UC or a Western
Association of Schools and Colleges (WASC) accredited independent or
private postsecondary education institution.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 6
Recommendation 50 calls for determination and definition of how the
costs of postsecondary education should be distributed among the state, the
federal government, and students and families. Then it recommends that a
new “fiscally” responsible and appropriately balanced student fee policy be
introduced that would preserve access to higher education opportunity for all
of California’s students, particularly those from low-income and
underrepresented groups.”
Recommendation 51 recommends that the state “should adopt a student fee
policy which stabilizes students fees, so that to the extent feasible, fees would
increase in a moderate and predictable fashion when needed, and resist the
pressure to buy out student fee increases or reduce fees at CCCs, CSU, and
UC during strong economic times (The California Master Plan for Education,
2002).
A brief analysis of the policy perspective resulting from the
recommendations in the new Education Master Plan include the focus on; a seamless
educational system, the importance of career education and the needs of adult
learners, student outcome measures, ongoing assessment including exit exams,
reaffirmation of the eligibility criteria for the tri-partite postsecondary education
segment and the looming certainty of increased shared for the cost of higher
education for students and their families with a concomitant decrease of resources
from the state. It is important to note that the unfortunate coincidence in timing of
this most recent review with one of the worst periods in the California’s economic
history seems to have unduly constrained the visionary aspects ascribed to the
original Master Plan language and may have produced instead a regression to the
means in terms of policy constructs that do not consider the scope of demographic,
economic and societal changes facing the state. It is for this reason that it is unclear
whether this latest iteration of the Master Plan will be a strategic force in developing
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 7
a new educational direction that will prepare the youth and working citizens of
California for their place in the highly technologically driven future economy; or the
unfortunate marker signaling the decline of what was once perceived as one of the
most enlightened and progressive systems of public education.
Educational Access for Latinos in California
In California access to postsecondary education continues to be significantly
lower for Latinos than for non-Hispanic White students. One of the most significant
obstacles to access for Latinos is the low completion rates of the prescribed high
school, college preparatory course pattern also known as the “A-F” curriculum Since
1999, completion rates for both student groups has increased however at 41 percent
the rate is nearly double for non-Hispanic Whites than it is for Latinos at 22 percent
(CPEC, Student Profiles, 2000,1-5 & 1-6). It is important to note that since this
study, the faculty of the University of California redefined policies regarding the list
of approved courses for entrance to the university and by adding a “college
preparatory elective” or “g” category to the list. The new pattern is now known as
“A-F” as follows:
(a) History and Social Science
(b)
English
( c )
Mathematics
(d)
Laboratory Science
(e )
Language other than English
(f)
Visual and Performing Arts
(g)
College Preparatory Electives (University of California, Office of the
President, 2002)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 8
Los Angeles County: K-12 Education Enrollment and Graduation Statistics
Within the County of Los Angeles, 81 school districts serve nearly 1.7
million students at more than 1,700 school sites. The districts range in size from 87
students in Gorman Elementary to nearly 720,000 students in the Los Angeles
Unified School District (LAUSD). At the postsecondary education level the county
is home to 155 colleges and universities of which 27 are public institutions
(University of California (UC), California State University (UC)) and 128 are
private, independent colleges and universities.
Based on the most recent statistics from the Los Angeles County Office of
Education, for the 2000-2001 academic year, LAUSD the largest of the districts in
the county graduated a total of 316,124 students and of these only 112,469 or 35.6
percent met the requirement to enter either the UC or the CSU. In addition, Figure 2
below illustrates the difference in eligibility rates among ethnic group clearly noting
that Latinos share the very bottom of that scale with Native Americans. In sheer
numbers however, the picture is far more striking because of the 103,795 Latino
graduates in 2000-2001 only 22,772 or 22.9 percent were UC or CSU eligible
leaving 80,023 graduates with community college as their only option for the pursuit
of postsecondary education.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 9
Figure 2
Los Angeles Unified School District Graduates Eligible for UC and CSU
Percentage Eligible Number of LAUSD 12th Graduates Completing all Courses Required for UC or CSU
of Total Graduates Entrance by Ethnic Group, 2000-01
Per Group
70.00%
60.00%
50.00%
40.00%
30.00%
20.00%
10.00%
0.00%
American Asian Pacific Filipino Hispanic African White TOTAL
Indian or Islander or Latino American (Non
Alaskan Hispanic)
Native
Ethnicity
Source: California Department of Education, Educational Demographic Unit
Postsecondary Education Opportunities in the County of Los Angeles
According to the California Postsecondary Education Commission, there are
155 institutions of postsecondary education in the County of Los Angeles. These
institutions include:
• Two- and Four-Year Specialized Schools in the Arts and Sciences
• Traditional Liberal Arts Colleges
• Comprehensive Universities
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 0
• Major Research Universities
• Free-Standing Graduate and Professional Schools
• Campuses for Working Adults
Of particular interest to the study of transfer opportunities of urban community
colleges are the 27 institutions in the area that are publicly funded including the
University of California at Los Angeles and five California State University
campuses in addition to 21 community colleges. As of fall 2000 the combined
enrollment for these colleges and universities totaled 491,688 students with 36,890,
105,849 and 348,929 for the UC, CSU and CC respectively.
In Figure 3, enrollment by ethnicity illustrates that despite the size of the
California public postsecondary education system, access and equity continues to be
a problem for Latino students. In addition, since the mid 1990s, when the UC
Regents voted to end any consideration of race, ethnicity or gender in the
university’s admission decisions, many policymakers have expressed concerns over
a precipitous drop in student diversity at many UC campuses. In 2001 the Regents
rescinded their controversial ban, but the prohibition continues under Proposition
209 approved by the voters in 1996 which forbids discrimination or preferential
treatment based on race, ethnicity or gender in public education, hiring or
contracting.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31
Figure 3
Enrollment hv Ethnicity in California Four Year Colleges and Universities
□ N ative
Am erican
□ Filipino
■ Black
□ A sian
■ Latino
□ W hite
Source: CPEC 2000
The Transfer Function and the Unique Role of Community Colleges
Given the disproportionately low number of Hispanics that are eligible for the
University of California (UC) and California State University (CSU) systems it
becomes patently clear that the greatest single source for continued opportunity is the
California Community Colleges. In fact, California Community Colleges represent a
pivotal point of access not only at the state but also at national levels. Fully 29
percent of all Latinos enrolled in higher education in the United Sates attend and
Enrollment by Ethnic Groups in the Fall of 2000
i'ir 1 0 0 %
California State
University
University of
California
Independent
Colleges and
Universities
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 2
California Community College. And even if private institutions are included 74
percent of Latinos enrolled in a California college or university are enrolled in
community colleges (California Community Colleges, Pocket Profile 2002,
Community College League of California).
Transfer to four-year colleges and universities and attainment of the
baccalaureate and graduate degrees becomes possible for most Latinos only through
the vehicle of community colleges. The tables below present information on transfer
trends in the state and confirm low number of transfers to UC and CSU by ethnicity
in addition to the very low incremental gains made from year to year.
Table 1
California Community College Full Year Transfers to UC. bv Ethnicity
Ethnicity
Academic Year
1990-
1991
1991-
1992
1992-
1993
1993-
1994
1994-
1995
1995-
1996
1996-
1997
1997-
1998
1998-
1999
Native
American
126 139 121 107 129 137 124 102 97
Ethnicity
Academic Year
1990-
1991
1991-
1992
1992-
1993
1993-
1994
1994-
1995
1995-
1996
1996-
1997
1997-
1998
1998-
1999
Filipino 203 198 229 291 306 310 333 340 298
Black 272 288 274 306 364 386 318 293 228
Latino 1054 1175 1205 1335 1452 1503 1430 1300 1302
Asian
Pacific
Islander
1553 1563 1721 2287 2610 2767 2863 2806 2377
White 6318 5984 5751 5927 5614 4888 4664 4487 4000
Source: CPEC Student Profiles 2000
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
33
Table 2
California Community College Full Year Transfers to CSU. b y Ethnicity
Ethnicity
Academic Year
1990-
1991
1991-
1992
1992-
1993
1993-
1994
1994-
1995
1995-
1996
1996-
1997
1997-
1998
1998-
1999
Native
American
556 541 500 554 539 641 604 565 475
Filipino 1297 1245 1204 1431 1739 1840 1867 1626 1711
Black 2567 2480 2241 2441 2564 2836 2799 2442 2105
Latino 5694 5934 5780 6780 7437 8334 8661 8464 8201
Asian
Pacific
Islander
4552 4470 4416 5427 6212 6562 6741 6346 6230
White 26446 24480 21061 21068 21402 20931 19623 18341 18375
Source: CPEC Stuc ent Profiles 200()
Because California Community Colleges represent the most such a critical avenue
for educational access for Latinos, there has been much criticism and focus on data
which from a simplistic perspective do not yield the number of transfer students in
proportion to enrollment. There are many factors that must be understood in order to
interpret transfer data appropriately primary among these is the understanding that
transfer, while being a primary mission of community colleges, it is not the only
mission. California Community Colleges, as established in the Master Plan for
Education, are also charged with providing career education, adult education,
continuing education and economic development programs.
The California community colleges have long recognized that the
complexities and multiple variables associated with transfer make it difficult to
establish an official transfer rate, it is for this reason that an official transfer rate has
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 4
not been identified for the system. Crude and oversimplified measures of
transfer have been established by outside entities as a means to speak about transfer
patterns. Of particular concern is the use of non-related counts to generate a ratio of
current year transfers to current year of enrollment, often resulting in “rates” that are
in an unfavorable 3 to 5 percent range for the system. The use of such ratio implies
that all transfers come from a single-year cohort of students whose expectation is to
complete their entire transfer curriculum in one year; this goal is entirely
unattainable. The current transfers to current enrollment ratio also implies that all
students enrolled in the California Community Colleges have establish, transfer is
only one of the missions and therefore only one of the reasons students enroll in
community colleges.
In order to address the increased pressure for a valid measure of transfer that
would consider as many of the variable which make this measurement complex, the
State Chancellor’s Office established the Transfer Technical Data Workgroup
(TDTW) to define possible transfer rate scenarios that could serve as the “official”
transfer rate of the system. The group began its work in 2001 and presented
recommendations for the adoption of a set of factors that combined to produce the
best possible transfer rate data. The focal point for the discussion to establish a
methodology recognized that using a rate denominator that is too inclusive (all
students) negatively impacts the rate by including students without intent to transfer.
On the other hand, creating a denominator that is too exclusive (for instance,
counting only students who complete 56 lower division units and transfer level
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35
English and math) would produce greatly inflated rates by excluding students
who have intent to transfer but who do not succeed. The definition of “intent to
transfer” must balance between these elements to provide a reasonable estimate of
the percentage of students in the CCC system that can be called “transfer students.”
In addition the TDTW considered other factors to set the context for their
work as follows:
Counting Methodology and Source of Transfer Data. Given the
methodological ease of which to define a cohort and the superiority of the
transfer data, TDTW decided that longitudinal tracking using electronic data
matching was the preferred method of producing rates.
Cohort Tracking Period. The tracking period was determined to be six years
because historical studies show that the vast majority of transfers occur
within this time period. Although transfers do occur after the sixth year, this
number is small.
Defining “Intent to Transfer. ” Once a cohort methodology was decided
upon, the main focus of the group was to narrow the cohort from all students
to only those who had “intent to transfer.”
Given this context, three basic principles were established by the TDTW as
context for their recommendation as follows: (1) only the universe of students who
have the “intent to transfer” would be included (2) despite the complexity of factors
the rate would have to be comprehensible to lay audiences (3) and recognition that
there is no perfect measure of transfer rate.
Several different scenarios were tested by the TDTW, which resulted in the
identification of the best combination of factors to yield the most valid measure of
transfer. These scenarios are defined below:
Scenario 1 - defining “Intent to Transfer” using students stated intent. In this
scenario, what a student states as his educational goal is used to define the
denominator.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
Scenario 2 - defining “Intent to Transfer” using behavioral patterns. In this
scenario, a student’s course taking patterns were analyzed as including those
students who had attempted transfer level math and/or English courses.
Additionally, levels of units earned (3,6,12, 24) were examined separately
and in combination.
Scenario 3 - defining “Intent to Transfer” using a combination of both. In
this scenario, a combination of goal and behavior was used; however, results
were more closely aligned to those using goal alone.
After testing each of these scenarios and the combination of variables within
each scenario, the TDTW recommended the adoption of scenario three; namely, the
combination of a behavioral rate established by math and English course taking
patterns and a threshold of 12 units of completed units of credit. Several reasons
were cited in support of this particular methodology; it captured over 80% of all
students who actually transferred, it yielded a 34. 6 percent estimated statewide
transfer rate and it allowed for a “value added” factor by including students that had
completed at least 12 units of college credit thus eliminating remedial students from
the mix. A summary of the proposed transfer rate methodology is illustrated in the
table 3.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
37
Table 3
Summary and Comparison of Statewide Transfer Rates for All Students vs. Students
with “Transfer Intent”
Methodology
Cohort
Size
Percent
Who Show
“Intent to
Transfer”
Number o f
Actual
Transfers
in Six
Years
Percent o f
Transfers
Accounted by
the
Methodology
Transfer
Rate
All First-Time
Students, Fall 1994
269,353 39,227 14.6%
First Time Students, Fall
1994 who: Attempted
transfer level math or
English; and Completed at
least 12 units at any CCC
93,310 34.6% 31,447 80.2% 33.7%
Source: California Community Colleges Chancellor’s Office MIS Unit (March 2002,
Study Session on Transfer Measurement and Capacity of Receiving Institutions)
Sacramento, California.
The examination of these scenarios by the TDTW resulted in the selection of
the most inclusive and valid measurement of transfer rate, namely the identification
of that cohort of students defined by a combination of behavioral course taking
characteristics (English and math) and the completion of twelve units of credit. This
methodology captures over 80% of all students who actually transferred and the
number of credit units completed addressed the “value added” impact of community
college instruction by eliminating all remedial students.
Using the proposed transfer rate methodology, Table 4 below depicts the
transfer rates of the system by gender and ethnicity for the fall of 1994, consistent
with the six-year period established for the transfer of these students in Fall 2000.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
38
Table 4
Transfer Rates in the Year 2000 bv Gender and Ethnicity Fall 1994 Cohort of
Students with “Transfer Intent”
PERCENTAGE
GENDER TRANFER RATE
Male 33.5
Female 36.2
ETHNICITY
Asian/Filipino/
Pacific Islander
43.1
Hispanic 25.5
Black 23.5
Native American 26.3
White 37.2
Unknown 33.0
Source: CCC Chancellor’s Office MIS Unit
Transfer Preparedness: Another Measure of Transfer
In order to identify policies that can maximize the number of Latino transfers
to four year colleges and universities it is important to focus on those policies,
programs and services that are under the control of community colleges and
conversely those that are outside their purview and may only be impacted by four-
year colleges and universities. Simply stated while college policies and actions may
assist in the transfer function, colleges do not have full control over the transfer
function; ultimately the students themselves decide if, where, and when they are
going to transfer. Policies and actions of receiving institutions also affect transfer
including such variables course offerings, impacted programs, course scheduling and
socio-cultural pressures.
It is for these reasons that the CCC as a system has endeavored to collect
historical counts of three measures that accommodate for this external variables by
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 9
measuring student cohorts using the agreed upon six-year timetable from the
first time of enrollment. These three measures include:
Transfer Directed - students annually crossing the threshold of having
completed both transfer level math and English (regardless of units earned);
Transfer Prepared - the number of students annually crossing the threshold of
having completed 56 CSU or UC transferable units (regardless of whether
they have completed transfer level math or English); and
Transfer Ready - the combination of transfer directed and transfer prepared;
the number of students annually crossing the threshold of having completed
56 CSU or UC transferable units and both transfer level math and English.
The table below displays the data gathered for the period between 1999-1998
and 2000-2001.
Table 5
Annual Counts of Transfer Directed. Transfer Prepared, and Transfer Ready
1997-2001
Year
Transfer Directed
(completed
transfer English
and math)
Transfer Prepared
(completed 56
UC/CSU
transferable units)
Transfer Ready
(completed 56
UC/CSU units and
English and math)
1997-1998* 92,673 106,951 35,539
1998-1999 104,538 107,980 40,369
1999-2000 86,565 96,501 31,117
2000-2001 108,782 116,114 42,745
*Note: This cohort includes students who enrolled for the first time in the 1991-
1992 academic year to allow for the agreed upon six-year time frame for
transfer.
Source: California Community Colleges Chancellor’s Office MIS Unit (March 2002,
Study Session on Transfer Measurement and Capacity of Receiving Institutions)
Sacramento, California.
It is a well known phenomenon that there is a drop at three levels of the
transfer pipeline beginning with the number of students that complete all transfer
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 0
requirements at the community college, followed by a decline in those that
actually are admitted to four year colleges and universities and finally another
decline reported in the number that actually enroll as sophomores.
Equity and Access: The Failed Promise for Latino Students
Educational policy analysis as a process for review and study is based on
certain tenets and structures that are well taught in the academy. Invariably the
process requires a look at multiple sources, evaluation of multiple measures, the
identification of trends over time and balanced review of both sides of a given issue.
The culmination of this process is also vetted in a prescribed structure which
includes the identification of various alternatives and their attendant evaluation
including cost benefit analysis, and measures of impact which ultimately lead to a set
of weighed recommendations for future policy consideration. However, a review of
the literature focusing on student achievement, diversity, equity and access since the
passage of the Civil Rights Act of 1964, is so fraught with discontinuity, disparity
and ill logic on the part of policy makers that dispassionate analysis is nearly
impossible. The only conclusion that may be reached is that there is no real premium
placed on student equity outcomes and in fact the roots of institutional racism and
social disparity run deeper than any of us can fathom. These statements are not the
alarmist extreme of liberal ideology but rather the unavoidable conclusion resulting
from the review of multiple studies, vast sources of data and a chronology of neglect
that in the case of Latinos clearly demonstrate that educational policy, programs and
interventions continue to widen the achievement gap, deter from student success and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
perpetrate fraud on students who are led to believe that they have equity of
access and are nevertheless denied it; even after they have fulfilled their side of the
bargain.
Enrollment growth, enrollment projections and institutional capacity are key
to the discussion of equity and access to postsecondary education. Indeed there is a
direct correlation to enrollment growth and decline across all disciplines that can be
directly attributed to the need for enrollment and fluctuating economic patterns. In
reality, pure demand for higher education is always shaped, constrained and
dampened by other forces both short-term and longer. According to George Prather,
a researcher with the Los Angeles Community College District, the variation of year-
to-year participation around the regression line, once all data errors are removed, is
determined by the short-term forces. Chief among these are the economic cycle and
institutional policy choices (Prather, 1998).
In other words when colleges need students and the economy provides the
necessary resources to support public postsecondary institutions, obstacles are
minimized, admission standards are relaxed and special student support services are
increased. The rhetoric that accompanies these periods abounds in references to
equity, profiles of underrepresented student populations and support for student
success initiatives. The converse happens when there is retrenchment and
institutions find themselves impacted. Under these circumstances, admission
standards are revisited, pre-requisites proliferate, outreach and support services are
diminished or eliminated and policies resurface questioning prior enrollment demand
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 2
studies and the value of equity arguments with affirmative action being chief
among these.
The vicious cycle of poverty which is fueled by this pattern of decreased
access has been an important topic of discussion for the Educational Testing Service
which estimates that if Hispanics and African-Americans had the same education and
commensurate earnings as whites, there would be an upsurge in national wealth of
$113 billion annual for African-Americans and $118 billion for Hispanics.
(Educational Testing Service 2002). This estimate is directly supported in the
observations made by Frank Levy, professor of Economics at the Massachusetts
Institute of Technology who theorizes that a widening gap is evolving fueled by the
concurrence of two factors which may create an irrevocable set of circumstances
never before experienced in this country; namely, a twenty-year long period of slow
wage growth followed by a falling growth rate of per capita income, and a
continuing trend in which higher and lower income families have become
increasingly concentrated in separate geographic areas (January 1995). As a result of
this set of this set of factors, locally run schools become increasingly stratified, so
that poor and working class children cannot acquire the necessary education also
described by economists as “human capital” thus perpetuating and widening the
education/earnings gap. This phenomenon is also described by Levy as “the
college/high school earnings gap.” Levy continues to warn of an impending crisis
facing a nation that counts on mass upward mobility but continues to yield slow
growth in education and in earnings for its middle class. A projection which is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 3
further anchored in Levy’s intergeneration analysis of earnings focuses on a
time span over two decades which demonstrates that it is not possible to maintain
even a
static economic position without increasing one’s educational level over that of the
parents.
Experts in the social sciences also support Levy noting that as late as 1965
central cities had significant numbers of middle class families and today more than
one third of inner city residents are poor and contain nearly 45 percent of all poor
children in this country. This neighborhood-sorting phenomenon is particularly
pronounced for Black and Hispanic families resulting in significant educational and
subsequent spiraling economic inequality (Jargowski, 1994).
Campus Climate and the Maj ority/Minority Phenomenon
The literature related to student equity centers on the recognition of the
importance of educating students from all walks of life. Increasing diversity is a fact
of American life and likewise of college life. And in the most populous areas of the
country, persons of color make up the majority of the population, a phenomenon also
known as a “majority/minority.” However, despite these demographics, even with
the increase in access to colleges and universities students of color and in particular,
Latinos and Blacks continue to lag significantly behind white students. While
academicians search for numerous discrete factors and combinations of factors to
understand this state of affairs few focus on one potentially important and rather
sensitive factor; namely, the social climate on campus.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 4
Today as in the past, racism in the United States is systemic and, thus, a
complex social reality with several major dimensions: (1) an array of discriminatory
practices; (2) privileges and resources that accrue to whites from institutionalized
discrimination over many generations; (3) racist ideologies, prejudices, and emotions
that defend these unjustly gained privileges; and (4) an assortment of social
institutions that generally embed and reproduce racial inequalities. System racism is
made up of everyday patterns of stereotyping and discrimination that are webbed
throughout most major institutions (American Council on Education (ACE, 2001).
Surveys at many campuses reveal that, generally speaking students and
faculty of color face a different campus world from that of Whites. Racist incidents
are regularly reported at U.S. colleges and universities. Reports of such actions come
from campuses nationwide including: the University of California, Los Angeles;
University of Michigan; California Lutheran University; University of Colorado;
Denver; Harvard University; Yale Law School; and, University of Texas, Austin to
name but a few (Allen, Solorzano et. al 2000).
Several strategies to improve campus climate have been proposed including:
recruitment of faculty and staff of color, recruitment of more students of color,
expansion of mentoring programs and developing family, school and community
partnerships (ACE, 2001). A study conducted by Spradley (1996) with adult African
American male graduates from an urban commuter baccalaureate institution
confirmed the importance of peer support on campus through study groups and
classroom interactions as critical to academic success. The importance of emotional
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 5
peer support is also documented by other researchers (Quinnan, 1997). In
addition, faculty members who are accessible, concerned with students and
committed to quality instruction in the classroom environment play a significant role
in learning and persistence (Spradley, 1996). Faculty discussions that expand
student’s understanding of the content by placing information within a relevant
content of their lives increases motivation and persistence (Vella, 1994).
Those whites with little meaningful contact with people of color will often be
fearful or suspicious of them. Social isolation is serious because it means
that whites, including those within decision making power over society
usually do not possess the experience and knowledge necessary to understand
the hard realities of racial discrimination and racial inequality that face
African Americans and other Americans of color (Bobo, 2001).
In California, concerns about the impact of campus climate on educational
equity drew the attention of the State Legislature with the passage AB 4071, the
“Higher Education Equity Assessment Act of 1988 (Chapter 690 of the Statutes of
1988). The Act directed CPEC to determine the feasibility of developing an
institutional assessment tool that could measure “campus climate” within the public
postsecondary institutions of higher education in the state. Campus climate was
described as the combination of the interpersonal group dynamics that comprise the
experience of participants in collegiate settings. Specifically, CPEC was directed to
develop a survey instrument, link findings from programs and existing admissions
and retention to understand the levels of attrition and to develop comparable data to
establish profiles for the UC, CSU and CCCs (CPEC, 1992, Report 92:2).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
46
CPEC convened a panel of intersegmental experts to aid in the
development of the survey and tested the elements through the systematic use of
focus group interviews conducted statewide. Several factors emerged as most salient
and were identified as critical areas for assessment to determine campus climate
within any given institutions as follows:
• Faculty composition and philosophy
• Student-faculty interaction
• Curriculum content and pedagogical approaches
• Academic support service availability
• Student life
• Interactions among students
• Campus image
• Student expectation of the campus prior to enrollment
• Campus leaders’ philosophy and implementing practices; and,
• Campus-local community interaction.
The final outcome of the work undertaken by CPEC was the production of a
“Resource Guide for Assessing Campus Climate” which was published and widely
disseminated to colleges and universities in the state in the Fall of 1992.
Accompanying the guide were a list of institutional expected benefits. Specifically
CPEC theorized that a well-designed and implemented examination of the
perceptions of individuals within a campus environment should reap significant
institutional benefits. Among them:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 7
1. The opportunity to discuss often unacknowledged and unrecognized
issues. And the means to engage faculty, staff and administrators as well
as students in a meaningful discussion.
2. A mechanism for shifting discussions from idiosyncratic instances to
collective appraisals of institutional life. In order to differentiate between
singular isolated, and transitory incidents and perceptions and common,
recurrent, and consistent perceptions that may be wide spread throughout
the fabric of the institution.
3. The potential to enhance institutional effectiveness and efficiency in
terms of such factors as students flow measures as increased retention and
graduation rates.
4. The identification of strengths and weaknesses such that institution
decision makers may:
• Identify those perceptions that can be institutional affected;
• Determine which institutional program, policies, and practices are
enhancing or delaying the campus’ achievement of its goals -
particularly that of educational equity; and,
• Prioritize the actions that the institutions should initiate and continue
in order to meet those goals.
5. Through multi-year cyclical assessments, the opportunity to build
longitudinal information on the effectiveness of specific, planned
interventions designed to achieve institutional goals , as well as, the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
48
identification of unplanned institutional changes that may affect the
attainment of those goals.
6. The opportunity to anticipate and consider the implications of rapidly
changing demographics in program planning and evaluation.
7. The opportunity to pre-empt major embarrassing crisis on campus before
they occur.
8. Enhance the capacity and skills of all institutional members to participate
in the increasingly multicultural, complex world that exists beyond the
campus.
Interactionist Theories of Student Success
Fishbein and Ajzen laid the foundational work, which has served as the basis
for subsequent theoretical variations of student interactionist theories of success.
Henry and Smith who made a set of critical observations made one of the most
succinct descriptions of the central hypothesis of interactionist theories.
When both academic and environmental variables are favorable, students
should persist. When both variables are unfavorable, students are likely to
drop out. When academic variables are positive but environmental variables
are negative, the favorable effects of academic variables on student goal
attainment are suppressed or attenuated. ..Students may drop out of college
despite strong academic performance if they perceive low levels of utility,
satisfaction, or goal commitment, or if they experience high levels of stress
(1993 p. 29).
Since that time, variations of the model designed to test for different outcomes have
been developed by a host of other researchers (Ajzen & Fishbein, 1975; Spady,
1970; Tinto, 1975; Pascarella, 1980; Bean and Metzner, 1985; Madden & Ajzen,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 9
1992; Bagozzi, 1992; Cabrera, Nora, and Castaneda, 1992, Vallerand, Deshaies,
Cuerrir, Pelletier & Mongeau, 1992; Becker and Gibson 1999; Hagedom, 2002). For
the purpose of this study the Fishbein Ajzen basic construct serves once again as the
basis for another revision that incorporates the work of Benjamin (1994) in order to
establish a construct for student life. More specifically this revised model focuses on
community college urban campuses with high levels of student diversity and large
numbers of non-native speakers of English (Hagedom and Castro, 2000). This
revised model is also known as the Community College Model for Student Life and
Course Completion (Hagedom, 2002). This model seems to include the most
comprehensive set of student interaction factors and particularly lends itself to this
study because of its inclusion of non-traditional student variables including part-time
students, students with outside commitments and working students. In addition, the
model includes variables such as finances, employment, family responsibilities, and
opportunity to transfer; along with social integration, academic variables and other
background characteristics.
Conclusions
The review of the literature supports the importance of the study by
confirming the critical juncture, which describes the serious condition of inequity for
Latinos. Both quantitative and qualitative research documents clearly establish the
fact that by any measure, demographic, economic, educational and socio-cultural,
Latinos face insurmountable odds. Studies contradict the premise that educational
gains have been made by Latinos at par with that of other ethnic groups by noting in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
fact, that the opposite has happened and while all groups have made gains since
the passage of the Civil Rights Act of 1964, educational gaps have widened for
Latinos. Other critical factors juxtaposed in the review describe the current
downturn in the economy and the lack of physical capacity to accommodate
enrollment growth; all within the context of a growing chilling climate which de-
emphasizes the importance of outreach and has reversed in policy and in action the
tenets of “affirmative action” efforts designed to level the playing field for all
minorities. Findings also emphasize the unparalleled growth of Latinos nationwide
and in places like California in particular with Los Angeles County as the place
where the demographic future of the nation has arrived and the failure to prepare
Latinos is a reality. Inevitably, the review leads to the conclusion that serious
intervention is needed and any and all efforts to maximize access must be found.
Community colleges and the transfer function emerge as a significant vehicle
and institution of choice for many Latinos, the majority of which graduate from high
school, but have not met the UC and CSU requirements for entrance. In this vein the
review places special focus on the transfer function of community colleges and
details at length the complexities associated with a valid measure of transfer given
the diverse and non-traditional nature of community college students, the multiple
missions of the colleges and the parameters which ultimately circumscribe the ability
of two-year institutions to transfer students because of policies that are controlled by
four-year college and universities. Findings indicate that enrollment opportunities
and patterns vary widely and not primarily depending on the academic ability of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
students but more so on variables associated with capacity, resources and
impaction of programs.
Examination of the body of literature that focuses on campus climate and the
sensitive issue of “institutional racism” concludes that the interpersonal group
dynamics of an institution play a powerful role in the persistence of students of color.
The importance of this particular finding is magnified by the realization that few
Latinos survive the educational pipeline and the thus the gravity of further attrition
due; not to student capabilities, but rather to a campus climate that is hostile and
uninviting is truly catastrophic.
Finally, the review concludes with the confirmation of theoretical models that
support the methodology and research design for this study. Specifically, findings
support the use of interactionist theories and models that establish the critical
interplay between both academic and environmental factors. Further, the literature
confirms the evolution of such models by numerous researchers concluding with the
recent development of the Community College Model for Student Life and Course
Completion that most closely fits the diverse and non-traditional nature of
community college students.
Implications
Policy Implications
It is clear that this nation may be past the proverbial crossroads regarding the
future of education for Latinos. Since the 1980s to the present more than two
decades of data, reports, warnings and even social unrest have gone unheeded.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
52
Demographic profiles, economic trends and the spiraling trend of poverty
associated with Latinos find this nation facing what may the establishment of a
permanent underclass. The assuaging assumptions that affirmative action to
diminish inequity have worked, are now proven wrong by any outcome measure. It
is baffling that nearly forty years after the passage of the Civil Rights Act of 1964,
Latinos find themselves much worse off.
In California, the downturn from the crossroads is even more precipitous as
community colleges, the primary access to postsecondary education for Latinos face
the worst funding year in its entire history; this despite the fact that they are already
the segment with the lowest funding per student.
Urgent and radical alternatives must be identified and multiple solutions
sought to stem the downward spiral clearly established by researchers, social
scientists and educators. Implications are significant not only for Latinos but for an
entire nation which historically depends on its ability to support an upwardly mobile
middle class; made all the more urgent by the significant demographic growth of this
group.
California policy makers and legislators must recognize that the recent
review of the California Master Plan for Education has abridged the promise of
access that made this state an international economic power. The Plan contains no
visionary policies nor does it acknowledge the seriousness of the situation for
Latinos. It is the latest example of the dynamics of denial currently in place.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
53
It is because of the absence of leadership at the macro levels of policy
and planning that the only hope may be regional and local partnerships and solutions.
Community Colleges must continue to struggle and further emerge as a critical
bridge to partner with K-12 and four-year colleges and universities. Any and all
avenues to maximize and strengthen proven practices, along with the identification
of untried and radical alternatives, must be identified.
Recommendations for Future Research
Based on the findings of this literature review, it is clear that more research
linking economic and academic variables is necessary. If “human capital” is the
outcome of educational access and the engine that mobilizes the nation; then
research strategies that focus on ways to link these two areas of study and policy
formation must be found.
Primary language and the prominence of non-native English speakers cannot
be ignored and further examination that goes beyond the context of bilingual
education must be undertaken. Further, research that establishes the ubiquitous
nature of non-traditional students must be embraced in order to recognize that the
“traditional” student no longer exists and neither should traditional forms of
education.
Finally, campus climate and the courageous recognition of institutional
racism should be pursued in systematic research projects that allow for the
identification of discrete factors that may allow for the dismantling of unconscious
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5 4
yet self-sustaining institutional and bureaucratic processes that in concert act to
make students feel unwelcome and ostracized.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
55
CHAPTER III
METHODOLOGY
Introduction
The purpose of the study is to determine which set of variables constitutes
the most effective combination of factors to increase academic success for Latino
community college students in preparation for transfer to four-year colleges and
universities. Further, the study examines the impact of several sets of academic
success variables including; student background variables, college variables, and
demographic variables. In particular, demographics are examined to assess the
performance of Latinos in colleges where they represent the majority of the student
population to ascertain the degree to which the dynamics of campus climate impact
academic success.
Research Question
The conceptual framework for the research questions in this study emerged
from a review of the literature of the condition of education for Latino students,
which is replete with data regarding the educational gap between this group and all
other demographic population subgroups. These reports address two major areas of
study (1) identification of specific factors that impact academic success both
positively and negatively and the confluence of environmental that combine with
academic variables to enhance or impede student performance, and; (2) specific
factors associated with campus climate defined as a collage of the interpersonal and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
56
group dynamics that comprise the experience of participants in the college
setting (CPEC, 1992). While the literature, which supports the first construct, does
include factors associated with interpersonal relations, for the purpose of this study,
given the distinct ethnic and linguistic characteristics associated with Latino
students, it was important to add another context more directly linked to the issue of
ethnicity. For this reason the “Model of Institutional Adaptation to Student
Diversity” (Richardson, 1990) was also used. Within the model, institutional
mission and state policy environments are seen as the essential exogenous variables
that stimulate the adaptation process, with the need for adaptation being stimulated
first by changes in the state policy environment. The extent of institutional
adaptation to accommodate quality and diversity is most directly affected by the
impact of institutional interventions on six dimensions of the organizational culture.
For the purpose of this study only one aspect within that dimension is reviewed,
namely the degree to which the campuses included in this study have made a
commitment to faculty diversity in proportion to the changing demographics of their
students. Further, the study questions focused on the phenomenon of “majority/
minority” campuses to ascertain any relationship between campus demographics and
academic success. In other words, in colleges where Latinos are not the
demographic minority does it follow that academic achievement gaps associated
with under-representation also disappear?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5 7
Community College Model for Student Life
and Course Completion (CCSL)
The model selected to provide the context for the examination of academic
variables and their relationship to academic success is the Community College
Model for Student Life and Course Completion. The variables selected are those
established in most “interactionist’ model theories (Fishbein and Azjen 1975; Bean
and Metzner, 1985, Benjamin, 1994, Tinto, 1998) as reinterpreted in the Community
College Model of Student Life and Retention as depicted in Figure 4, below. The
importance of this model design for this study is its unique adaptation to non-
traditional community college students as described by the researchers below:
Our results clearly demonstrate that the student life (CCSL) as a construct
mediates the relationship between (which variables) and the ability for
persistence (defined as course completion). However, for a diverse and urban
population, we found that proficiency with the English language and
obstacles to postsecondary education are influential in course completion
(Hagedom, L.S., et al (2002).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Figure 4. C om m unity College M odel for S tudent Life and Course Com pletion
20
Beliefs and
Academic
Attitude
o n se quenqes
igh School
GPA
Student
S felf-Percepti
4 7 -06
Cam pus
Poverty
spirations
English
Abitit
Student
Life
Course
Academic
G ender
eeiings of
O bstacles
Children
eekly Hou
mploymen
Normative
EeM s
Subjective
Norms
Source: Hagedorn, L.S.,et al (2002)
©O
5 9
An application of these models generates the following research
questions:
QUESTION 1: What academic, environmental and social integration
variables maximize academic success leading to transfer preparedness for Latino
community college students?
QUESTION 2: What is the relation between the level of representation of
Latino students on campus to overall academic success?
QUESTION 3: What is the relation between the level of ethnic representation
of faculty on campus to overall academic success of Latino students?
The research questions generated the following hypothesis:
Hypothesis 1 : There are specific academic, environmental and social
integration variables that alone and in combination maximize academic success and
the preparation of Latino community college students for transfer.
Hypothesis 2: The level of Latino student representation on campus is
important to their level of academic success.
Hypothesis 3: The level of representation of faculty from ethnic minority
backgrounds is important to the academic success of Latino students.
Methodology
Research Population
This study utilized data collected through the TRUCCS project. In the fall of
1999, the TRUCCS research team developed a 47-item questionnaire specifically
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
designed for urban community college campuses with diverse student
enrollments and also with large number of students for whom English is not a first
language. The questionnaires were developed to include items and scales particular
to community college students reflecting the influence of the literature of
interactionist theories of student success including Bean and Metzner (1985), de los
Santos and Wright (1990), Hagedom and Castro (1999), and others.
Research Design
This is a secondary analysis of data that has been gathered, analyzed and
validated through the TRUCCS Project. The description of the research population
is taken directly from the reports produced by the research team (Hagedom et al
2002). The final sample for this study consists of 4, 967 students from all nine
colleges within the Los Angeles Community College District that participated in the
TRUCCS survey and for whom transcript data could be assessed. For purposes of
this study these students become the experimentally accessible population from
which the sample of all Latino students is obtained. The number of Latino students
within this sample is 2,570 or 51.7 percent with the remaining 2,397 students
comprising 48.3 percent of the total.
This data base used in this study has been validated and refined through data
analysis processes using sampling design that maximized variation in the
independent variables in the sample to allow researchers to make internally valid
comparisons of subgroups (Hagedom, L.S. and Maxwell, B, 1999). This is an
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
61
inferential study using a quasi-experimental design based on regression analysis
design establishing three groupings based on the ethnic representational value (RV),
of Latinos enrollment at each of the campuses in the Los Angeles Community
College District. Four colleges with Latino enrollment of 50 percent or higher are
designated “high”. Three colleges with 40-30 percent are designated “moderate” and
two colleges with 20 to 30 percent are designated “low.” In addition, this study
tested a series of factorial designs including forward entry regression to test various
models that demonstrate causal relationships between the level and percentage of
representation of Latinos in a given campus and their level of academic success
within the context of academic, environmental and social integration factors. Scales
previously identified by factor analysis and submitted to reliability analysis will
represent these constructs. The study seeks to identify which factors or combination
of factors have the highest correlation with academic success. The study used course
completion as proxy for academic success. Student success was determined by the
number of courses successfully completed with an A, B, or C grade over the total
number of courses attempted with grades of A, B, C, D, F, I (incomplete), NC (no
credit) or W (withdrawal).
Number o f courses completed w/an A. B C or P
Number of courses attempted w/an A, B, C, D, F, NC, W, or I
This ratio has proven to be the best measure of student success because it reduces the
factor to the lowest common denominator thus minimizing “noise” or the
interference of other attributes that might contribute or detract from student success.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 2
Given the established validity of the TRUCCS data, it is expected that
generalizations may be made to Latino community college students who are the
target population of this study.
Instrumentation
In the fall of 1999, the TRUCCS research team developed a 47-item
questionnaire designed specifically for an urban campus with significant student
diversity and large numbers of non-native speakers of English.
The survey instrument was piloted and revised and in the Spring 2000
semester, it was administered to 5,000 students across 241 classrooms and nine
colleges within the Los Angeles Community College District. Participating
classrooms were identified through a stratified random sampling method that relied
heavily on three levels of English courses (2 levels below transfer, 1 level below
transfer, and transfer level), occupational programs stratified by gender
predominance, remedial courses, regular courses, learning communities, and
traditional gateway courses. In addition, transcript data was acquired from the Los
Angeles Community College District for all students who signed the requisite
consent forms (96 percent of the sample). The final sample for this study consists of
4,4333 students from the Los Angeles Community College District that participated
in the TRUCCS survey and for whom transcript data could be assessed.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
63
Data Analysis
This study uses both descriptive statistics and inferential statistics to
investigate the existence and extent of relationships among selected variables.
Descriptive statistics including frequencies, means, and standard deviations for the
dependent variable that is successful course completion used as a proxy for academic
success are also examined.
Paths were hypothesized to follow the examples of the Community College
Student Life Model (CCSL). For example, several paths were opened through a
series of factorial analysis to explore the relationship between academic success and
a series of significant variables identified in the CCSL model. These factors were
grouped into four basic clusters: background, college, faculty and staff
demographics.
Several tables were designed to respond to the research questions and test the
attendant hypothesis. The tables were of two types the first set provides descriptive
demographic data and the second set include tables of means to compare successful
course enrollment across institutions designated as having high (50 percent or more)
moderate (30 percent to 50 percent) or low (20 percent to 30 percent) Latino student
enrollment. One and two way analysis of variance were used to test the difference of
means across several factors including: student performance as measured by GPA,
successful course completion, and remedial or transfer level enrollment patterns for
Latinos in relationship to native and non native English speaking ability. Six
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 4
separate analysis of variance to test enrollment patterns in English and math
courses were conducted to ensure that these patterns would be mutually exclusive
and collectively exhaustive by course.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
65
CHAPTER IV
RESULTS
Introduction
In this chapter, observed outcomes are analyzed in the context of the research
questions presented in Chapter One. Descriptive findings are presented first which
include demographic information necessary to frame the remainder of the findings.
Statistical findings follow which are interpreted in relation to the extent to which
they support or failed to support the research hypotheses. As all three-research
hypotheses specifically predicted the relationship among the variables of interests,
findings are evaluated in relation to the one-tailed significance tests of zero-order
correlation coefficients consistent with the directional nature of the hypotheses.
Results are discussed within the context of the interactionist model of student
success presented as the conceptual framework for this study.
The first section of this chapter presents the construct validity and reliability
with the principal component analysis of each measure. The second section presents
the descriptive statistics of observed variables, and examines the relationship among
observed variables. The third section presents a model summary of observed
variables that examines several independent variables and their relation to student
success. The fourth section presents the results of one way analysis of covariance
(ANOVA) and path analysis to determine if the results support the hypothesis that
there are specific academic environmental and social integration variables that alone
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
66
and in combination maximize academic success and that the level of Latino
student and minority faculty representation on campus also impact student
achievement.
Construct Validity and Reliability
Principal Component Analysis
Construct validity of the independent variable was established as valid and
reliable through prior research conducted by Dr. Linda Serra Hagedom (2002). Dr.
Hagedom used structural equation modeling that tested the relationship between
beliefs, intentions and course completion resulting in the Community Model for
Student Life and Course Completion (CCSL) (Hagedom, L.S. et al 2002).
Based on the CCSL model there are nine subscales and three additional
variables included in this analysis all of which were submitted to reliability item
analysis which yielded a >.7 Cronbach’s Alpha. These subscales include: Beliefs
and Consequences, Attitude, Subjective Norms, Aspirations, Normative Beliefs,
Determination, Academic Integration, Obstacles and Student’s English Ability. The
additional variables are the proportion of minority faculty, full and part time, as well
the RV value per institution for a total of twelve independent variables (IV). The
variables were then entered in a block regression. Successful course completion (i.e.
courses completed with a C or better) is the dependent variable (DV). Appendix A,
includes the factorial tables produced to validate this analysis. The table below lists
the twelve subscales selected for the study grouped in the four clusters (background,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 7
college, student demographics, faculty demographics) used for the block entries
in this study.
Table 6
Pattern Matrix of Scales Selected
Scale and Cronbach’s
Alpha Items/Variables Comprising Scale
BACKGROUND
Beliefs and Consequences
Alpha = .7093
Influences on decision to come to the particular college:
Graduates get good jobs
Students transfer to good 4-year schools
To get a better job
To get a college degree
To enroll in a special program or certificate
Attitude
Alpha = .7301
Strongly Agree to Strongly Disagree
Understanding what is taught is important
I always complete homework assignments
Success in college large due to effort
I can leam all skills taught in college
Enjoy challenging class assignments
Expect to do well/earn good grades
Subjective Norm
Alpha = .7378
Why in college
My parents wanted me to come
Other family member wanted me to come
High School or other counselor advised me
My friends are attending here
My employer encouraged me to enroll here
Aspirations
Alpha = .7240
As things stand, do you think you will
Transfer to a 4-year college/university
Get a bachelor’s degree
Normative Beliefs
Alpha = .7781
Reasons for coming
Zscore: This college has good social activities
Zscore: Offers special educational programs
Zscore: This college has a good reputation
What people think about the college
Zscore: You
Zscore: Your closest friends
Zscore: Your spouse or partner
Zscore: Your parents or guardians
Zscore: Your other relatives
Zscore: Your high school teachers
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
68
Scale and Cronbach’s
Alpha Items/Variables Comprising Scale
Determination
Alpha = .7395
Strongly agree to strongly disagree
Satisfied when I work hard to achieve
I am very determined to reach my goals
Important to finish courses in program o f studies
Keep trying even when frustrated by task
COLLEGE
Academic Integration
Alpha = .7473
How often or how many times
Talk with instructors before or after class
Talk with instructors during office hours
Help another student understand homework
Study in small groups outside o f class
Speak with an academic counselor
Telephone/email/student about studies
Obstacles
Alpha = .7123
How large a problem
Parking
Transportation
Family responsibilities
Job-related responsibilities
Paying for college
Scheduling classes for next semester
Understanding the English language
Difficulty o f classes
STUDENT DEMOGRAPHICS
English Ability
Alpha = .9231
Ability in English
Read
Write
Understand a college lecture
Read a college text book
Write an essay exam
Write a term paper
Participate in class discussion
Communicate with instructor
Age*
Ethnicity*
FACULTY
DEMOGRAPHICS
Full Time Faculty*
Part Time Faculty*
Note: These items simply represent factors entered and therefore do not contain
Cronbach’s Alpha reliability scores.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 9
Statistical Analysis
Descriptive Statistics: Exploratory Factor Analysis
The tables that follow are of two types. The first set provides descriptive
demographic data as the necessary context for subsequent statistical analysis. The
second set include several tables of means comparing successful course enrollment
across high, moderate and low RV institutions for all Latino students and course
enrollment patterns for Latinos who are native and non native speakers of English in
gate keeping courses of English and math.
One of the first steps undertaken was to assign a “representational” value to
each of the nine colleges in the Los Angeles Community College District based on
the percentage of the total Latino student enrollment in a given campus. The second
step assigned a representational value (RV) to these institutions of, high, medium or
low. Academic Success (successful course completion) is the dependent variable
(DV) and ethnic representational value is the independent variable (IV).
Descriptive Tables of the Demographic Context
Table 7 presents the number of Latinos residing in the service area of each of
the LACCD campuses in contrast to the number of Latino students enrolled. Los
Angeles Mission, Los Angeles Valley, Los Angeles Harbor, West Los Angeles and
Los Angeles Pierce enroll more Latino students on their campus than the percentage
residing in their service area. On the other hand, East Los Angeles, Los Angeles
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
70
Trade Tech., Los Angeles City and Los Angeles Southwest enroll fewer
students on their respective campuses than the percentage represented in the service
area.
Table 7
Los Angeles Community College District Student Headcount by Ethnicity By
College for 2002 Spring Semester
Name of
College
Hispanic
Student
Headcount
Total
Headcount
Percentage of
Total
Headcount
Percentage of
Pop. in Service
Areas
Campus 1
(LAMC)
5,991 9,143 65.5 47.8
Campus 2
(ELA)
20,174 32,630 61.8 70.6
Campus 3
(LATT)
7,863 14,328 54.8 66.5
Campus 4
(LAVC)
7,384 19,078 51.5 33.4
Campus 5
(LAHC)
4,168 10,356 40.2 31.5
Campus 6
(LACC)
8,346 24,761 33.7 45.1
Campus 7
(LASW)
2,532 8,573 29.5 35.6
Campus 8
(WLA)
2,692 11,232 23.9 16.4
Campus 9
(LAPC)
4,259 19,368 21.9 18.7
LACCD
G rand Total
63,629 50,209 42.3 42.3
Sources: California Community Colleges State Chancellor’s Office
(misweb.cccco.edu/mis/onlinestat/studdemo_coll_rpt.cfin), LACCD Websites
Table 8 establishes groupings of college depending on their level of Latino
representation. Those colleges with a percentage of Latino enrollments that is higher
than 50 percent are designated as colleges with high RV, colleges with Latino
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71
enrollment between 30 percent and 49 percent are moderate RV campuses and
those colleges with 20 percent to 29 percent are in the low category.
Table 8
Demographic Descriptors/Establishing Representational Values (RV)
College
Percentage o f Latino
Enrollment
Proposed Grouping Based
on Ethnic Representation
Value (RV)
Campus 1 65.5 HIGH
Campus 2 61.8 HIGH
Campus 3 54.8 HIGH
Campus 4 51.5 HIGH
Campus 5 40.2 MODERATE
Campus 6 33.7 MODERATE
Campus 7 29.5 MODERATE
Campus 8 23.9 LOW
Campus 9 21.9 LOW
LACCD Grand Total 42.3
As may be seen campuses one through 4 are designated with High Representational
Value (HRV), campuses five through seven are Moderate Representational Value
(MRV) institutions and campuses eight and nine are Low Representational Value
(LRV) colleges.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Tables 9 and 10 list the colleges by level of Latino student enrollment in comparison to the number of full and part time and
part-time minority faculty employed by the college that may serve as role models.
Table 9
LACCD 2001 Full-Time Faculty Demographics
College by
Percent of
Latino
Enrollment
Total
Number
ofFull
Time
Faculty
Number and
Percentage of
Minority
Faculty
Number and
Percentage of
White
Faculty
Number and
Percentage of
Black Faculty
Number and
Percentage of
Latino
Faculty
Number and
Percentage of
Asian/Pacific
Islander Faculty
Number and
Percentage of
Native
American
Faculty
Number and
Percentage of
Male Faculty
Number and
Percentage of
Female
Faculty
High # % # % # % # % # % # % # % # %
l 165 56 33.9 109 66.0 16 9.6 28 16.9 12 7.2 0 0.0 90 54.6 75 45.5
2 402 210 52.2 192 47.7 21 5.2 128 31.8 60 14.9 1 0.2 369 67.0 133 33.0
3 208 101 48.5 107 51.4 46 22.1 31 14.9 22 10.5 2 0.9 104 50.0 104 50.0
4 236 59 25.0 177 7.50 21 O O
0 0
22 9.3 14 5.9 2 0.8 126 53.4 110 53.4
Moderate
5 217 70 30.2 147 67.7 20 9.3 28 12.9 21 9.6 1 0.4 127 58.6 90 41.4
6 322 104 32.2 218 67.7 42 13.9 25 7.7 36 11.1 1 0.3 291 59.4 131 40.6
7 113 82 72.5 31 27.4 67 0.6 8 7.0 7 6.1 0 0.0 50 44.3 63 55.7
Low
8 132 52 39.3 80 60.6 28 21.2 13 9.8 10 7.5 1 0.7 65 49.3 67 50.7
9 369 53 14.3 316 85.6 12 3.2 17 4.6 23 6.2 1 0.2 229 62.1 140 37.9
to
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 10
LACCD 2001 Part-Time Faculty Demographics
College by
Percent of
Latino
Enrollment
Total
Number
of Part
Time
Faculty
Number and
Percentage
o f Minority
Faculty
Number and
Percentage
o f White
Faculty
Number and
Percentage
o f Black
Faculty
Number and
Percentage
o f Latino
Faculty
Number and
Percentage of
Asian/Pacific
Islander
Faculty
Number and
Percentage
o f Native
American
Faculty
Number and
Percentage of
Male Faculty
Number and
Percentage
of Female
Faculty
H ig h # % # % # % # % # % # % # % # %
1 95 32 33.6 143 66.3 7 7.3 18 18.9 7 7.3 0 0.0 44 66.4 51 53.6
2 256 121 47.2 135 52.7 11 4.2 73 28.5 37 14.4 0 0.0 133 52.0 123 48.9
3 264 139 52.6 125 47.3 64 24.2 45 17.9 28 10.6 2 0.7 173 666.3 89 33.7
4 275 61 22.1 214 77.8 16 5.8 23 8.3 21 7.6 1 0.3 167 60.8 108 39.2
Moderate
5 142 52 36.8 89 63.1 19 13.4 17 12.0 15 10.6 1 0.7 65 45.1 76 53.9
6 255 90 35.2 165 64.7 35 13.2 29 11.3 25 9.8 1 0.3 145 56.9 110 43.1
7' 146 103 70.5 43 29.4 79 0.5 14 9.5 9 6.1 1 0.6 90 61.7 56 38.3
Low
8 259 107 41.3 152 58.6 65 25.0 19 7.3 23 8.8 0 0.0 171 66.5 87 33.5
9
201 30 14.9 171 85.0 7 3.4 15 7.4 7 3.4 1 0.4 114 56.8 87 43.2
74
Descriptive Statistics for Academic Achievement by RV Level
Tables 11 through 14 present descriptive statistics about the characteristics of
the Latino student sample in relation to the representational value (RV) of campus
groupings. These tables are designed to address most directly one of the main
research questions established in this study:
What is the relationship between the levels of representation of Latino
students on campus to overall academic success as measured by the proportion of
courses completed with grades of C or better and overall GPA?
Table 11 gives descriptive means and standard deviations of academic performance
measures (proportion of courses completed with grades of C or better, and
cumulative GPA) under levels of RV. Table 12 shows descriptive statistics of
enrollment patterns for Latino students in remedial and non-remedial gate keeping
courses (English and Mathematics). Table 13 presents patterns of non-remedial
course enrollment for native and non-native speakers of English. Table 14 shows
descriptive statistics for remedial course enrollment by native and non-native
speakers of English. The following section describes one and two-way analyses of
variance used to test the differences of the means shown in Tables 11 -14.
One-way Analyses of Variance for Academic Achievement by RV Level
The question examined in Table 11 is whether RV has any significance as a
predictor for the proportion of successful courses completed and academic
performance for all Latino students.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
75
Table 11
Mean Academic Performance for Colleges with High. Moderate and Low Latino
Student Representation
College
Representational
Value (RV)
Groupings
(Percentage of
Latino
Enrollment)
Mean Proportion
o f Courses
Completed with
C or better
Mean Number of
Courses
Completed with
grades of C or
better.
Cumulative
Grade Point
Average
(C.G.P.A.)
Mean
(S.D.)
Mean
(S.D.)
Mean
(S.D.)
HighRV 0.84 13.62 2.42
>50 % (0.16) (10.46) (0.78)
Moderate RV 0.84 14.75 2.42
<50% and >30% (0,16) (10.48) (0.79)
Low RV 0.84 9.98 2.26
<30% and >20% (0.19) (8.48) (0.86)
One-Way ANOVA of proportion of courses completed with C or better by RV
Table 11 (a)
Descriptives
Measure o f Academic Success N Mean
Standard
Deviation Standard Error
Courses 1 HighRV 1492 .84203 .16131 4.18E-03
Completed 2. Moderate RV 601 ,83774 16492 6.73E-03
with C or 3. Low RV 226 .81881 19139 1.27E-02
Better/ Courses
Attempted
Total 2319 .83865 16546 3.44E-03
GPA 1 High RV 1576 2.4242 .7842 1.975E-02
2. Moderate RV 608 2.4230 .7935 3.218E-02
3. Low RV 244 2.2626 .8559 5.480E-02
Total 2428 2.4076 .7952 1.614E-02
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
76
Table 11 (b)
ANOVA
Measure of Academic Success Sum of Squares D f Mean Square F Sign.
Courses
Completed
with C or
Better/Courses
Attempted
Between Groups
Within Groups
Total
.106
63.356
63.463
2
2316
2318
5.322E-02
2.736E-02
1.945 .143
GPA Between Groups
Within Groups
Total
5.710
1528.859
1534.568
2
2425
2427
2.8555
.630
4.528 .011
A one-way ANOVA of the proportion of courses completed with grades of C or
better by RV was not significant. On the other hand, a one-way ANOVA of GPA by
RV was significant (F=4.53, df=2, p=0.011). Post hoc, pair-wise comparisons of
levels of RV indicate significantly higher GPAs for the High and Moderate RV
groups as compared to the Low group (a difference of 0.16, p-0.003 and a difference
of 0.16, p=0.008 respectively). There was no significant difference in GPA for the
High and Moderate RV groups.
One-way Analyses of Variance for Enrollment in Gate Keeping Courses by RV Level
The question examined in Table 12 is whether college RV has any
relationship to the proportion of all Latino students who enroll in remedial/pre-
collegiate and non-remedial/transfer gate keeping courses, namely English and math.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
77
Table 12
Latino Students Remedial and Non-remedial Enrollment Patterns in Gate Keeping
Courses
College
Representational
Value (RV)
Groupings
(Percentage of
Latino
Enrollment)
English Math
Proportion
Not
Enrolled
in Any
English
Proportion
of
Enrollment
in
Remedial
Proportion
of
Enrollment
in
Non-
remedial
Proportion
Not
Enrolled
in Any
Math
Proportion
o f
Enrollment
in
Remedial
Proportion
of
Enrollment
in Non-
Remedial
HighRV
>50 %
0.18
(0.39)
0.49
(0.50)
0.32
(0.47)
0.24
(0.43)
0.37
(0.48)
0.39
(0.49)
Moderate RV
<50% and >
30%
0.15
(0.36)
0.51
(0.50)
0.34
(0.47)
0.25
(0.43)
0.42
(0.49)
0.34
(0.47)
LowRV
<30% and
>20%
0.20
(0.40)
0.57
(0.50)
0.23
(0.42)
0.33
(0.47)
0.34
(0.47)
0.33
(0.47)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 12 (a)
One-way ANOVAs: English and Math (non-taken, remedial, non-remedial) by RV
Descriptives
N Mean Sid. Deviation Std. Error
95% Confidence Interval for
Mean
Minimum Maximum Lower Bound Upper Bound
ENGANY1 No English 1 High 1668 .18 .39 9.50E-03 .17 .20 0 1
courses attempted
2 Moderate 647 .15 .36 1.42E-02 .13 .18 0 1
3 Low 255 .20 .40 2.51 E-02 .15 .25 0 1
Total 2570 .18 .38 7.55E-03 .16 .19 0 1
ENGRM1 One or more 1 High 1668 .49 .50 1.22E-02 .47 .52 0 1
remedial English courses
2 Moderate 647 .51 .50 1.97E-02 .47 .55 0 1
3 Low 255 .57 .50 3.10E-02 .51 .63 0 1
Total 2570 .51 .50 9.86E-03 .49 .53 0 1
ENGNR1 Non-remedial 1 High 1668 .73 .44 1.09E-02 .71 .75 0 1
English courses and
2 Moderate 647 .77 .42 1.65E-02 .74 .80 0 1
remedial courses
3 Low
255 .62 .49 3.05E-02 .56 .68 0 1
Total
2570 .73 .44 8.78E-03 .71 .75 0 1
MTHANY1 No Math 1 High 1668 .24 .43 1.05E-02 .22 .26 0 1
courses attempted
2 Moderate 647 .25 .43 1.70E-Q2 .21 .28 0 1
3 Low 255 .33 .47 2.95E-02 .27 .39 0 1
Total
2570 .25 .43 8.55E-03 .23 .27 0 1
MATHREM1 One or more 1 High 1668 .37 .48 1.18E-02 .34 .39 0 1
remedial Math courses
2 Moderate 647 .42 .49 1.94E-02 .38 .46 0 1
3 Low 255 .34 .47 2.97E-02 .28 .40 0 1
Total 2570 .38 .48 9.56E-03 .36 .40 0 1
MATHNR2 Non-remedial 1 High 1668 .39 .49 1.20E-02 .37 .42 0 1
Math Courses only
2 Moderate 647 .34 .47 1.86E-02 .30 .37 0 1
3 Low 255 .33 .47 2.96E-02 .28 .39 0 1
Total 2570 .37 .48 9.54E-03 .35 .39 0 1
79
ANOVA
Sum of
Squares df ^/lean Square F Sig.
ENGANY1 No English Between Grouf
courses attempted Within Groups
Total
.601
375.779
376.380
2
2567
2569
.301
.146
2.053 .129
ENGRM1 One or mon Between Grou[
remedial English cours within Groups
Total
1.380
641.033
642.412
2
2567
2569
.690
.250
2.763 .063
ENGNR2 Non-remedi; Between Grou|
English Courses only within Groups
Total
2.331
553.115
555.446
2
2567
2569
1.165
.215
5.408 .005
MTHANY1 No Math
courses attempted
Between Grou|:
Within Groups
Total
1.764
481.359
483.123
2
2567
2569
.882
.188
4.703 .009
MATHREM1 One or n Between Grou[
remedial Math courses within Groups
Total
1.624
601.775
603.399
2
2567
2569
.812
.234
3.463 .032
MATHNR2 Non-reme< Between Grouf
Math Courses only within Groups
Total
1.962
598.677
600.639
2
2567
2569
.981
.233
4.205 .015
Six separate analyses of variance were conducted for three types of enrollment in
English and mathematics courses over levels of RV:
1) No English courses attempted
2) One or more remedial English courses attempted
3) Only transfer/non-remedial English courses attempted
4) No mathematics courses attempted
5) One or more remedial mathematics courses attempted
6) And only transfer/non-remedial mathematics courses attempted.
These patterns of course taking are mutually exclusive and collectively exhaustive by
course.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
80
Of the six one-way ANOVAs, numbers 3 through 6 were significant at
the 0.05 level. The one-way ANOVA and subsequent post hoc tests in Analysis 3
revealed that those students in the High and Moderate RV categories enrolled in
significantly more transfer/non-remedial English courses than did students in the
Low RV group (p=0.008 and p=0.004 respectively). There was no significant
difference in enrollment between the High and Moderate RV levels. In Analysis 4
the proportion of Latino students who enrolled in no mathematics courses differed
significantly by RV level (F=4.70, df=2, p=0.009). Post hoc tests revealed that a
significantly lower proportion of students in the High and Moderate RV groups as
opposed to the Low RV group enrolled in no mathematics courses. There was no
significant difference between the High and Moderate groups. In Analysis 5 the
proportion of Latino students who enrolled in one or more remedial mathematics
courses differed significantly by RV level (F=4.703, df=2, p=0.009). Post hoc tests
showed that a significantly greater proportion of students in the Moderate RV group
as opposed to the High and Low groups enrolled in remedial mathematics courses.
There was no significant difference between the High and Low RV groups. Analysis
6 showed that Latino students who enrolled only in transfer level mathematics
courses differed with respect to RV levels (F=4.20, df=2, p=0.015). Those in the
High RV category enrolled in more transfer level mathematics courses than those in
the moderate RV category. There was no significant difference between the High and
Low and Moderate and Low RV groups.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
81
Two-way ANOVA of Transfer Level English and Mathematics Enrollment by
Native Language and RV
The focal point for Table 13 is to examine the relationship between
Representational Value, language proficiency and enrollment with regards to gate
keeping English and mathematics courses. In other words is there a relationship
between the RV status of an institution and the placement of students in transfer
level English and math courses based on whether or not they were native or non
native speakers of English.
Table 13
Transfer Level English and Mathematics Course Enrollment for Native and Non-
Native Speakers of English by RV Level
College
Representational
Value (RV)
Groupings
(Percentage o f Latino
Enrollment
FRESHMAN ENGLISH
(BA/Transfer Level)
MATH
(BA/Transfer Level)
Native Non-Native Native Non-Native
HighRV
>50 %
0.63 0.58 0.29 0.32
(0.48) (0.49) (0.45) (0-47)
0.68 0.59 0.4 0 0.33
Moderate RV
<50% and > 30%
(0.47) (0.49) (0.49) (0.47)
LowRV
<30% and >20%
0.59 0.49 0.15 0.13
(0.49) (0.50) (0.36) (0.34)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
82
Table 13 (a)
Two-Way ANOVA (Transfer English by RV and Native Language)
Descriptive Statistics
Dependent Variable: ENGTRAN1 One or more English transfer courses
RV
Q20 Is English your
Mean Std. Deviation N
1 High 1 No .58 .49 1108
2 Yes .63 .48 461
Total .59 .49 1569
2 Moderate 1 No .59 .49 430
2 Yes .68 .47 186
Total .62 .49 616
3 Low 1 No .49 .50 171
2 Yes .59 .49 74
Total .52 .50 245
Total 1 No .57 .49 1709
2 Yes .64 .48 721
Total .59 .49 2430
Table 13 (b)
T ests of Betw een-Subjects Effects
Dependent Variable: ENGTRAN1 One or more English transfer courses
Source
Type III Sum
of Squares df Mean Square F Sig.
Corrected Model 4.321a 5 .864 3.597 .003
Intercept 420.900 1 420.900 1751.983 .000
RV 1.344 2 .672 2.796 .061
Q20 2.064 1 2.064 8.591 .003
RV * Q20 .171 2 8.551 E-02 .356 .701
Error
582.346 2424 .240
Total 1440.000 2430
Corrected Total 586.667 2429
a. R Squared = .007 (Adjusted R Squared = .005)
Two-way ANOVA (Transfer level Math by RV and Native Language)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
83
Table 13 (c)
Descriptive Statistics
Dependent Variable: MTHTRAN1 One or more transfer math courses
RV
Q20 Is English your
Mean Std. Deviation N
1 High 1 No .32 .47 1108
2 Yes .29 .45 461
Total .31 .46 1569
2 Moderate 1 No .33 .47 430
2 Yes .40 .49 186
Total .35 .48 616
3 Low 1 No .13 .34 171
2 Yes .15 .36 74
Total .14 .35 245
Total 1 No .30 .46 1709
2 Yes .31 .46 721
Total .30 .46 2430
Table 13 (d)
T ests o f B etw een-Subjects E ffects
Dependent Variable: MTHTRAN1 One or more transfer math courses
Source
Type III Sum
of Squares df Mean Square F Sig.
Corrected Model 9.123a 5 1.825 8.763 .000
Intercept 87.567 1 87.567 420.535 .000
RV 7.420 2 3.710 17.818 .000
Q20 .122 1 .122 .586 .444
RV * Q20
1.015 2 .507 2.437 .088
Error
504.743 2424 .208
Total
738.000 2430
Corrected Total 513.867 2429
a- R Squared = .018 (Adjusted R Squared = .016)
An overall two-way analysis of variance predicting enrollment in transfer
level English by the three levels of RV and two levels of Language Proficiency ("Is
English your native language?" was significant (F=3.60, df=5, p=0.003). English
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8 4
language proficiency was the only significant effect. In general, native English
speakers increased the likelihood of enrolling in one or more transfer level English
courses by eight percent. There was no significant interaction between language
proficiency and representational value.
With respect to transfer level mathematics course taking, an overall two-
analysis of variance for three levels of RV and two levels of Language Proficiency
was significant (F=8.76, df=5, p=0.0005). In contrast to the previous model
predicting English transfer course enrollment, the model for mathematics showed
only a main effect for Representational Value (F=3.71, df=2, p=0.0005). Students in
the Moderate RV category showed significantly higher enrollment in transfer level
mathematics courses than those in either the Low or High RV categories. Students in
the High RV category were more likely to enroll in transfer level math courses than
students in the Low RV category. There was no significant interaction.
Table 14 examines language proficiency, college RV and enrollment in
remedial/pre-collegiate gate keeping courses.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
85
Table 14
Remedial Level English and Mathematics Course Enrollments for Native and
Non-native Speakers of English by RV Level
Patterns of Remedial Course
Enrollment for Native and Non-
Native Speakers o f English by
College Representational Value
(RV) Groupings
Remedial English Remedial Math
Native Non-native Native Non-Native
HighRV
0.42 0.52 0.37 0.36
>50 %
(0.49) (0.50) (0.48) (0.48)
Moderate RV
0.39 0.56 0.40 0.42
<50% and >30%
(0.49) (0.50) (0.49) (0.49)
0.42 0.63 0.30 0.33
Low RV
<30% and >20%
(0.50) (0.49) (0.46) (0.47)
The two-way ANOVA predicting pre-collegiate English enrollment was significant
(F=8.65, df=5, p=0.Q005). Language Proficiency was the only main effect (F=31.40,
df=l, p=0.0005). Native speakers of English were enrolled in significantly fewer
remedial courses than non-native speakers (mean difference, -0.161; p=0.0005). The
overall two-way analysis of variance for remedial mathematics course enrollment by
RV and language proficiency was not significant.
Two-way ANOVA (remedial English and math by RV and Native Language)
Table 14(a) Descriptive Statistics
Q20 Is English your
RV Mean Std. Deviation N
1 No 1 High .52 .50 1108
2 Moderate .56 .50 430
3 Low .63 .49 171
Total .54 .50 1709
2 Yes 1 High .42 .49 461
2 Moderate .39 .49 186
3 Low .42 .50 74
Total .41 .49 721
Total 1 High .49 .50 1569
2 Moderate .51 .50 616
3 Low .56 .50 245
Total .50 .50 2430
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 14 (b)
Tests o f B etw een-S ubjects E ffects
Dependent Variable: ENGRM1 One or more remedial English courses
Source
Type III Sum
of Squares df Mean Square F Sig.
Partial Eta
Squared
Noncent.
Parameter
Observed
Power3
Corrected Model 10.648b 5 2.130 8.650 .000 .018 43.248 1.000
Intercept 285.271 1 285.271 1158.624 .000 .323 1158.624 1.000
Q20 7.731 1 7.731 31.398 .000 .013 31.398 1.000
RV .480 2 .240 .975 .377 .001 1.950 .221
Q20 * RV .728 2 .364 1.478 .228 .001 2.955 .317
Error 596.825 2424 .246
Total 1223.000 2430
Corrected Total 607.474 2429
a. Computed using alpha = .05
b- R Squared = .018 (Adjusted R Squared = .016)
00
Os
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 14 (c)
D escrip tive S ta tistic s
Dependent Variable: MATHREM1 One or more remedial Mat i courses
Q20 Is English your r v
Mean Std. Deviation N
1 No 1 High .36 .48 1108
2 Moderate .42 .49 430
3 Low .33 .47 171
Total .37 .48 1709
2 Yes 1 High .37 .48 461
2 Moderate .40 .49 186
3 Low .30 .46 74
Total .37 .48 721
Total 1 High .36 .48 1569
2 Moderate .41 .49 616
3 Low .32 .47 245
Total .37 .48 2430
00
< 1
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 14 (d)
T e s ts o f B e tw e e n -S u b je c ts E ffects
Dependent Variable: MATHREM1 One or more remedia Math courses
Source
Type III Sum
of Squares df Mean Square F Sig.
Partial Eta
Squared
Noncent.
Parameter
Observed
Powef3
Corrected Mode 1.918b 5 .384 1.643 .145 .003 8.213 .576
Intercept 158.341 1 158.341 678.083 .000 .219 678.083 1.000
Q20 5.713E-02 1 5.713E-02 .245 .621 .000 .245 .078
RV 1.497 2 .749 3.206 .041 .003 6.413 .614
Q20 * RV .129 2 6.452E-02 .276 .759 .000 .553 .094
Error 566.035 2424 .234
Total 905.000 2430
Corrected Total 567.953 2429
a- Computed using alpha = .05
b- R Squared = .003 (Adjusted R Squared = .001)
00
00
89
Model Summary
Multiple Regression of Successful Course Completion on Variables
Table 15 gives the results of a block forward entry multiple regression of
academic success (proportion of successful course completion to total courses
attempted) on a series of factorial constructs exploring the relationship between a
series of variables identified in the CCSL model. These independent variables are
grouped in five basic clusters: Student background (norms and beliefs),
Commitment to college (academic integration and perception of obstacles), student
demographics, faculty demographics, and representational value (with the levels of
RV entered as three dummy variables). The multiple regression model is designed to
examine each of the three research questions established in this study:
• What academic, environmental and social integration variables maximize
academic success for Latino community college students?
• What is the relationship between the levels of representation of Latino
students on campus to overall academic success?
• What is the relationship between the level of ethnic representation of
faculty on campus to overall academic success of Latino students?
Means, variances and standard deviations of the observed variables for Table
15 are presented in the statistical analysis conducted in Tables 15 (a) through 15 (c).
The regression model after the final entry was significant ( F=3.670, df=12,
p=0.0005). Upon entry of the first block Determination is retained as a single
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
90
independent variable (t=3.31, p=0.001). The value of R2 for this initial model is
equal to 0.007. The second block entry introduces only the Beliefs and
Consequences variable (t=-2.08, p=0.038); this concept enters with a negative weight
(beta = -0.053, t=-2.08, p= 0.038). The value of R2 increases by only 0.003. The third
block enters Age with a negative weight (beta = -0.10, t = -3.93, p = .0005). R2
increases to 0.019. The final block adds the proportion of minority faculty on
campus. In the fourth block, the Beliefs and Consequences construct is no longer
significant at the 0.05 level. The value of R2 for the final model is 0.024. The only
significant factors that remain after the final entry are: (1) self determination on the
part of the student, (2) positive correlation between the age (maturity) of the student
and performance, and (3) a positive correlation with the proportion of minority full
time faculty employed by the institution.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
91
Table 15
Block Entry Analysis of the Impact of Institutional, Individual and Demographic
Variables on Latino Student Success: Regression with Scale Values
Coefficients8
Unstandardized
Coefficients
Standardi
zed
Coefficien
ts 95% Confidence Interval for B
Model B Std. Error Beta t Sig. Lower Bound Upper Bound
1 (Constant) .730 .043 16.937 .000 .646 .815
BANDC -1.54E-03 .001 -.055 -1.819 .069 -.003 .000
ATTITUDE -7.56E-04 .001 -.018 -.512 .609 -.004 .002
SUBNORM -7.71E-04 .001 -.031 -1.009 .313 -.002 .001
ASPIR 3.564E-Q3 .002 .040 1.524 .128 -.001 .008
NORMBELI 8.427E-04 .001 .038 1.168 .243
o
©
.002
DETERM 5.298E-03 .002 .091 2.548 .011 .001 .009
2 (Constant) .738 .045 16.253 .000 .649 .827
BANDC -1.51E-03 .001 -.054 -1.781 .075 -.003 .000
ATTITUDE -9.44E-04 .001 -.023 -.635 .526 -.004 .002
SUBNORM -8.32E-04 .001 -.033 -1.079 .281 -.002 .001
ASPIR 3.498E-03 .002 .039 1.490 .136 -.001 .008
NORMBELI 7.482E-04 .001 .034 1.028 .304
©
o
.002
DETERM 5.218E-03 .002 .089 2.506 .012 .001 .009
ACADINTG 9.635E-04 .001 .026 .988 .323 -.001 .003
OBSTACLE -3.73E-04 .001 -.012 -.457 .648 -.002 .001
3 (Constant) .788 .052 15.277 .000 .687 .889
BANDC -1.05E-03 .001 -.038 -1.233 .218 -.003 .001
ATTITUDE -5.33E-05 .002 -.001 -.035 .972 -.003 .003
SUBNORM -1.06E-03 .001 -.042 -1.359 .174 -.003 .000
ASPIR 2.064E-03 .002 .023 .861 .390 -.003 .007
NORMBELI 4.390E-04 .001 .020 .600 .548 -.001 .002
DETERM 5.869E-03 .002 .101 2.819 .005 .002 .010
ACADINTG 9.744E-04 .001 .027 1.002 .316 -.001 .003
OBSTACLE -1.74E-04 .001 -.005 -.208 .835 -.002 .001
GENDER
Q29 Age on December
-7.53E-03 .009 -.022 -.886 .376 -.024 .009
31 of this year
-1.09E-02 .003 -.097 -3.650 .000 -.017 -.005
ENGLISH -7.2GE-04 .001 -.019 -.690 .490 -.003 .001
4 (Constant) .761 .052 14.549 .000 .858 .863
BANDC -1.08E-03 ,001 -.039 -1.268 .205 -.003 .001
ATTITUDE -3.42E-04 .002 -.008 -.226 .821 -.003 .003
SUBNORM -1.15E-03 .001 -.046 -1.480 .139 -.003 .000
ASPIR 2.405E-03 .002 .027 1.004 .316 -.002 .007
NORMBELI 4.813E-04 .001 .022 .659 .510 -.001 .002
DETERM 6.060E-03 .002 .104 2.917 .004 .002 ,010
ACADINTG 8.392E-04 .001 .023 .864 .388 -.001 .003
OBSTACLE -1.96E-04 .001 -.006 -.235 .814 -.002 .001
GENDER
Q29 Age on December
-8.29E-03 .008 -.025 -.976 .329 -.025 .008
31 of this year
-1.15E-02 .003 -.102 -3.836 .000 -.017 -.006
ENGLISH -8.12E-04 .001 -.021 -.780 .436 -.003 .001
FACFT 9.969E-G4 .000 .075 2.962 .003 .000 .002
a- Dependent Variable: COMPPROP Academic success = (courses with C or better and P) / attempted
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
92
Table 15 (a)
Descriptive Statistics
Mean Std. Deviation N
COMPPROP Academic
success = (courses
with C or better and P) /
.83923 .16644 1567
attempted
BANDC 26.22 5.98 1567
ATTITUDE 35.70 4.06 1567
SUBNORM 14.54 6.68 1567
ASPIR 8.39 1.86 1567
NORMBELI 30.24 7.53 1567
DETERM 24.91 2.85 1567
ACADINTG 10.77 4.54 1567
OBSTACLE 16.03 5.20 1567
GENDER .58 .49 1567
Q29 Age on December
31 of this year
5.96 1.48 1567
ENGLISH 26.26 4.31 1567
FACFT 38.936 12.504 1567
FACPART 39.137 11.899 1567
Table 15 (b)
Model Sum m ary
C hange Statistics
Model R R Square
Adjusted
R Square
Std. Error of
the Estimate
R Square
C hange F C hange df1 df2 Sig. F Change
1 .111® .012 .008 .16573 .012 3.221 6 1560 .004
2 ,114b .013 .008 .16578 .001 .557 2 1558 .573
3 .149° .022 .015 .16517 .009 4.835 3 1555 .002
4 .166d .028 .020 .16476 .005 8.773 1 1554 .003
a. Predictors: (Constant), DETERM, SUBNORM, ASPIR, BANDC, NORMBELI, ATTITUDE
b. Predictors: (Constant), DETERM, SUBNORM, ASPIR, BANDC, NORMBELI, ATTITUDE, OBSTACLE, ACfi
c. Predictors: (Constant), DETERM, SUBNORM, ASPIR, BANDC, NORMBELI, ATTITUDE, OBSTACLE, ACA
0 2 9 Age on D ecem ber 31 of this year, ENGLISH
d. Predictors: (Constant), DETERM, SUBNORM, ASPIR, BANDC, NORMBELI, ATTITUDE, OBSTACLE, ACA
0 2 9 Age on D ecem ber 31 of this year, ENGLISH, FACFT
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
93
ANOVA6
Model
Sum of
Squares df Mean Square F Sig.
1 Regression .531 6 8.849E-02 3.221 .004a
Residual 42.849 1560 2.747E-02
Total 43.380 1566
2 Regression .562 8 7.019E-02 2.554 ,009b
Residual 42.819 1558 2.748E-02
Total 43.380 1566
3 Regression .957 11 8.703E-02 3.190 .000°
Residual 42.423 1555 2.728E-02
Total 43.380 1566
4 Regression 1.195 12 9.962E-02 3.670 .000d
Residual 42.185 1554 2.715E-02
Total 43.380 1566
a- Predictors: (Constant), DETERM, SUBNORM, ASPIR, BANDC, NORMBELI,
ATTITUDE
b. Predictors: (Constant), DETERM, SUBNORM, ASPIR, BANDC, NORMBELI,
ATTITUDE, OBSTACLE, ACADINTG
c- Predictors: (Constant), DETERM, SUBNORM, ASPIR, BANDC, NORMBELI,
ATTITUDE, OBSTACLE, ACADINTG, GENDER, Q29 Age on December 31 of this
year, ENGLISH
d- Predictors: (Constant), DETERM, SUBNORM, ASPIR, BANDC, NORMBELI,
ATTITUDE, OBSTACLE, ACADINTG, GENDER, Q29 Age on December 31 of this
year, ENGLISH, FACFT
e. Dependent Variable: COMPPROP Academic success = (courses with C or better
and P) / attempted
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9 4
CHAPTER V
SUMMARY CONCLUSIONS AND RECOMMENDATIONS
This chapter includes an introduction and review of the purpose of the study,
a summary of findings and conclusion regarding those findings including a
discussion about the significance of the investigation and finally, implications for
further research.
Introduction
The education gap for Latinos, the largest growing minority group in the
nation continues to widen at alarming rates. The literature confirms that a crisis of
unparalleled proportions exists created by a series of factors in place simultaneously
that when combined, produced decreased educational opportunity and a projected
circle of poverty for Latinos in this country. Declining economic trends in the areas
of employment and growth, the emergence of a knowledge based economy, eroding
fiscal support for all segments of education from Kindergarten to graduate schools
and policy retrenchment in the areas of affirmative action and targeted educational
outreach programs—all combine to portend serious outcomes not just for Latinos but
for the nation as a whole.
The Purpose of the Study
The purpose of this study is to identify any strategies to reduce the
achievement gap for Latinos focusing on community colleges as the most promising
avenue for this group. In particular, California community colleges are targeted
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9 5
because this segment alone enrolls one-third of all Latinos in public
postsecondary education in the entire nation and three-fourths of all Latinos in the
state. It is for this reason that community colleges represent the primary vehicle for
transfer to four- year colleges and universities and findings resulting from this study
have the potential of significant impact on future policy to stem this looming crisis.
Summary of Findings
Hypothesis 1: There are specific academic, environmental and social
integration variables that alone and in combination maximize academic success and
the preparation of Latino community college students for transfer.
The block entry regression table is the primary source in support of the
hypothesis that interactionist variables alone and in combination maximize student
success. Findings in this study indicate that of all the factors tested which form part
of the CCSL scales, only three variables were found to be statistically significant,
Determination, Age and proportion of minority faculty representation. Because the
model proved valid, the hypothesis is true within the limitation of these three key
variables.
Hypothesis 2: The level of Latino student representation on campus is
important to their level of academic success.
A one way analysis of variance (ANOVA) was performed on several
measures to test this hypothesis including a direct measure of student performance
for all Latino students followed by an analysis of enrollment patterns in remedial and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9 6
non-remedial course work for native speakers of English and non-native
speakers of English in direct correlation to the representational value of Latinos on
campus. Results of this investigation were consistent with the hypothesis that the
level of Latino student enrollment/representation on campus is positively correlated
with successful course completion, greater proportion of enrollment in transfer/non-
remedial courses and conversely lower levels of enrollment in remedial/precollegiate
English and math courses. In other words, the higher the proportion of Latinos on
campus the greater their opportunity for academic success. Specific support for this
conclusion is summarized below.
1. Course Completion and GPA
Specifically, Latino students in the HRV group showed significantly
greater proportion of course completion than did those in the LRV category
(p=0.0005). Likewise post hoc, pair wise comparison levels of RV indicate
significantly higher GPAs for the HRV and MRV group as compared to the
LRV group (p=0.0003 and p=0.0008 respectively).
2. Enrollment in Transfer and Remedial Gate Keeping Courses
English- An examination of course enrollment patterns in English and
math, also known as “gate keeping courses” was conducted through six
separate analysis of variance and four of the six analysis were found
significant at the 0.05 level. In the area of English, HRV and MRV colleges
enrolled significantly more Latinos in transfer/non-remedial courses than
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9 7
LRV institutions and there was no significant difference between MRV
and LRV colleges.
Math - In the area of math, LRV colleges were found to have a higher
proportion of Latinos that had never enrolled in any math courses in
comparison to HRV and MRV institutions colleges (p=0.005). An
examination of the proportion of Latinos enrolled in remedial math also
differed significantly by RV. In this instance, the data show that MRV
institutions enrolled the largest proportion of students in remedial math.
Finally, the general positive pattern of HRV colleges was sustained in the last
analysis as HRV colleges emerged again as the institutions with the largest
proportion of Latinos enrolled in transfer level math courses.
3. English Ability and Course Enrollment Patterns
Transfer English-The data showed a statistical difference based on
student language proficiency (native and non-native speakers of English). In
general, native speakers of English increased their likelihood of enrolling in
transfer level English courses. This pattern held through regardless of the RV
value of the institution.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9 8
Transfer Math—The extreme pattern of positive support for
academic success was confirmed once again under this measure as MRV and
HRV colleges were found to enroll the largest numbers of Latinos in transfer
level mathematics. And in this case, language proficiency had no significant
effect.
Remedial English-Language ability was the most important
determinant of the proportion of students enrolled in remedial English. As
expected, the proportion of students enrolled in remedial English was higher
for non-native speakers and lower for native speakers of English. In this
instance, however it is important to note that the RV value of the institution
was not found a significant factor.
Remedial Math—While there was a general pattern associated with
RV in which HRV colleges enrolled a smaller proportion of Latinos in
remedial math in comparison to MRV and LRV institutions the difference is
not statistically significant. Neither was language proficiency a significant
predictor. The conclusion then is that enrollment in remedial math is
independent of RV and language proficiency.
In conclusion, the hypothesis that Latino student representation on campus is
important to their level of academic success is proven valid. Of the twelve measures
linked to student success detailed above only three were found to have no significant
correlation to RV with the remaining nine measures resulting in a significant pattern
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
99
positively related to RV. In summary, student GPA, enrollment levels in
transfer courses and conversely low enrollment in remedial coursework all were
positively correlated with RV value. Equally telling is the profile of LRV
institutions were minority students did not perform well in most of the measures of
success.
An important observation to make is that English ability played a predictable
role in the proportion of students placed in transfer level English in comparison to
those placed in remedial English. In general, native English speakers increased the
likelihood of enrolling in one or more transfer level English course. Likewise, non
native speakers of English were found in greater proportion in remedial English
courses. The RV designation of the institutions however did not seem to play any
role in predicting placement in English courses. In the case of mathematics,
language proficiency was not a significant predictor of enrollment in transfer or
remedial level mathematics. In this instance the RV designation of the institution
was the key variable so that LRV institutions placed fewer students in transfer math
and the largest proportion of students in remedial math independent of language
proficiency. The converse was true for HRV and MRV institutions.
Hypothesis 3: The level of representation of faculty from ethnic minority
backgrounds is important to the academic success of Latino students.
The demographic tables introduced in Chapter 4, indicated that HRV and
MRV institutions employed a higher percentage of minority full time faculty than
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
LRV colleges. This factor is significant in that one of the few variables, which
remained as a strong predictor of student success in the block entry regression table,
was the proportion of minority faculty. Only two other factors were retained as
significant at the end of the regression namely, student age and determination. The
hypothesis therefore is proven true. The proportion of minority faculty is an
important predictor of student success for minority students. This factor is further
reaffirmed by the findings in the ANOVA and Two-way ANOVA analysis
developed in response to the second hypothesis in this study as described above.
Discussion
Construct Validity
The results of the principal component analysis clearly established that the
CCSL model mediated the relationship between key variables and student academic
success. In addition, the secondary analysis conducted in this study further submitted
the scales used from the CCSL model and the added variables to a reliability item
test which yielded a >.7 Cronbach’s Alpha.
Implications
Interactionist Factors Maximizing Student Success
This study found that while the predictive ability of the block regression
analysis was relatively low (0.024), nevertheless, the theoretical context of
interactionist factors that combine to predict academic success was valid and in this
instance yielded three critical factors, positively correlated as predictors of students
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
101
success for Latino students: student self determination, age (maturity) and the
proportion of minority faculty. The most important inference to be drawn from this
conclusion is that individual determinism and personal contact with minority faculty
along with student maturity are more important than language proficiency or the
proportion of peer minority student representation.
Value of Latino Representation and Peer Interaction
An exploration of the single variable of minority peer student interaction and
academic success left no doubt that the proportion of Latino students on campus was
positively correlated with improved academic success through a preponderance of
outcome measures. This finding supports the theoretical constructs associated with
campus climate.
Importance of Minority Faculty as Role Models
A significant outcome of this study is the reaffirmation of the importance of
the role of faculty to academic success and in this instance the role of minority
faculty for the academic success of Latino students. Ultimately, the regression
analysis retained the proportion of minority faculty as a more important variable than
the proportion of Latinos on campus. While this finding is not unexpected, it does
reaffirm the importance of minority role models for Latino students.
Unexpected Findings
Student determination and age as an important predictor of student success
supports the notion that pedagogy that makes use of these factors is needed. This
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0 2
finding is particularly important for community colleges where the average age
of students is 27.
Conclusion
Implications for Future Policy
This study presents a convincing set of evidence regarding the positive
effects of majority/minority institutions for Latino students. Peer interaction and the
attendant supportive campus climate do result in measurable outcomes of academic
success as gauged by this study. Community colleges as institutions of first choice
for Latinos at the state and national level are reaffirmed in this study as possessing a
critical characteristics in the ultimate preparation of these students for transfer,
namely the increasing demographics of minority representation on campus.
Likewise, the presence of minority faculty on camps is strongly linked to academic
success. Given the current retrenchment climate in the area of affirmative action, it
is important to argue the case of minority faculty hires in terms of sound pedagogical
practices linked directly to student outcome measures of academic excellence.
Student demographics linked to age of students is yet another characteristics
that can be explored in future policy discussions as a positive factor in the role that
community colleges play in the providing access to Latino students. The average
age of community college students is 27. Most of these students have families and
work full or part time. This very factor however proves positive in terms of student
success for Latinos. Policy considerations that focus on teaching adult learners
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
103
become significant for future consideration. The term androgogy in contrast to
pedagogy is being used with increased frequency in community college circles as is
the term “adult learners.” There is need for further exploration of the factors
associated with these premises as a promising avenue to maximize educational
opportunities for Latino community college student.
Peer interaction and a communities of learners - Based on the positive
findings in this study regarding the value of peer interaction, it becomes important to
consider the principles of learning communities designed to enhance academic
performance. Peer interaction within the formalized setting of learning communities
should prove fertile ground for maximizing student success.
Language barriers in context - This study would dispute that language
proficiency is the most critical factor for Latino underachievement. In fact, the
results obtained through the regression analysis did not retain English proficiency as
more important to academic success than the proportion of minority faculty or
student self-determination. And while it is very important to acquire English
proficiency, in the case of Latinos this is not the sole area and perhaps not even the
primary area for consideration in educational program planning if it is done in the
absence of the other factors found significant in this study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0 4
Recommendations for Future Research
As in any research study, findings always generate additional questions for
further research. In this instance, there are several areas of further study that should
be considered. Listed below are some of the more obvious areas of focus.
What are the differences in performance between Latino and white
community college students?
Does institutional racism play a part in the placement of Latino students in
remedial courses?
What role do minority faculty play in the revision and adaptation of
community college curriculum?
How are the needs of adult learners accommodated in the community college
instructional practices?
What educational program strategies should be designed to maximize the
positive correlation between individual motivation and determination and student
success?
What are the particular skills or competencies that may be identified for all
faculty members that build on the characteristics of minority faculty in relationship
to their relevance for minority students?
Limitations of the Study
1. This study is limited to the validity of the data collected by the TRUCCS
study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
105
2. It is limited in application to Latino student populations similarly
enrolled in urban community college campuses.
3. It is limited to the reliability of the instruments used.
It is within the limitations of the study and the context of the literature, which framed
this study that the policy implications and recommendations for future research are
offered. In particular, the generalizability of the study is limited to campuses and
student cohorts, which are most similar to those identified in the TRUCCS project.
Finally, given the demographic trends and projections for California and the rest of
the nation, it is expected that these findings will be increasingly applicable as the
demographic profiles in other areas become most like those in the Los Angeles
Community College District.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0 6
REFERENCES
Ajzen, I. & Fishbein, M. (1969). Prediction of goal-directed behavior: Attitude,
intention, and perceived behavioral control. Journal o f Experimental and
Social Psychology 22,453-474.
Ajzen, I. & Madden, T. (1986). The prediction of behavioral indicators in a choice
situation. Journal of Experimental Social Psychology 5,400-416.
Allen, W. & Solorzano, D. (2001). “Affirmative Action, Educational Equity and
Campus Racial Climate: a case Study of the University of Michigan Law
School.” La Raza Law Journal, 12, 237-363.
American Council on Education, Office of Minorities in Higher Education. (2001).
The continuing Significance of Racism: U.S. Colleges and Universities.
(Occasional Paper). Washington, D.C: Feagin, Joe R.
Appiah, A. (1994). Identity, authenticity, survival: Multicultural societies and social
reproduction, multiculturalism. A. Gutman. Princeton, Princeton University
Press: 149-164.
Bean, J.P. & Metzner, B.S. (1985). Interaction effects based on class level in an
explanatory model of college student dropout syndrome. American
Educational Research Journal, 22, 35-64.
Benjamin, M. (1994). The quality of student life: Toward a coherent
conceptualization. Social Indicators Research 31 (3), 205-264.
Bobo, Larry. (2001).Enduring two-ness. Public Perspective. Roper Center Review.
May/June, 12-16.
Burrell, L.F. (1981), Is There a Future for Black Students on Predominantly White
Campuses? Integrateducation ,18,(107-108), pp. 23-27.
Cabrera, A.F., Nora, a., and Castaneda, M.B. (1992). The convergence between two
theories of college persistence. Journal of Higher Education. 63: 143-164.
California Community Colleges (2002). General Information Web Site/About Us:
www. cccco. edu/about/about. him. Chancellor's Office of the California
Community Colleges.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
107
California Community Colleges Chancellor’s Office (2002 MONTH/ study
session on transfer measurement and capacity of receiving institutions.
(Agenda Item No. 7).
California Department of Education (2000). 12th grade graduates completing all
courses required for university of California and/or California state
university entrance. Sacramento, California: Department of Research.
California Department of Education (2001). California basic educational data
system / CBEDS (1997-1998). Sacramento, California: Department of
Research.
California Department of Education (July, 2002), High school graduation
requirements. [On-line]. Available: http://www.cde.ca.gov/shsd/hsgr
(7/12/2002).
California Department of Education (July, 2002), California school directory. [On
line] Available:
http://www.cde.ca.eov/schooldir and http://www. at-la. com/(a)la-edu/private. htm
(7/12/2002).
California Department of Finance (February, 2003). Demographic Research Unit.
Official dialogue for the current Master Plan, June 2002. [On-line].
Available: http://www.dof.ca.eov (2/5/03).
California Postsecondary Education Commission (1998). A master plan for higher
education in California, 1960-1975. Sacramento, California: CPEC Research
Unit.
California Postsecondary Education Commission (2000). Performance indicators of
California higher education, 1999. Sacramento, California: CPEC Research
Unit.
California Postsecondary Education Commission (2000). Student profiles (1999-
2000). Sacramento, California: CPEC Research Unit.
California Postsecondary Education Commission (2001). Higher education
enrollment demand, 1998-2010. Sacramento, California: CPEC Research
Unit.
California Postsecondary Education Commission (2001). On-line college guide to
the California community colleges, California state university, and university
of California systems. Sacramento, California: CPEC Research Unit.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
108
California Postsecondary Education Commission (2001). Questions and
answers about enrollment demand projections to 2010, Sacramento,
California: CPEC Research Unit.
California Postsecondary Education Commission (November, 2002). About the
Commission. [On-line], Available:
httn://www.cvec.ca.2ov/SecondPaees/CommissionHistorv.asD
Chancellor's Office California Community Colleges (2002, March 1). Transfer
capacity and readiness in the California community colleges. Sacramento,
California: Student Services & Special Programs Division and Technology,
Research, and Information Systems Division: 49.
Commission for the Review of the Master Plan for Higher Education (1987). The
master plan renewed: unity, equity, quality, and efficiency in California
postsecondary education. Sacramento, California: The California Legislature.
Community College League of California (2002). California community colleges:
pocket profile. Sacramento, California: Community College League of
California.
Conoley, Colleen & Parrisher, D. (1978). Minority/majority perceptions of guidance
services. Integrateducation. 16, (93) pp.37-39.
De los Santos, A.G. & Wright, I. (1990). Maricopa’s swirling students: earning one-
third of Arizona’s state’s bachelor’s degrees. Community, Technical, and
Junior College Journal, 60 (6), 32-34.
Delpit, L. (1995). Other people's children: Cultural conflict in the classroom. New
York: The New Press.
Educational Testing Service Policy Information Center (2002 MONTH). The closing
of the education frontier. (ISSUE NO. XXX). Princeton, New Jersey: Paul
Barton.
Fishbein, M. (1963). An investigation of the relationships between beliefs about an
object and the attitude toward that object. Human Relations 16(3), 233-239.
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior: An
introduction to theory and research. Reading. MA: Addisson-Wesley.
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An
introduction to theory and research. Reading, MA: Addison-Wesley.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
109
Gandara, P. (1995). Over the ivy walls: The educational mobility of low-
income Chicanos. New York, NY: SUNY Press.
Gandara, P. (1999). Staying in the race: The challenge for Chicanos/as in higher
education. The elusive quest for equality. Harvard Educational Review. J.
Morales.
Hagedom, L.S. & Castro, C.R. (1999). Paradoxes: California’s experience with
reverse transfer students. In B.K. Townsend (Ed.), Understanding the Impact
of Reverse Transfer Students on Community Colleges. Summer, 1999 ed.,
Vol. 106. pp. 15-26). San Francisco, CA: Jossey-Bass.
Hagedom, L.S. & Castro, C.R. (2000). Spending the summer at a California
community college. Academic Exchange Quarterly, 4(2), 23-32.
Hagedom, L.S., Maxwell, W., Rodriguez, P., Hocevar, D., & Fillpot, J. (2000). Peer
and student-faculty relations of women and of men in a community college.
Community College Journal of Research and Practice, 24(7), 587-598.
Hagedom, L.S., et al (2002). Assessing the interplay of the quality of student life and
retention of urban community college students. Presentation to the American
Educational Research Association (AERA). New Orleans, LA.
Henry, T.C. and Smith, G.P. (1993). Planning student success and persistence:
Implementing a state system strategy. 22: 27-35.
Jargowsky, Paul A. (1994) Take the money and run: Economic segregation in U.S.
metropolitan areas. Working paper. School of Social Sciences, University of
Texas at Dallas.
Los Angeles Almanac (October, 2002). Historical Demographic. [On-line].
Available:
http://www.losangelesalmanac.com/topics/Population/index.htm 10/7/2002
8:17 p.m.
Los Angeles Unified School District (July, 2002), Fingertip Facts 2001-2002 LAUSD. [On
line]. Available:
http://www.lausd.kl2.ca.us/lausd/offices/Office o f Communications/2000-
fineertip.htm (7/12/2002).
Lundquist, S (2002). Achieving equity and excellence in 21st century American
higher education: The California master plan and beyond. Unpublished
doctoral dissertation, Claremont University.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
110
Madden, M.J., Ellen, A.P.S., Ajzan, I. (1992). A comparison of the theory of
planned behavioral and the theory of reasoned action. Personality and Social
Psychology Bulletin. 198, (1), 3-9.
Pascarella, E. and P. T. Terenzini, Eds. (1991). How college affects students. San
Francisco, CA: Jossey-Bass.
Pascarella, E.T. (1980). Student-faculty informal outcomes. Review o f Educational
Research. 50: 545-595.
Pascarella, E.T., Edison, M. I. & Nora, A., Hagedom, L.S., Terenzini, P.T. (1998).
Does work inhibit cognitive development during the first year of college?
Educational Evaluation and Policy Analysis (EEPA Journal), 20(2), 75-93.
Pascarella, E.T., Edison, M.I., & Nora, A., Hagedom, L.S., Terenzini, P.T. (1998).
Does community college versus four-year college attendance influence
students’ educational plans? Journal of College Development 39(2), 179-
193.
Prather, George. (1998). Tidal wave II? reflections on the purpose and methodology
of enrollment projections. Poster session presented at the California
Association for Institutional Research Annual Conference, San Diego,
California.
Quinnan, T.W. (1997). Adult students at risk, culture bias in higher education.
Westport, Connecticut: Bergin & Garvey.
Reider E. Larry. (March, 2000) Kem County Superintendent’s recommendations for
the Master Plan. [On-line]. Available:
http://www.kem.org/mastemlan/financing.html
Richardson, R. and E. F. Skinner (1991). Achieving Q\quality and diversity:
Universities in a multicultural society. New York, NY: MacMillan.
Richardson, R. and L. Bender, Eds. (1987). Fostering minority access and
achievement in higher education. San Francisco, CA: Jossey-Bass.
Santos, A. d. 1 . and A. Rigual (1994). Progress of Hispanics in American higher
education. Minorities in higher education. Phoenix, AZ: Oryx Press.
Spady, W. G. (1970). Dropouts from higher education: An interdisciplinary review
and synthesis. Interchange, 19 (1), 109-121.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I l l
Spradley, P.A. (1996). A multiple variable analysis of the persistence of adult
African American male graduates from a baccalaureate degree program.
Unpublished doctoral dissertation, Columbia University, Teacher’s College.
State of California, Department of Finance (2001, March). Race/ethnic population
estimates: Components of change for California counties, April 1990 to July
1999. Sacramento, California.
State of California, Department of Finance (2003, January). County population
estimates and components of change, July 1, 2001-2002, with historical 2000
and 2001 estimates. Sacramento, California,.
State of California Joint Committee to Develop a Master Plan for Education (2002,
August). The California Master Plan for Education. Sacramento, California.
Terenzini, P.T. & Pascarella, E.T. (1980). Toward the validation of Tinto’s model of
college student attrition: A Review of recent studies. Research in Higher
Education, 12, 271-282.
The California Higher Education Policy Center (1995, September) Tidal wave II: An
evaluation of enrollment projections for California higher education
(Technical Report #95-6). Sacramento, California: Breneman, D., Estrada, L.
and Hayward, G.
The Center for Studies in Higher Education. The history of the California master
plan for higher education. [On-line]. Available:
http://sunsite.berkelev.edu/uchistorv/archives/masterplan/index.html
(9/26/02)
The Federal Reserve Bank of New York (1995, January). The future path and
consequences of the U.S. earnings/education gap. Economic Policy Review,
(39-41) New York, NY: Levy, F.
The U.S. Department of Education (1999). The condition of education. Washington,
D.C.: National Center for Education Statistics.
Tinto, V. (1975). Dropout from higher education: A theoretical synthesis of recent
research. Review of Educational Research. 45 (1), 89-123.
University of California, Office of the President (February, 2002). Guide to “a-g”
Requirements and Instruction for Updating.[On-line]. Available:
www.ucop.edu/a-gGuide
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
112
U. S. Census Bureau (1998). Current population survey, The American Council
on Education, Re-printed as an ACE Fact Sheet.
U. S. Census Bureau (2000) Current population survey. United way of greater Los
Angeles 1999. State of the County Report. Los Angeles, C A .
U. S. Office of Civil Rights (1988-1994). Elementary and secondary school civil
rights compliance reports. Washington, D.C.: U.S. Office of Civil Rights.
U.S. Department of Education (1996). Math assessment, table B-6. Washington,
D.C. The National Center for Education Statistics (NAEP).
U.S. Department of Education (1996). Math tables. Washington, D.C., The National
Assessment of Educational Progress (NAEP).
U.S. Department of Education (1998). National assessment of educational progress:
Math assessment, table B-6. Washington, D.C.: The National Center for
Education Statistics.
U.S. Department of Education (1998, 2000-2002). Annual educational statistics by
age and race. Washington, D.C.: The National Center for Education
Statistics, Office of Educational Research and Improvement.
Vallerand, R.Deshaies, P., Cuerrier, J., Pelletier, L. & Mongeau, C. (1992). Ajzen’s
and Fishbein’s theory of reasoned action as applied to moral behavior: A
confirmatory analysis. Journal of Personality and Social Psychology 62(1),
98-109.
Vella, J. (1994). Learning to listen, learning to teach: the power of dialogue in
educating adults. San Francisco, California: Jossey-Bass.
Wilds, D. and R. Wilson (2000). Minorities in higher education: Seventeenth annual
status report. Washington, D.C.: American Council on Education.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX A
Community College Student Life and Course Completion
Model Factorial Validation Tables
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 1. Items and Scales used in the model
Typology Category Scale name and Cronbach’s
Alpha
Items Range, Mean (standard
deviation)
Conditioning Campus Pov erty Level The percentage of people in the campus Mean 18.3332
district below the poverty level as Std. 8.1478
determined by the 1990 Census Deviation
Range 24.20
Minimum 8.20
Maximum 32.40
Age How old will you be on December 31 of this Mean 6.28
year? Std. 1.74
1= 16 years or younger, Deviation
2=17, Range 9
3=18, . . . Minimum 1
10 55 or older Maximum 10
Gender Your gender: Mean 1.62
l=Male Std. .48
2=Female Deviation
Range 1
Minimum 1
Maximum 2
Children How manv children/stepchildren are living in Mean 1.42
vour household? Std. .69
l=None Deviation
2=1-2 Range 3
3=3-4 Minimum 1
4=5 or more Maximum 4
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
English Ability
Alpha = .9231
Ability in English (1= not at all, 2=with
difficulty, 3= fairly well, 4=very well)
Read
Write
Understand a college lecture
Read a college text book
Write an essay exam
Write a term paper
Participate in class discussions
Communicate with instructor
Mean 3.2907
Std. .5937
Deviation
Range 3.00
Minimum 1.00
Maximum 4.00
Independent High school GPA Self reported average grade in high school?
1=D or lower (Poor)
2=C- (Below Average)
3=C (Average)
4=C+ (Above Average)
5= B- (Good)
6= B (Very Good)
7= B+ (Excellent)
8= A- (Superior Quality)
9 -A o r A + (Extraordinary)__________
Mean 5.51
Std. 1.82
Deviation
Range 8
Minimum 1
Maximum 9
College GPA Grade point average for Spring 2001 from
LACCD transcripts.
Mean
Std.
Deviation
Range
Minimum
Maximum
Academic Integration
Alpha = .7473
How often or how many times
Talk w/instructor before or after class
Talk with instructor during office hours
Help another student understand
homework
Study in small groups outside o f class
Speak with an academic counselor
Telephone/email /student about studies
2.6605
.9487
4.00
.00
4.00
Mean 1.9033
Std. .8498
Deviation
Range 5.00
Minimum 1.00
Maximum 6.00
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Obstacles
Alpha = .7123
How large a problem (1= not a problem to
5=very large problem).
Parking
Transportation
Family responsibilities
Job-related responsibilities
Paying for college
Scheduling classes for next semester
Understanding the English language
Mean 2.0132
Std. .7042
Deviation
Range 4.00
Minimum 1.00
Maximum 5.00
Mediating Normative Beliefs Reasons for coming (1= very unimportant to Mean 2.956E-02
Alpha = .7781 7— very important) Std. .6391
Z score: This college has good social Deviation
activities Range 5.06
Z score: Offers special educational Minimum -3.39
programs Maximum 1.67
Z score: This college has a good
reputation
What people think about the college (1=
poor college to 4=excellent college)
Z score: You
Z score: Your closest friends
Z score: Your spouse or partner
Z score: Your parents or guardians
Z score: Your other relatives
Beliefs and Influences on decision to come to the Mean 5.2207
consequences particular college (1= very unimportant to 7= Std. 1.3394
Alpha = .7093 very important): Deviation
Graduates get good jobs Range 6.00
Students transfer to good 4-yr schools Minimum 1.00
To get a better job Maximum 7.00
To get a college degree
To enroll in a special program or
certificate
O i
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Academic Attitude
Alpha = .7301
l=Strongly disagree to 7= Strongly agree
Understanding what is taught is
important
I always complete homework
assignments
Success in college largely due to effort
I can learn all skills taught in college
Enjoy challenging class assignments
Expect to do well/earn good grades
Mean 5.9600
Std. .7355
Deviation
Range 6.00
Minimum 1.00
Maximum 7.00
Subjective Norm
Alpha = .7378
Why in college (1= very unimportant to 7 =
very important):
My parents wanted me to come
Other family member wanted me to
come
HS or other counselor advised me
My friends are attending here
My employer encouraged me to enroll
here
Mean 2.9964
Std. 1.4490
Deviation
Range 6.00
Minimum 1.00
Maximum 7.00
# weekly hours employment 1= none to 9= 46 hours or more Mean
Std.
Deviation
Range
Minimum
Maximum
Student Self Perception How do you think o f yourself?
0= Primarily as a parent who is going
to college, or Primarily as an employee
who is going to college
1= Solely as a student, or Primarily as a
student who is employed_____________
5.28
2.89
Mean .5750
Std. .4944
Deviation
Range 1.00
Minimum .00
Maximum 1.00
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Goal Orientation
Alpha =.7395
l=Strongly disagree to 7= Strongly agree
Satisfied when I work hard to achieve
I am very determined to reach my goals
Impt to finish courses in pgm o f studies
Keep trying even when frustrated by
task
Mean 6.2231
Std. .7588
Deviation
Range 6.00
Minimum 1.00
Maximum 7.00
Aspirations
Alpha = .7240
As things stand, do you think you will (1=
definitely not to 5= definitely)
Transfer to a 4-year college/university
Get a bachelor's degree
Mean 4.1577
Std. .9954
Deviation
Range 4.00
Minimum 1.00
Maximum 5.00
Feelings o f belonging I feel I belong at this college (l=Strongly
disagree to 7= Strongly agree)
Mean 5.05
Std. 1.50
Deviation
Range 6
Minimum 1
Maximum 7
Dependent Student life Latent Construct
Course completion Ratio of courses completed with a C or better
(or pass) divided by the number o f courses
enrolled (Spring 2001)
Mean .8825
Std. .2521
Deviation
Range 1.00
Minimum .00
Maximum 1.00
119
Table 2
Overall Fit Indices and Parsimony Fit Indices for the Model of Community College Life
and Retention
Overall Fit Indices
Overall Model Fit NFI RFI IFI TLI
.900 .900 .844 .900 .845
Parsimony Fit Indices.
MODEL PRATIO PNFI PCFI
Tested Model .642 .578 .578
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
120
Table 3
Squared Multiple Correlations
M iinniwijpaa— ■ — — ■
Endogenous Constructs
Squared Multiple Correlations
Academic Attitude
0.614
academic integration
0.143
Aspirations 0.466
Beliefs and consequences 0.205
Children 0.103
College GPA
0.264
English Ability 0.306
Feelings o f Belonging 0.167
Goal orientation 0.856
High school GPA 0.343
Normative Beliefs 0.008
Number o f hours o f work 0.014
Obstacles 0.156
Student self perception 0.148
Subjective Norms 0.313
Dependent Variables
Student Life 0.399
Course Completion 0.511
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 4
Standardized Direct Paths to all Endogenous Variables in the Community College Model of Student Life and Course
Completion
Age Gender Poverty children English Ability
#
Hrs. Work Norm Belief
Student
Self-Concept
children 0.289 0.112 0 0 0 0 0 0
English Ability 0 0.458 0.291 0 0 0 0 0
# Hrs. Work 0 -0.045 0 0.023 0.131 0 0 0
Norm Belief -0.05 0.054 0.058 0 0 0 0 0
Student
Self-Concept 0 0 0 -0.33 0.093 -0.19 0 0
Belief & Conseq 0.125 0 0.057 0 0.079 0 0.433 0
Belonging 0.226 -0.002 0 0 0.092 0 0.329 0
Acad Attitude 0.252 0.151 0 0 0.583 0 0 0.133
Subjctv Norm 0 0 0 0 -0.32 0 0.472 0
HSGPA 0 0.258 0 0 0.418 0 0 0
Goal Orient 0 0 0 0 0.138 0 0 0
Acad Intgrtn 0 0.071 0 0 0.264 0 0 0.05
Aspire 0.068 0 0 0.001 0.363 0 0 0.106
College GPA 0.196 0 0 0 0 0 0 0
Obstacles 0 0 0 0.059 -0.389 0.122 0 0
Student Life 0.016 0 0 0 0.06 0 0 0
Course
Completion 0 0 0 0 0 0 0
0
to
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 4 Continued.
Standardized Direct Paths to all Endogenous Variables in the Community College Model o f Student Life and Course Completion
Belief &
Conseq Belonging
Acad
Attitude
Subjctv
Norm HSGPA Goal Orient Acad Intgrtn Aspire
College
GPA Obstacles
Children 0 0 0 0 0 0 0 0 0 0
English
Ability 0 0 0 0 0 0 0 0 0 0
# Hrs. Work 0 0 0 0 0 0 0 0 0 0
Norm Belief 0 0 0 0 0 0 0 0 0 0
Student Self-
Concept 0 0 0 0 0 0 0 0 0 0
B elief &
Conseq 0 0 0 0 0 0 0 0 0 0
Belonging 0 0 0 0 0 0 0 0 0 0
Acad Attitude 0.24 0 0 0 0 0 0 0 0 0
Subjctv Norm 0 0 0 0 0 0 0 0 0 0
HSGPA 0 0 0 0 0 0 0 0 0 0
Goal Orient 0 0 0.827 0 0 0 0 0 0 0
Acad Intgrtn 0 0.128 0 0.237 0 0 0 0 0 0
Aspire 0 0 0.374 0 0 0 0 0 0 0
College GPA 0 0 0.218 0 0.198 0.06 -0.031 0.063 0 0
Obstacles 0 0 0 0 0 0 0 0 0 0
Student Life 0 0 0 0 0 0 0 0 0.135 -0.571
Course
Completion 0 0 0 0 0 0 0 0 0 0.541
123
Table 5a
Standardized Direct. Indirect, and Total Effects on Student Life
Construct Direct Effect Indirect Effect Total Effect
Academic Attitude
0 0.039 0.039
Academic
Integration
0 -0.004 -0.004
Age 0.016ns 0.026 0.042
Aspirations 0 0.009 0.009
Beliefs and
Consequences
0 0.009 0.009
Children 0 -0.037 -0.037
College GPA 0.135 0 0.135
English Ability 0.06 0.25 0.31
Feelings of
Belonging 0 -0.001
-0.001
Gender 0 0.154 0.154
Goal Orientation
0 0.008
0.008
High School GPA
0 0.027
0.027
Normative Beliefs 0 0.003 0.003
Number hrs work 0 -0.071 -0.071
Obstacles -0.571 0 -0.571
Poverty 0 0.091 0.091
Student Self
Perception
0 0.006 0.006
Subjective Norms 0 -0.001 -0.001
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
124
Table 5b
Standardized Direct. Indirect, and Total Effects on Course Completion
Construct Direct Effect Indirect Effect Total Effect
Academic attitude
0 0.036
0.036
Academic Integration
-0.004 -0.004
Age 0 0.048 0.048
Aspirations 0
0.008 0.008
Beliefs and Consequences
0 0.009
0.009
Children 0 0 0
College GPA 0 0.122 0.122
English Ability 0 0.078 0.078
Feelings o f Belonging
0 0
0
Gender 0 0.047 0.047
Goal Orientation
0 0.007
0.007
High School GPA
0 0.024
0.024
Normative Beliefs. 0 0.003 0.003
Number hrs work 0 0.002 0.002
Obstacles 0.541 -0.515 0.026
Poverty 0 0.023 0.023
Student self perception 0 0.005
0.005
Subjective norms
0 -0.001
-0.001
Student Life
.90 0
.90
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
125
APPENDIX B
Data Analysis Descriptive Statistical Output Tables
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
126
Table 11
One-Wav ANOVA of proportion of courses completed with C or better by RV
Descriptives
N Mean Std. Deviation Std. Error
COMPPROP Academic 1 High 1492 .84203 .16131 4.18E-03
success = (courses
2 Moderate 601 .83774 .16492 6.73E-03
with C or better and P) /
3 Low 226 .81881 .19139 1.27E-02
attempted
Total 2319 .83865 .16546 3.44E-03
GPA 1 High 1576 2.4242 .7842 1.975E-02
2 Moderate 608 2.4230 .7935 3.218E-02
3 Low 244 2.2626 .8559 5.480E-02
Total 2428 2.4076 .7952 1.614E-02
ANOVA
Sum of
Squares df Mean Square F Sig.
COMPPROP Academic
success = (courses
with C or better and P) /
attempted
Between Groups
Within Groups
Total
.106
63.356
63.463
2
2316
2318
5.322E-02
2.736E-02
1.945 .143
GPA Between Groups 5.710 2 2.855 4.528 .0 1 1
Within Groups 1528.859 2425 .630
Total 1534.568 2427
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 2 7
Multiple C om parisons
D ependent Variable (DRV (J)R V
Mean
Difference
(l-J) Std. Error Sig.
95% Confidence Interval
.ower Bounc Jpper Bounc
COM PPROP Acad LSD 1 High 2 Modera
su ccess = (courses 3 Low
2909E-03
3213E-Q2*
.99E-03
.18E-02
.591
.049
13789E-02
I.1780E-05
.9961E-02
L6364E-02
with C or better and 2 M odera i High
attem pted 3 Low
.291E-03
3922E-02
.99E-03
.29E-02
.591
.143
99607E-02
38613E-03
.1379E-02
L4230E-02
3 Low 1 High
2 Modera
1,321 E-02*
.892E-02
.18E-02
.29E-02
.049
.143
63642E-02
42304E-02
17798E-05
3.3861 E-03
Bonferror 1 High 2 Modera
3 Low
2909E-03
3213E-02
.99E-03
.18E-02
1.000
.148
48528E-02
07061 E-03
’.3435E-02
S.1497E-02
2 Modera 1 High
3 Low
.291 E-03
3922E-02
’ .99E-03
.29E-02
1.000
.428
34345E-02
19967E-02
4853E-02
1.9841 E-02
3 Low 1 High
2 Modera
.321 E-02
.892E-02
.18E-02
.29E-02
.148
.428
14966E-02
98410E-02
i.0706E-03
L1997E-02
GPA LSD 1 High 2 Modera
3 Low
.125E-03
.1616*
791 E-02
462E-02
.976
.003
’.3210E-02
5.448E-02
7.546E-02
.2687
2 Modera 1 High
3 Low
1246E-03
.1605*
791 E-02
017E-O2
.976
.008
'.5459E-02
4.247E-02
7.321 E-02
.2785
3 Low 1 High
2 Modera
-.1616*
-.1605*
462E-02
317E-02
.003
.008
-.2687
-.2785
I.4478E-02
I.2474E-02
Bonferror 1 High 2 Modera
3 Low
.125E-03
.1616*
791E-02
462E-02
1.000
.009
I.9688E-02
3.073E-02
9.194E-02
.2925
2 Modera 1 High
3 Low
1246E-03
.1605*
791E-02
D17E-02
1.000
.023
I.1937E-02
1.632E-02
8.969E-02
.3046
3 Low 1 High
2 Modera
-.1616*
-.1605*
462E-02
D17E-02
.009
.023
-.2925
-.3046
I.0733E-02
L6317E-02
'•The m ean difference is significant at the .05 level.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
128
o
O
.85.
a >
c
05
C D ??
%
e
H
8
< D "O
O g
O S
S |
u o
c ^
c 05
0 ) w
T J X
< i
O to
.84.
.83.
.82.
.8 1
High
" " V
V
X .
X
\
\
X .
X.
X .
x„
X ..
X .
X .
X .
X.
X
X
Moderate
RV
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission
Low
1 2 9
One-way ANOVA: Course completion by RV
Descriptive*
COMPLETE
N Mean Std. Deviation Std. Error
95% Confidence Interval for
Mean
Minimum Maximum Lower Bound Upper Bound
1 High 1492 13.6180 10.4592 .2708 13.0868 14.1491 1.00 74.00
2 Moderate 601 14.7537 10.4811 .4275 13.9141 15.5934 1.00 70.00
3 Low 226 9.9823 8.3777 .5573 8.8841 11.0805 1.00 45.00
Total 2319 13.5580 10.3553 .2150 13.1363 13.9797 1.00 74.00
ANOVA
COMPLETE
Sum of
Squares df Mean Square F Sig.
Between Groups 3754.227 2 1877.114 17.758 .000
Within Groups 244809.7 2316 105.704
Total 248563.9 2318
Multiple Comparisons
Dependent Variable: COMPLETE
LSD
(DRV (J) RV
Mean
Difference
d-J)
Std. Error Sig.
95% Confidence Interval
Lower Bound Upper Bound
1 High 2 Moderate -1.1358* .4967 .022 -2.1098 -.1617
3 Low 3.6357* .7339 .000 2.1966 5.0748
2 Moderate 1 High 1.1358* .4967 .022 .1617 2.1098
3 Low 4.7714* .8022 .000 3.1983 6.3446
3 Low 1 High -3.6357* .7339 .000 -5.0748 -2.1966
2 Moderate -4.7714* .8022 .000 -6.3446 -3.1983
*• The mean difference is significant at the .05 level.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
130
Table 12: One-way ANOVAs: English and Math
(non-taken, remedial, non-remedial) by RV
Descriptivss
N Mean Std. Deviation Std. Error
95% Confidence Interval for
Mean
Minimum Maximum Lower Bound Upper Bound
ENGaNYI No English 1 High 1668 .18 .39 9.50E-03 .17 .20 0 1
courses attempted
2 Moderate 647 .15 .36 1.42E-02 .13 .18 0 1
3 Low 255 .20 .40 2.51 E-02 .15 .25 0 1
Total 2570 .18 .38 7.55E-03 .16 .19 0 1
ENGRM1 One or more 1 High 1668 .49 .50 1.22E-02 .47 .52 0 1
remedial English courses
2 Moderate 647 .51 .50 1.97E-02 .47 .55 0 1
3 Low 255 .57 .50 3.10E-02 .51 .63 0 1
Total 2570 .51 .50 9.86E-03 .49 .53 0 1
ENGNR1 Non-remedial 1 High 1668 .73 .44 1.09E-02 .71 .75 0 1
English courses and
2 Moderate 647 .77 .42 1.65E-02 .74 .80 0 1
remedial courses
3 Low
255 .62 .49 3.05E-02 .56 .68 0 1
Total
2570 .73 .44 8.78E-03 .71 .75 0 1
M THANY1 No Math 1 High 1668 .24 .43 1.05E-02 .22 .26 0 1
courses attempted
2 Moderate 647 .25 .43 1.70E-02 .21 .28 0 1
3 Low 255 .33 .47 2.95E-02 .27 .39 0 1
Total
2570 .25 .43 8.55E-03 .23 .27 0 1
MATHREMt One or more 1 High 1668 .37 .48 1.18E-02 .34 .39 0 1
remedial Math courses
2 Moderate 647 .42 .49 1.94E-02 .38 .46 0 1
3 Low 255 .34 .47 2.97E-02 .28 .40 0 1
Total 2570 .38 .48 9.56E-03 .36 .40 0 1
MATHNR2 Non-remedial 1 High 1668 .39 .49 1.2QE-02 .37 .42 0 1
Math Courses only
2 Moderate 647 .34 .47 1.86E-02 .30 .37 0 1
3 Low 255 .33 .47 2.96E-02 .28 .39 0 1
Total 2570 .37 .48 9.54E-03 .35 .39 0 1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
131
ANOVA
Sum of
Squares df Mean Square F Sig.
ENGANY1 No English Between Grouf
courses attempted Within Groups
Total
.601
375.779
376.380
2
2567
2569
.301
.146
2.053 .129
ENGRM1 One or mort Between Grout
remedial English cours within Groups
Total
1.380
641.033
642.412
2
2567
2569
.690
.250
2.763 .063
ENGNR2 Non-remedi; Between Group
English Courses only within Groups
Total
2.331
553.115
555.446
2
2567
2569
1.165
.215
5.408 .005
MTHANY1 No Math Between Grout
courses attempted within Groups
Total
1.764
481.359
483.123
2
2567
2569
.882
.188
4.703 .009
MATHREM1 One or rr Between Group
remedial Math courses within Groups
Total
1.624
601.775
603.399
2
2567
2569
.812
.234
3.463 .032
MATHNR2 Non-remec Between Group
Math Courses only within Groups
Total
1.962
598.677
600.639
2
2567
2569
.981
.233
4.205 .015
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
132
One-way ANOVA: GPA by RV
Descriptive*
GPA
N Mean Std. Deviation Std. Error
95% Confidence Interval for
Mean
Minimum Maximum Lower Bound Upper Bound
1 High 1576 2.4242 .7842 I.975E-02 2.3854 2.4629 .00 4.00
2 Moderate 608 2.4230 .7935 3.218E-02 2.3598 2.4862 .00 4.00
3 Low 244 2.2626 .8559 5.480E-02 2.1546 2.3705 .00 4.00
Total 2428 2.4076 .7952 I.614E-02 2.3760 2.4393 .00 4.00
ANOVA
GPA
Sum of
Squares df Mean Square F Sig.
Between Groups 5.710 2 2.855 4.528 .011
Within Groups 1528.859 2425 .630
Total 1534.568 2427
Multiple Comparisons
Dependent Variable: GPA
LSD
(I) RV (J) RV
Mean
Difference
d-J)
Std. Error Sig.
95% Confidence Interval
Lower Bound Upper Bound
1 High 2 Moderate 1.125E-03 3.791 E-02 .976 -7.3210E-02 7.546E-02
3 Low .1616* 5.462E-02 .003 5.448E-02 .2687
2 Moderate 1 High -1.1246E-03 3.791 E-02 .976 -7.5459E-02 7.321E-02
3 Low .1605* 6.017E-02 .008 4.247E-02 .2785
3 Low 1 High
-.1616* 5.462E-02 .003 -.2687 -5.4478E-02
2 Moderate -.1605* 6.017E-02 .008 -.2785 -4.2474E-02
* ■ The mean difference is significant at the .05 level.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LSD
Multiple Com parisons
Dependent Variable (I) RV
(J) RV
Mean
Difference
(l-J) Std. Error Sig.
ENGANY1 No English 1 High 2 Moderate 3.16E-02 1.77E-02 .074
courses attempted
3 Low -1 .53E-02 2.57E-02 .551
2 Moderate 1 High -3.16E-02 1.77E-02 .074
3 Low -4.70E-02 2.83E-02 .097
3 Low 1 High 1.53E-02 2.57E-02 .551
2 Moderate 4.70E-02 2.83E-02 .097
ENGRM1 One or more 1 High 2 Moderate -1.60E-02 2.31 E-02 .488
remedial English courses
3 Low -7.85E-02* 3.36E-02 .019
2 Moderate 1 High 1.60E-02 2.31 E-02 .488
3 Low -6.25E-02 3.69E-02 .091
3 Low 1 High 7.85E-02* 3.36E-02 .019
2 Moderate 6.25E-02 3.69E-02 .091
ENGNR2 Non-remedial 1 High 2 Moderate -1.56E-02 2.15E-02 .468
English Courses only
3 Low 9.39E-02* 3.12E-02 .003
2 Moderate 1 High 1.56E-02 2.15E-Q2 .468
3 Low .11* 3.43E-02 .001
3 Low 1 High -9.39E-02* 3.12E-02 .003
2 Moderate -.11* 3.43E-02 .001
MTHANY1 No Math 1 High 2 Moderate -6.89E-03 2.01 E-02 .731
courses attempted
3 Low -8.90E-02* 2.91 E-02 .002
2 Moderate 1 High 6.89E-03 2.01 E-02 .731
3 Low -8.21 E-02* 3.20E-02 .010
3 Low 1 High 8.90E-02* 2.91 E-02 .002
2 Moderate 8.21 E-02* 3.20E-02 .010
MATHREM1 One or more 1 High 2 Moderate -5.04E-02* 2.24E-02 .025
remedial Math courses
3 Low 2.97E-02 3.26E-02 .362
2 Moderate 1 High 5.Q4E-02* 2.24E-02 .025
3 Low 8.01 E-02* 3.58E-02 .025
3 Low 1 High
-2.97E-02 3.26E-02 .362
2 Moderate -8.01 E-02* 3.58E-02 .025
MATHNR2 Non-remedial 1 High 2 Moderate 5.73E-02* 2.24E-02 .010
Math Courses only
3 Low 5.94E-02 3.25E-02 .068
2 Moderate 1 High -5.73E-02* 2.24E-02 .010
3 Low 2.06E-03 3.57E-02 .954
3 Low 1 High -5.94E-02 3.25E-02 .068
2 Moderate -2.06E-03 3.57E-02 .954
* • The mean difference is significant at the .05 level.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 13 Two-Way ANOVA (Transfer English by RV
and Native Language)
Between-Subjects Factors
Value Label N
RV 1 High 1569
2 Moderate 616
3 Low 245
Q20 Is English your 1 No 1709
native language?
2 Yes 721
Descriptive Statistics
Dependent Variable: ENGTRAN1 One or more English transfer courses
RV
Q20 Is English your
Mean Std. Deviation N
1 High 1 No .58 .49 1108
2 Yes .63 .48 461
Total .59 .49 1569
2 Moderate 1 No .59 .49 430
2 Yes .68 .47 186
Total .62 .49 616
3 Low 1 No .49 .50 171
2 Yes .59 .49 74
Total
.52 .50 245
Total 1 No .57 .49 1709
2 Yes .64 .48 721
Total
.59 .49 2430
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Tests of Between-Subjects Effects
Dependent Variable: ENGTRAN1 One or more English transfer courses
Source
Type III Sum
of Squares df Mean Square F Sig.
Corrected Model 4.321a 5 .864 3.597 .003
Intercept 420.900 1 420.900 1751.983 .000
RV 1.344 2 .672 2.796 .061
Q20 2.064 1 2.064 8.591 .003
RV * Q20 .171 2 8.551 E-02 .356 .701
Error 582.346 2424 .240
Total 1440.000 2430
Corrected Total 586.667 2429
a- R Squared = .007 (Adjusted R Squared = .005)
1. Grand Mean
Dependent Variable: ENGTRAN1 One or more English
transfer courses
Mean Std. Error
95% Confidence Interval
Lower Bound Upper Bound
.594 .014 .566 .621
Estimates
Dependent Variable: ENGTRAN1 One or more English transfer
courses
RV Mean Std. Error
95% Confidence Interval
Lower Bound Upper Bound
1 High
.606 .014 .579 .632
2 Moderate .635 .022 .593 .677
3 Low .540 .034 .473 .607
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
136
Pairwise Comparisons
Dependent Variable: ENGTRAN1 One or more English transfer courses
(I)RV (J) RV
Mean
Difference
d-J)
Std. Error Sig.a
95% Confidence Interval for
Difference8
Lower Bound Upper Bound
1 High 2 Moderate -2.971 E-02 .025 .243 -7.959E-02 2.017E-02
3 Low 6.552E-02 .037 .074 -6.455E-03 .138
2 Moderate 1 High 2.971 E-02 .025 .243 -2.017E-02 7.959E-02
3 Low 9.523E-02* .040 .018 1.618E-02 .174
3 Low 1 High -6.552E-02 .037 .074 -.138 6.455E-03
2 Moderate -9.523E-02* .040 .018 -.174 -1.618E-02
Based on estimated marginal means
* The mean difference is significant at the .05 level.
s. Adjustment for multiple comparisons: Least Significant Difference (equivalent to no
adjustments).
Univariate Tests
Dependent Variable: ENGTRAN1 One or more English transfer courses
Sum of
Squares df Mean Square F Sig.
Contrast 1.344 2 .672 2.796 .061
Error 582.346 2424 .240
The F tests the effect of RV. This test is based on the linearly independent
pairwise comparisons among the estimated marginal means.
Estimates
Dependent Variable: ENGTRAN1 One or more English transfer courses
Is English your 95% Confidence Interval
native language? Mean Std. Error Lower Bound Upper Bound
1 No
.552 .016 .521 .583
2 Yes
.635 .024 .589 .682
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
137
Pairwise Comparisons
Dependent Variable: ENGTRAN1 One or more English transfer courses
(I) Is English yoi (J) Is English yot
native language native language'
Mean
Difference
(W)
Std. Error Sig.a
6% Confidence Interval fo
Differenci
_ower Bound JpperBound
1 No 2 Yes 8.313E-02* .028 .003 -.139 -2.752E-02
2 Yes 1 No 3.313E-02* .028 .003 2.752E-02 .139
Based on estimated marginal means
‘ •The mean difference is significant at the .05 level.
a. Adjustment for multiple comparisons: Least Significant Difference (equivalent to no adjust
Univariate Tests
Dependent Variable: ENGTRAN1 One or more English transfer courses
Sum of
Squares df Mean Square F Sig.
Contrast 2.064 1 2.064 8.591 .003
Error 582.346 2424 .240
The F tests the effect of Is English your native language?. This test is based on
the linearly independent pairwise comparisons among the estimated marginal
means.
4. RV * Is English your native language?
Dependent Variable: ENGTRAN1 One or more English transfer courses
RV
Is English your
native language? Mean Std. Error
95% Confidence Interval
Lower Bound Upper Bound
1 High 1 No
.578 .015 .549 .606
2 Yes
.633 .023 .589 .678
2 Moderate 1 No .593 .024 .547 .639
2 Yes
.677 .036 .607 .748
3 Low 1 No
.485 .037 .412 .559
2 Yes
.595 .057 .483 .706
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
138
Multiple Comparisons
Dependent Variable: ENGTRAN1 One or more English transfer courses
LSD
(I) RV (J) RV
Mean
Difference
d-J)
Std. Error Sig.
95% Confidence Interval
Lower Bound Upper Bound
1 High 2 Moderate -2.45E-02 2.33E-02 .293 -7.02E-02 2.12E-02
3 Low 7.56E-02* 3.37E-02 .025 9.62E-03 .14
2 Moderate 1 High 2.45E-02 2.33E-02 .293 -2.12E-02 7.02E-02
3 Low .10* 3.70E-02 .007 2.75E-02 .17
3 Low 1 High -7.56 E-02* 3.37E-02 .025 -.14 -9.62E-03
2 Moderate -.10* 3.70E-02 .007 -.17 -2.75E-02
Based on observed means.
*• The mean difference is significant at the .05 level.
Estimated Marginal Means of One or more English transfer courses
ro
I
5
Is English your native
slanguage
T 3
0 }
1 5
E
' 4 =
C f l
UJ
No
High Moderate Low
RV
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
139
Table 13 Two-way ANOVA (Transfer level Math by RV
and Native Language)
Between-Subjects Factors
Value Label N
RV 1 High 1569
2 Moderate 616
3 Low 245
Q20 Is English your 1 No 1709
native language?
2 Yes 721
Descriptive Statistics
Dependent Variable: MTHTRAN1 One or more transfer math courses
RV
Q20 Is English your
Mean Std. Deviation N
1 High 1 No .32 .47 1108
2 Yes .29 .45 461
Total .31 .46 1569
2 Moderate 1 No .33 .47 430
2 Yes .40 .49 186
Total .35 .48 616
3 Low 1 No .13 .34 171
2 Yes .15 .36 74
Total .14 .35 245
Total 1 No .30 .46 1709
2 Yes .31 .46 721
Total .30 .46 2430
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Tests of Between-Subjects Effects
Dependent Variable: MTHTRAN1 One or more transfer math courses
Source
Type III Sum
of Squares df Mean Square F Sig.
Corrected Model 9.123® 5 1.825 8.763 .000
Intercept 87.567 1 87.567 420.535 .000
RV 7.420 2 3.710 17.818 .000
Q20 .122 1 .122 .586 .444
RV * Q20 1.015 2 .507 2.437 .088
Error 504.743 2424 .208
Total 738.000 2430
Corrected Total 513.867 2429
a- R Squared = .018 (Adjusted R Squared = .016)
1. Grand Mean
Dependent Variable: MTHTRAN1 One or more transfer
math courses
Mean Std. Error
95% Confidence Interval
Lower Bound Upper Bound
.271 .013 .245 .297
Estimates
Dependent Variable: MTHTRAN1 One or more transfer math courses
RV Mean Std. Error
95% Confidence Interval
Lower Bound Upper Bound
1 High .305 .013 .280 .330
2 Moderate .366 .020 .326 .405
3 Low
.142 .032 7.932E-02 .204
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
141
Pairwise Comparisons
Dependent Variable: MTHTRAN1 One or more transfer math courses
(DRV
(J) RV
Mean
Difference
d-J)
Std. Error
s ig a
95% Confidence Interval for
Difference8
Lower Bound Upper Bound
1 High 2 Moderate -6.048E-02* .024 .011 -.107 -1.404E-02
3 Low .164* .034 .000 9.650E-02 .231
2 Moderate 1 High 6.048E-02* .024 .011 1.404E-02 .107
3 Low .224* .038 .000 .150 .298
3 Low 1 High -.164* .034 .000 -.231 -9.650E-02
2 Moderate -.224* .038 .000 -.298 -.150
Based on estimated marginal means
* ■ The mean difference is significant at the .05 level.
a Adjustment for multiple comparisons: Least Significant Difference (equivalent to no
adjustments).
Univariate Tests
Dependent Variable: MTHTRAN1 One or more transfer math courses
Sum of
Squares df Mean Square F Siq.
Contrast 7.420 2 3.710 17.818 .000
Error 504.743 2424 .208
The F tests the effect of RV. This test is based on the linearly independent
pairwise comparisons among the estimated marginal means.
Estimates
Dependent Variable: MTHTRAN1 One or more transfer math courses
Is English your
95% Confidence interval
native language? Mean Std. Error Lower Bound Upper Bound
1 No .261 .014 .232 .289
2 Yes
.281 .022 .238 .324
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
142
Pairwise Comparisons
Dependent Variable: MTHTRAN1 One or more transfer math courses
(I) Is English yor (J) Is English yor
native language' native language'
Mean
Difference
(l-J) Std. Error Sig.a
)5% Confidence Interval fo
Differencl
_ower Bound Upper Bound
1 No 2 Yes 2.021 E-02 .026 .444 -7.199E-02 3.156E-02
2 Yes 1 No 2.021 E-02 .026 .444 -3.156E-02 7.199E-02
Based on estimated marginal means
a-Adjustment for multiple comparisons: Least Significant Difference (equivalent to no adjustm
Univariate Tests
Dependent Variable: MTHTRAN1 One or more transfer math courses
Sum of
Squares df Mean Square F Sig.
Contrast .122 1 .122 .586 .444
Error 504.743 2424 .208
The F tests the effect of Is English your native language?. This test is based on
the linearly independent pairwise comparisons among the estimated marginal
means.
4. RV * Is English your native language?
Dependent Variable: MTHTRAN1 One or more transfer math courses
RV
Is English your
native language? Mean Std. Error
95% Confidence interval
Lower Bound Upper Bound
1 High 1 No .319 .014 .293 .346
2 Yes .291 .021 .249 .332
2 Moderate 1 No .328 .022 .285 .371
2 Yes
.403 .033 .338 .469
3 Low 1 No
.135 .035 6.607E-02 .203
2 Yes .149 .053 4.463E-02 .253
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
143
Multiple Comparisons
Dependent Variable: MTHTRAN1 One or more transfer math courses
LSD _______________
(1) RV (J) RV
Mean
Difference
d-J)
Std. Error Sig.
95% Confidence Interval
Lower Bound Upper Bound
1 High 2 Moderate -3.96 E-02 2.17E-02 .068 -8.22E-02 2.92E-03
3 Low .17* 3.13E-02 .000 .11 .23
2 Moderate 1 High 3.96E-02 2.17E-02 .068 -2.92E-03 8.22E-02
3 Low .21* 3.45E-02 .000 .14 .28
3 Low 1 High -.17* 3.13E-02 .000 -.23 -.11
2 Moderate -.21* 3.45E-02 .000 -.28 -.14
Based on observed means.
*• The mean difference is significant at the .05 level.
Estimated Marginal Means of One or more transfer math courses
.5
.4
E S '—
.3
is English your native
language
.2
No
« Yes . 1
High Moderate Low
RV
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
144
Table 14 Two-way ANOVA (remedial English and math
by RV and Native Language)
Between-Subjects Factors
Value Label N
Q20 Is English your 1 No 1709
native language?
2 Yes 721
RV 1 High 1569
2 Moderate 616
3 Low 245
Multivariate Test^1
Effect Value F Hypothesis df Error df Sig.
Intercept Pillai's Trace .422 885.555a 2.000 2423.000 .000
Wilks' Lam bda .578 885.555a 2.000 2423.000 .000
Hotelling's Trace .731 885.555a 2.000 2423.000 .000
Roy's Largest Root .731 885.555a 2.000 2423.000 .000
Q20 Pillai's Trace .004 4.381a 2.000 2423.000 .013
Wilks' Lambda .996 4.381a 2.000 2423.000 .013
Hotelling's Trace .004 4.3813 2.000 2423.000 .013
Roy's L argest Root .004 4.3813 2.000 2423.000 .013
RV Pillai's Trace .015 8.853 4.000 4848.000 .000
Wilks' Lam bda .985 8.882a 4.000 4846.000 .000
Hotelling's Trace .015 8.911 4.000 4844.000 .000
Roy's L argest Root .015 17.818b 2.000 2424.000 .000
Q20 * RV Pillai's Trace .002 1.334 4.000 4848.000 .255
Wilks' Lambda .998 1.334a 4.000 4846.000 .255
Hotelling's Trace .002 1.334 4.000 4844.000 .255
Roy's L argest Root .002 2.477b 2.000 2424.000 .084
a - Exact statistic
b. The statistic is an upper bound on F that yields a lower bound on the significance level.
c. Design: lntercept+Q20+RV+Q20 * RV
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
145
Between-Subjects Factors
Value Label N
Q20 Is English your 1 No 1709
native language?
2 Yes 721
RV 1 High 1569
2 Moderate 616
3 Low 245
Descriptive Statistics
Dependent Variable: ENGRM1 One or more remedial English courses
Q20 Is English your
RV Mean Std. Deviation N
1 No 1 High .52 .50 1108
2 Moderate .56 .50 430
3 Low .63 .49 171
Total .54 .50 1709
2 Yes 1 High .42 .49 461
2 Moderate .39 .49 186
3 Low .42 .50 74
Total .41 .49 721
Total 1 High .49 .50 1569
2 Moderate .51 .50 616
3 Low .56 .50 245
Total
.50 .50 2430
Tests of Between-Subjects Effects
Dependent Variable: ENGRM1 One or more remedial English courses
Source
ype III Sur
)f Squares df lean Squar F Sig.
Partial Eta
Squared
Noncent.
3 arametei
Dbserved
Powet
Corrected M 10.648b 5 2.130 8.650 .000 .018 43.248 1.000
Intercept 285.271 1 285.271 1 58.624 .000 .323 158.624 1.000
Q20 7.731 1 7.731 31.398 .000 .013 31.398 1.000
RV
.480 2 .240 .975 .377 .001 1.950 .221
Q20 * RV .728 2 .364 1.478 .228 .001 2.955 .317
Error
596.825 2424 .246
Total
1223.000 2430
Corrected T < 607.474 2429
a-Computed using alpha = .05
b.R Squared = .018 (Adjusted R Squared = .016)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
146
1. Grand Mean
Dependent Variable: ENGRM1 One or more remedial
English courses__________________________ _____
Mean Std. Error
95% Confidence Interval
Lower Bound Upper Bound
.489 .014 .461 .517
Estimates
Dependent Variable: ENGRM1 One or more remedial English courses
Is English your
native language?
95% Confidence Interval
Mean Std. Error Lower Bound Upper Bound
1 No .569 .016 .538 .600
2 Yes .408 .024 .361 .455
Pairwise Comparisons
Dependent Variable: ENGRM1 One or more remedial English courses
(I) Is English you (J) Is English you
native language^ native language?
Mean
Difference
0-J)
Std. Error Sig.a
95% Confidence Interval for
Differenc!
Lower Bound Upper Bound
1 No 2 Yes .161* .029 .000 .105 .217
2 Yes 1 No -.161* .029 .000 -.217 -.105
Based on estimated marginal means
*• The mean difference is significant at the .05 level.
a- Adjustment for multiple comparisons: Least Significant Difference (equivalent to no adjustmen
Univariate Tests
Dependent Variable: ENGRM1 One or more remedial English courses
Sum of
Squares df \/lean Square F Sig.
Partial Eta
Squared
Noncent.
Parameter
Observed
Powe?
Contrasl
Error
7.731
596.825
1
2424
7.731
.246
31.398 .000 .013 31.398 1.000
The F tests the effect of Is English your native language?. This test is based on the linearly indepe
comparisons among the estimated marginal means.
a-Computed using alpha = .05
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
147
Estimates
Dependent Variable: ENGRM1 One or more remedial English courses
RV Mean Std. Error
95% Confidence Interval
Lower Bound Upper Bound
1 High .471 .014 .444 .498
2 Moderate .473 .022 .430 .515
3 Low .522 .035 .455 .590
P airw ise C o m p a riso n s
Dependent Variable: ENGRM1 One or more remedial English courses
(!)RV (J) RV
Mean
Difference
d-J)
Std. Error Sig.a
95% Confidence Interval for
Difference®
Lower Bound Upper Bound
1 High 2 Moderate -1.558E-03 .026 .952 -5.206E-02 4.894E-02
3 Low -5.126E-02 .037 .168 -.124 2.160E-02
2 Moderate 1 High 1.558E-03 .026 .952 -4.894E-02 5.2Q6E-Q2
3 Low -4,971 E-02 .041 .223 -.130 3.033E-02
3 Low 1 High 5.126E-02 .037 .168 -2.160E-02 .124
2 Moderate 4.971 E-02 .041 .223 -3.033E-02 .130
Based on estimated marginal means
a- Adjustment for multiple comparisons: Least Significant Difference (equivalent to no
adjustments).
Univariate Tests
Dependent Variable: ENGRM1 One or more remedial English courses
Sum of
Squares df vlean Square F Sig.
Partial Eta
Squared
Noncent.
Parameter
Observed
Powef
Contrast
Error
.480
596.825
2
2424
.240
.246
.975 .377 .001 1.950 .221
The F tests the effect of RV. This test is based on the linearly independent pairwise comparisons ai
marginal means.
a-Computed using alpha = .05
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
148
4. Is English your native language? * RV
Dependent Variable: ENGRM1 One or more remedial English courses
Is English your
native language? RV Mean Std. Error
95% Confidence Interval
Lower Bound Upper Bound
1 No 1 High .523 .015 .494 .553
2 Moderate .558 .024 .511 .605
3 Low .626 .038 .551 .700
2 Yes 1 High .419 .023 .373 .464
2 Moderate .387 .036 .316 .458
3 Low .419 .058 .306 .532
Multiple Comparisons
Dependent Variable: ENGRM1 One or more remedial English courses
LSD
(I) RV (J) RV
Mean
Difference
(l-J) Std. Error Sig.
95% Confidence Interval
Lower Bound Upper Bound
1 High 2 Moderate -1.38E-02 2.36 E-02 .558 -6.01 E-02 3.24E-02
3 Low -7.06E-02* 3.41 E-02 .038 -.14 -3.75E-03
2 Moderate 1 High 1.38E-02 2.36E-02 .558 -3.24E-02 6.01 E-02
3 Low -5.68E-02 3.75E-02 .130 -.13 1.67E-02
3 Low 1 High 7.06E-02* 3.41 E-02 .038 3.75E-03 .14
2 Moderate 5.68E-02 3.75E-02 .130 -1.67E-02 .13
Based on observed means.
*• The mean difference is significant at the .05 level.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Estimated Marginal Means
149
Estimated Marginal Means of One or more remedial English courses
Is English your native
'language
a No
Yes
High Moderate Low
RV
Between-Subjects Factors
Value Label N
Q20 Is English your 1 No 1709
native language?
2 Yes 721
RV 1 High 1569
2 Moderate 616
3 Low 245
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
150
Descriptive Statistics
Dependent Variable: MATHREM1 One or more remedial Math courses
Q20 Is English your
RV Mean Std. Deviation N
1 No 1 High .36 .48 1108
2 Moderate .42 .49 430
3 Low .33 .47 171
Total .37 .48 1709
2 Yes 1 High .37 .48 461
2 Moderate .40 .49 186
3 Low .30 .46 74
Total .37 .48 721
Total 1 High .36 .48 1569
2 Moderate .41 .49 616
3 Low .32 .47 245
Total .37 .48 2430
T ests o f Betw een-Subjects Effects
Dependent Variable: MATHREM1 One or more remedial Math courses
Source
'ype III Sum
of Squares df /lean Square F Sig.
Partial Eta
Squared
Noncent.
Param eter
Observed
Powe?
Corrected M o< 1.918b 5 .384 1.643 .145 .003 8.213 .576
Intercept 158.341 1 158.341 678.083 .000 .219 678.083 1.000
Q20 5.713E-02 1 5.713E-02 .245 .621 .000 .245 .078
RV 1.497 2 .749 3.206 .041 .003 6.413 .614
Q20 * RV .129 2 6.452E-02 .276 .759 .000 .553 .094
Error 566.035 2424 .234
Total 905.000 2430
Corrected Tot 567.953 2429
a. Computed using alpha = .05
b.R Squared = .003 (Adjusted R Squared = .001)
1. Grand Mean
Dependent Variable: MATHREM1 One or more remedial
Math courses
Mean Std. Error
95% Confidence Interval
Lower Bound Upper Bound
.364 .014 .337 .391
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
151
Estimates
Dependent Variable: MATHREM1 One or more remedial Math courses
Is English your
native language?
95% Confidence Interval
Mean Std. Error Lower Bound Upper Bound
1 No .371 .015 .341 .401
2 Yes .357 .023 .311 .403
Pairwise Comparisons
Dependent Variable: MATHREM1 One or more remedial Math courses
(I) Is English your (J) Is English your
native language? native language?
Mean
Difference
(l-J) Std. Error Sig.a
95% Confidence Interval for
Differenc#
Lower Bound Upper Bound
1 No 2 Yes 1.383E-02 .028 .621 -4.100E-02 6.866E-02
2 Yes 1 No -1.383E-02 .028 .621 -6.866E-02 4.100E-02
Based on estimated marginal means
a. Adjustment for multiple comparisons: Least Significant Difference (equivalent to no adjustments).
Univariate Tests
Dependent Variable: MATHREM1 One or more remedial Math courses
Sum of
Squares df Vlean Square F Sig.
Partial Eta
Squared
Noncent
Parameter
Observed
PoweP
Contrast
Error
713E-02
566.035
1
2424
5.713E-02
.234
.245 .621 .000 .245 .078
The F tests the effect of Is English your native language?. This test is based on the linearly independer
comparisons among the estimated marginal means.
a- Computed using alpha = .05
Estimates
Dependent Variable: MATHREM1 One or more remedial Math courses
RV Mean Std. Error
95% Confidence Interval
Lower Bound Upper Bound
1 High .366 .013 .340 .392
2 Moderate .411 .021 .369 .452
3 Low
.315 .034 .249 .381
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
152
Pairwise Comparisons
Dependent Variable: MATHREM1 One or more remedial Math courses
(I)RV
(J) RV
Mean
Difference
d-J)
Std. Error Siga
95% Confidence Interval for
Differenc#
Lower Bound Upper Bound
1 High 2 Moderate -4.494E-02 .025 .073 -9.412E-02 4.235E-03
3 Low 5.066E-02 .036 .162 -2.031 E-02 .122
2 Moderate 1 High 4.494E-02 .025 .073 -4.235E-03 9.412E-02
3 Low 9.560E-02* .040 .016 1.766E-02 .174
3 Low 1 High -5.066E-02 .036 .162 -.122 2.031 E-02
2 Moderate -9.560E-02* .040 .016 -.174 -1.766E-02
Based on estimated marginal means
* ■ The mean difference is significant at the .05 level.
a- Adjustment for multiple comparisons: Least Significant Difference (equivalent to no
adjustments).
Univariate Tests
Dependent Variable: MATHREM1 One or more remedial Math courses
Sum of
Squares df ^ean Square F Sig.
Partial Eta
Squared
Noncent.
Parameter
Observed
PoweP
Contras
Error
1.497
566.035
2
2424
.749
.234
3.206 .041 .003 6.413 .614
The F tests the effect of RV. This test is based on the linearly independent pairwise comparisons i
marginal means.
a-Computed using alpha = .05
4. Is English your native language? * RV
Dependent Variable: MATHREM1 One or more remedial Math courses
Is English your
native language? RV Mean Std. Error
95% Confidence Interval
Lower Bound Upper Bound
1 No 1 High .361 .015 .333 .389
2 Moderate .419 .023 .373 .464
3 Low
.333 .037 .261 .406
2 Yes 1 High .371 .023 .327 .415
2 Moderate .403 .035 .334 .473
3 Low .297 .056 .187 .407
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Estimated Marginal Means
153
Multiple Comparisons
Dependent Variable: MATHREM1 One or more remedial Math courses
LSD
(I) RV (J) RV
Mean
Difference
d-J)
Std. Error Sig.
95% Confidence Interval
Lower Bound Upper Bound
1 High 2 Moderate -5.00E-02* 2.30E-02 .030 -9.51 E-02 -4.98E-03
3 Low 4.15E-02 3.32E-02 .212 -2.36E-02 .11
2 Moderate 1 High 5.00E-02* 2.30E-02 .030 4.98E-03 9.51E-02
3 Low 9.15E-02* 3.65E-02 .012 1.99E-02 .16
3 Low 1 High -4.15E-02 3.32E-02 .212 -.11 2.36E-02
2 Moderate -9.15E-02* 3.65E-02 .012 -.16 -1.99E-02
Based on observed means.
* ■ The mean difference is significant at the .05 level.
Estimated Marginal Means of One or more remedial Math courses
.42
" \
.40
.38
.36
.34
als English your native
language
.32
No .30
.28
High Moderate Low
RV
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 15: Regression with Scale Values
Descriptive Statistics
154
Mean Std. Deviation N
COMPPROP Academic
success = (courses
with C or better and P) /
.83923 .16644 1567
attempted
BANDC 26.22 5.98 1567
ATTITUDE 35.70 4.06 1567
SUBNORM 14.54 6.68 1567
ASPIR 8.39 1.86 1567
NORMBELI 30.24 7.53 1567
DETERM 24.91 2.85 1567
ACADINTG 10.77 4.54 1567
OBSTACLE 16.03 5.20 1567
GENDER .58 .49 1567
Q29 Age on December
31 of this year
5.96 1.48 1567
ENGLISH 26.26 4.31 1567
FACFT 38.936 12.504 1567
FACPART 39.137 11.899 1567
Model Summary
Mode R R Square
Adjusted
R Square
Std. Error of
he Estimate
Change Statistics
R Square
Change : Change df1 df2 iig. F Changi
1 ,111a .012 .008 .16573 .012 3.221 6 1560 .004
2 . 114b .013 .008 .16578 .001 .557 2 1558 .573
3 .149° .022 .015 .16517 .009 4.835 3 1555 .002
4 ,166d .028 .020 .16476 .005 8.773 1 1554 .003
a. Predictors: (Constant), DETERM, SUBNORM, ASPIR, BANDC, NORMBELI, ATTITUDE
b. Predictors: (Constant), DETERM, SUBNORM, ASPIR, BANDC, NORMBELI, ATTITUDE, OBSTA
c.Predictors: (Constant), DETERM, SUBNORM, ASPIR, BANDC, NORMBELI, ATTITUDE, OBSTA
Q29 Age on December 31 of this year, ENGLISH
d. Predictors: (Constant), DETERM, SUBNORM, ASPIR, BANDC, NORMBELI, ATTITUDE, OBSTA
Q29 Age on December 31 of this year, ENGLISH, FACFT
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
155
ANOVAe
Model
Sum of
Squares df Mean Square F Sig.
1 Regression .531 6 8.849E-02 3.221 .004*
Residual 42.849 1560 2.747E-02
Total 43.380 1566
2 Regression .562 8 7.019E-02 2.554 ,009b
Residual 42.819 1558 2.748E-02
Total 43.380 1566
3 Regression .957 11 8.703E-02 3.190 .000°
Residual 42.423 1555 2.728E-02
Total 43.380 1566
4 Regression 1.195 12 9.962E-02 3.670 ,000d
Residual 42.185 1554 2.715E-02
Total 43.380 1566
a- Predictors: (Constant), DETERM, SUBNORM, ASPIR, BANDC, NORMBELI,
ATTITUDE
b. Predictors: (Constant), DETERM, SUBNORM, ASPIR, BANDC, NORMBELI,
ATTITUDE, OBSTACLE, ACADINTG
c. Predictors: (Constant), DETERM, SUBNORM, ASPIR, BANDC, NORMBELI,
ATTITUDE, OBSTACLE, ACADINTG, GENDER, Q29 Age on December 31 of this
year, ENGLISH
d- Predictors: (Constant), DETERM, SUBNORM, ASPIR, BANDC, NORMBELI,
ATTITUDE, OBSTACLE, ACADINTG, GENDER, Q29 Age on December 31 of this
year, ENGLISH, FACFT
e- Dependent Variable: COMPPROP Academic success = (courses with C or better
and P) / attempted
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
156
C oefficient^
Unstandardized
Coefficients
Standard!
zed
Coefficien
ts 95% Confidence Interval for B
Model B Std. Error Beta t Sig. Lower Bound Upper Bound
' I ......... (Constant) .730 .043 16.937 .000 .646 .815
BANDC -1.54E-03 .001 -.055 -1.919 .069 -.003 .000
ATTITUDE -7.56E-04 .001 -.018 -.512 .609 -.004 .002
SUBNORM -7.71 E-04 .001 -.031 -1.009 .313 -.002 .001
ASPIR 3.564E-03 .002 .040 1.524 .128 -.001 .008
NORMBELI 8.427E-04 .001 .038 1.168 .243 -.001 .002
DETERM 5.298E-03 .002 .091 2.548 .011 .001 .009
2 (Constant) ,738 .045 16.253 .000 .649 .827
BANDC -1.51E-Q3 .001 -.054 -1.781 .075 -.003 .000
ATTITUDE -9.44E-04 .001 -.023 -.635 .526 -.004 .002
SUBNORM -8.32E-04 .001 -.033 -1.079 .281 -.002 .001
ASPIR 3.498E-03 .002 .039 1.490 .136 -.001 .008
NORMBELI 7.482E-04 .001 .034 1.028 .304 -.001 .002
DETERM 5.218E-03 .002 .089 2.506 .012 .001 .009
ACADINTG 9.635E-04 .001 .026 .988 .323 -.001 .003
OBSTACLE -3.73E-04 .001 -.012 -.457 .648 -.002 .001
3 (Constant) .788 .052 15.277 .000 .687 .889
BANDC -1.05E-03 .001 -.038 -1.233 .218 -.003 .001
ATTITUDE -5.33E-05 .002 -.001 -.035 .972 -.003 .003
SUBNORM -1.06E-03 .001 -.042 -1.359 .174 -.003 .000
ASPIR 2.064E-03 .002 .023 .861 .390 -.003 .007
NORMBELI 4.390E-04 .001 .020 .600 .548 -.001 .002
DETERM 5.869E-03 .002 .101 2.819 .005 .002 .010
ACADINTG 9.744E-04 .001 .027 1.002 .316 -.001 .003
OBSTACLE -1.74E-04 .001 -.005 -.208 .835 -.002 .001
GENDER
Q29 Age on December
-7.53E-03 .009 -.022 -.886 .376 -.024 .009
31 of this year
-1.09E-02 .003 -.097 -3.650 .000 -.017 -.005
ENGLISH -7.20E-04 .001 -.019 -.690 .490 -.003 .001
4 (Constant) .761 .052 14.549 .000 .658 .863
BANDC -1.08E-03 .001 -.039 -1.268 .205 -.003 .001
ATTITUDE -3.42E-04 .002 -.008 -.226 .821 -.003 .003
SUBNORM -1.15E-03 .001 -.046 -1.480 .139 -.003 .000
ASPIR 2.405E-03 .002 .027 1.004 .316 -.002 .007
NORMBELI 4.813E-04 .001 .022 .659 .510 -.001 .002
DETERM 6.060E-03 .002 .104 2.917 .004 .002 .010
ACADINTG 8.392E-04 .001 .023 .864 .388 -.001 .003
OBSTACLE -1.96E-04 ,001 -.006 -.235 .814 -.002 .001
GENDER
Q29 Age on December
-8.29E-03 .008 -.025 -.976 .329 -.025 .008
31 of this year
-1.15E-02 .003 -.102 -3.836 .000 -.017 -.006
ENGLISH -8.12E-04 .001 -.021 -.780 .436 -.003 .001
FACFT 9.969E-04 .000 .075 2.962 .003 .000 .002
a- Dependent Variable: COM PPROP Academic su ccess = {courses with C or better and P) / attem pted
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Academic success and student-parents in the Los Angeles Community College District

PDF

Asian students in community colleges: A study of intersecting effects of student characteristics, construct models of retention, social reproduction and resulting variables as applied to the stud...

PDF

Factors influencing academic success of Chinese international students in Los Angeles community colleges

PDF

A study of the profile and leadership traits of vice presidents of instruction in the California community college system

PDF

Academic success for African -American male community college students

PDF

Evaluating the California Community Colleges Registry

PDF

Examining the factors that predict the academic success of minority students in the remedial mathematics pipeline in an urban community college

PDF

Community college students and achievement in mathematics: Predictors of success

PDF

Academic performance and persistence of Asian American students in the Los Angeles Community College District

PDF

Community college students: The effect of parenthood and selected variables on degree -seeking aspirations

PDF

Faculty characteristics: What are their relationships with academic outcomes of community college students?

PDF

Community college English courses: The road less traveled by community college students

PDF

Alumni student -athletes' attitudes towards educational philanthropy

PDF

Community college students' perceptions of collaborative learning in developmental writing classes: Identifying the factors that promote positive active learning

PDF

Fulfilling the mission to serve the underserved: A case study of a private, Catholic, urban college's two -year program

PDF

Improving the transfer rates of minority students: A case study

PDF

Exploring ethnic identity on a university campus: Filipino American students' perspectives

PDF

Increasing graduate success on the CAT version of the NCLEX -RN

PDF

"Home boy" to "school boy": Inside the PowerBuilders Workshops at CSULA. A qualitative examination of access, mobility, and leadership qualities of minorities in higher education

PDF

Achievement and retention of first-semester nursing students: The effects of a study skills course

Asset Metadata

Creator Cepeda, Rita Maria (author)

Core Title Academic, environmental and social integration variables that maximize transfer preparedness for Latino community college students: An application of academic success models to the study of Tran...

School Rossier School of Education

Degree Doctor of Education

Degree Program Education

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag education, bilingual and multicultural,education, community college,OAI-PMH Harvest,sociology, ethnic and racial studies

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Hagedorn, Linda Serra (committee chair), Picus, Lawrence O. (committee member), Sundt, Melora (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-358125

Unique identifier UC11339357

Identifier 3103866.pdf (filename),usctheses-c16-358125 (legacy record id)

Legacy Identifier 3103866.pdf

Dmrecord 358125

Document Type Dissertation

Rights Cepeda, Rita Maria

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA