University of Southern California Dissertations and Theses

The practice and effects of a school district's retention policies

(USC Thesis Other)

The practice and effects of a school district's retention policies

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Running Head: EFFECTS OF RETENTION 1

The Practice and Effects of a School District's Retention Policies

by

Christine Hamlin

___________________________________________________________________________

A Dissertation Presented to the
FACULTY OF THE USC ROSSIER SCHOOL OF EDUCATION
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF EDUCATION

May 2013

Copyright 2013 Christine Hamlin

EFFECTS OF RETENTION 2
Table of Contents
Dedication ....................................................................................................................................... 5
Acknowledgements ......................................................................................................................... 6
Abstract ........................................................................................................................................... 7
List of Tables .................................................................................................................................. 8
Chapter One: The Problem ........................................................................................................... 10
Statement of the Problem ......................................................................................................... 12
Purpose of the Study ................................................................................................................ 13
Research Questions .................................................................................................................. 13
Importance of the Study ........................................................................................................... 13
Definition of Terms .................................................................................................................. 14
Chapter Two: Literature Review .................................................................................................. 16
Early Landmark Meta-Analyses .............................................................................................. 16
Current Responses to the Early Meta-Analyses of Retention Studies ..................................... 19
Critique of Holmes, (1989) ................................................................................................. 21
Critique of Jimerson, (2001) ............................................................................................... 22
Methodological Considerations for New Retention Research ................................................. 23
Recent Methods to Correct Design Deficiencies ..................................................................... 24
International Findings .............................................................................................................. 25
Policy Considerations .............................................................................................................. 27
Intervention Services and Retention ........................................................................................ 29
Hypothesis ................................................................................................................................ 31
Chapter Three: Methodology ........................................................................................................ 32
Quantitative Research Design .................................................................................................. 33
Study 1: Pre/Post Independent Retrospective Control Group Design ................................. 33
Study 2: Correlation Design ................................................................................................ 34
Participants and Setting ............................................................................................................ 35
Academic Intervention ............................................................................................................. 36
Instrumentation and Procedures ............................................................................................... 38
Study 1 ................................................................................................................................. 38
Study 2 ................................................................................................................................. 38
Analysis .................................................................................................................................... 44
Limitations (Threats to Validity) ............................................................................................. 45
Internal Validity .................................................................................................................. 45
Study 1 ............................................................................................................................ 45
EFFECTS OF RETENTION 3
Study 2 ............................................................................................................................ 45
External Validity ................................................................................................................. 45
Study 1 ............................................................................................................................ 45
Study 2 ............................................................................................................................ 46
Statistical Conclusion Validity ............................................................................................ 47
Study 1 ............................................................................................................................ 47
Study 2 ............................................................................................................................ 47
Chapter Four: Results ................................................................................................................... 48
Research Question 1: How does retention of an SUSD student in kindergarten through second
grade affect the student’s future academic achievement? ........................................................ 49
Study 1 ................................................................................................................................. 49
Study 2 ................................................................................................................................. 51
Findings in Regards to CST Performance ................................................................................ 52
Findings in Regards to CELDT Performance .......................................................................... 53
Sample Characteristics in Regards to t-Tests using Report Card Reading Grades .................. 55
Sample Characteristics in Regards to Frequencies Using Report Card Reading Grades ........ 56
Sample Characteristics in Regards to Frequencies Using Report Card Reading Grades
Compared with CST scores ...................................................................................................... 58
Sample Characteristics in Regards to the Report Card Work/Study and Citizenship Grades . 60
Research Question 2: What is the correlation between selected demographics and retention?63
Study 2 ................................................................................................................................. 63
Sample Characteristics Related to Education Setting. ............................................................. 63
Sample Characteristics in Regards to Gender, Ethnicity, Age and Program Participation...... 64
Sample Characteristics in Regards to Family .......................................................................... 66
Sample Characteristics in Regards to Special Education ......................................................... 67
Sample Characteristics in Regards to Attendance ................................................................... 68
Sample Characteristics in Regards to Suspensions .................................................................. 69
Findings in Regards to Academic Risk Notification and Intervention .................................... 70
Sample Characteristics in Regards to the Student Study Team (SST) Process ....................... 71
Chapter Five: Discussion .............................................................................................................. 73
Academic Achievement as Measured by a Retrospective Control Group ............................... 73
Study 1 ................................................................................................................................. 73
Methodological Considerations ............................................................................................... 74
Internal validity. .................................................................................................................. 74
External validity. ................................................................................................................. 76
Statistical validity. ............................................................................................................... 76
EFFECTS OF RETENTION 4
Academic Achievement Measured by Teachers’ Classroom Expectations and State Testing 76
Study 2 ................................................................................................................................. 76
Report Card Reading Grades ................................................................................................... 77
Students’ Reading Progress as Measured by Grade Level Expectations ................................. 77
Methodological Concerns ........................................................................................................ 78
Internal validity. .................................................................................................................. 78
External validity. ................................................................................................................. 79
State Testing ............................................................................................................................. 79
Methodological Concerns ........................................................................................................ 80
Internal validity. .................................................................................................................. 80
External validity. ................................................................................................................. 80
Statistical conclusion validity. ................................................................................................. 80
Demographics and Retention ................................................................................................... 80
Age and Maturity ................................................................................................................ 80
Methodological Concerns ........................................................................................................ 81
Internal validity. .................................................................................................................. 81
External validity. ................................................................................................................. 82
Policy Considerations .............................................................................................................. 83
Parental Involvement ............................................................................................................... 83
Academic Performance Guidelines for Retention.................................................................... 84
Academic Intervention ............................................................................................................. 84
Special Education ..................................................................................................................... 85
Further Study ............................................................................................................................ 87
Birth Order ............................................................................................................................... 87
Behavior ................................................................................................................................... 87
Maturity as a Reason for Retention .......................................................................................... 88
Policy Dilemmas ...................................................................................................................... 88
Policy Recommendations ......................................................................................................... 89
Conclusion ............................................................................................................................... 90
References ..................................................................................................................................... 91
Appendix A: Social Promotion Retention Laws ........................................................................... 95
Appendix B: Pupil Record Card ................................................................................................... 97
Appendix C: Academic Intervention ............................................................................................ 98
Appendix D: Enrollment Form ................................................................................................... 133

EFFECTS OF RETENTION 5
Dedication
To Val Gordon, for being my friend throughout the limited time availability of a doctoral
program;
To Eva Yocum Aeschliman, for modeling that women can make significant contributions to the
world of higher education in a country where married women could not have a bank account in
their own name;
To Donald Aeschliman, for modeling both life-long learning and that the late fifties is a great
time to embark on a doctoral degree;
To my beloved children and grandchildren: Alyson, Rochelle, Sean, Xavier, Nathan, Ashley, and
Therese; for love, encouragement, and inspiration;
And to Greg Hamlin: The great love of my life, who so often has believed things of me before I
could believe it for myself. So many long, hard days couched in your care, insights,
encouragement, and faith- thank you!

EFFECTS OF RETENTION 6
Acknowledgements
Amanda Conrad was extremely helpful in the data gathering phase of this research:
Thank you for your attention to detail and your cheerfulness in the process.
My sincere appreciation to my doctoral committee:
To Dr. Julietta Shakhbagova: Thank you for steering me to think more globally about the
issue of retention in education systems around the world;
To Dr. Joan Lucid: Thank you for your willingness to discuss the practical ramifications
of my design and the application of my findings, as well as your regular encouragement
throughout this long process;
To Dr. Dennis Hocevar: I am deeply grateful for your never ending availability and your
expertise. Whether the discussion was philosophical, statistical, or practical, your insights were
always thought- provoking and intellectually stimulating. Thank you for taking on an
independent dissertation and for sharing my enthusiasm of the subject.

EFFECTS OF RETENTION 7
Abstract
During conferences to decide whether or not to retain a child, usually two questions are asked. The
first is, “Will retention bring success in school?” The other is, “If we retain him, will he perform
better in school than if we don’t retain him?” The goal of this research was to see if retained students
would make enough academic growth due to retention that they would to catch up to their peers, or
even surpass them as a result of being retained. Two separate studies were conducted to measure the
academic effects of retention. The first study identified the group of 1999-2000 students who were
retained under the changing California law that banned social promotion. Propensity scores were
used to create a retrospective control group of K-2 students from 1998-99 who would have been
retained if the law had been adopted a year earlier. The second study examined records of retained
students from the 2011-2012 school year. Performance pre- and post-retention was measured using
California Standards Test (CST) scores, California English Development Test (CELDT) scores, and
teachers’ scores. Results indicate that retention does not benefit students academically. Retained
students did not meet the level of CST performance of their non-retained peers; almost half (48%) of
the retained students did not show grade level proficiency post retention on CST; ELL students did
not gain an advantage in language learning due to retention; and teacher perception was that only
24% of the retained students worked within grade level expectations post-retention.
EFFECTS OF RETENTION 8
Tables
Table 1: Comparison of RUSD Retention Rates 35
Table 2: 2000 RUSD Guidelines for Retention by Grade Level 37
Table 3: Pupil Record Card and STAR Variables Coding 38
Table 4: Coding of Enrollment Form Variables 39
Table 5: Coding for Variables from the Cumulative File 40
Table 6: Progress Report Code Variable Coding 42
Table 7: Statistical Results of the t-Test of Pupil Record Card (PRC) Scores 50
Table 8: Statistical Results of the t-Test of California Standards Test (CST) Scores 51
Table 9: Statistical Results of the t-Test of Language Arts CST Scores 52
Table 10: A Comparison of Retained Students 2011-12 CSTs with All Performance Levels with
All Non-Retained RUSD Students 53
Table 11: Statistical Results of the t-Test of Individual CELDT Scores 54
Table 12: Statistical Results of the t-Test of Retained and Non-Retained Students’ CELDT
Scores 54
Table 13: Results of the t-Test of Composite Report Card Reading Scores: Retention Year
and 2011-12 55
Table 14: Results of the t-Test comparing Phonics and Comprehension Report Card Grades from
the Retention Year and 2011-2012 56
Table 15: Crosstabulation of Report Card Phonics Scores: Retention Year and 2011-2012 57
Table 16: Results of the Crosstabulation of Report Card Comprehension Grades comparing the
Retained year and 2011-12 58
Table 17: 2011-12 CST Language Arts Proficiency Level and Retained Year Phonics Grade
Crosstabulation 59
Table 18: 2011-12 CST Language Arts Proficiency Level and Retained Year Comprehension
Grade Crosstabulation 59
Table 19: A Comparison of Composite Work/Study Habits Scores: Retention Year and
2011-12 60
Table 20: Results of the Work/Study Habits t-Test Comparing Retained Year and 2011-12 61
Table 21: A Comparison of Composite Citizenship Scores: Retention Year and 2011-12 62
Table 22: Results of the Citizenship t-Test Comparing the Retained Year and 2011-12 Report
Card Scores 62
Table 23: Summary of Retained Student Enrollment by Grade Level 64
Table 24: Comparison of Study Group and District Demographics 65
EFFECTS OF RETENTION 9
Table 25: Crosstabulation of the Student’s Birth Order in the Family and Number of Children in
the Family 66
Table 26: Grade Levels At Which Students Were Tested for a Disability 67
Table 27: Comparison of Attendance rates Pre-Retention and Post-Retention 68
Table 28: Numbers of Suspensions Compared to Retention Year 69
Table 29: Comparison of Suspension Rates During the 2011-2012 70
Table 30: Intervention Participation as Noted in the Cumulative File 71
Table 31: Parent Participation in the SST Process as Noted in the Cumulative File 71

EFFECTS OF RETENTION 10
Chapter One: The Problem
Teachers enter the education profession because they want students to learn. Initial and
ongoing teacher professional development centers on understanding the content and skills
students need to know, and which pedagogical strategies will bring about the greatest learning.
Despite their best efforts and intentions, at some point each teacher is faced with the fact that a
student has not mastered the content and skills deemed appropriate for their grade level, and the
question is raised: Does this student need to be retained so that he can have an additional
opportunity to master the grade level standards, or should he be promoted with the hopes that he
will catch up at the next grade level?
Among educators, and in society as a whole, that question provokes many discussions.
On the one side, giving the student “the gift of time” makes a lot of sense; maturity plays a role
in a student’s ability to learn as well as the pace at which he learns (Jimerson & Kauffman,
2003). It seems unfair to the student to ask him to move on to more difficult work knowing that
he is beginning the next grade level with a handicap. It has been suggested that repeating a grade
may be developmentally appropriate and thus more meaningful for struggling students (Smith &
Shepard, 1988). On the other side, there is also the voice calling for the student to apply himself,
to take responsibility, and to buckle down and study, because there is no free ride in life (Hong &
Raudenbush, 2005).
Several longitudinal studies regarding student progress after retention seem to suggest
that students who are retained receive at most a two to three year benefit from being retained
(Frey, 2005). The studies also indicate that boys are more likely to be retained than girls, as are
students of color and poverty (Jimerson, 2001); students who have conflicts in their relationships
with teachers; students with low ego resilience; and students whose parents exhibit a low
EFFECTS OF RETENTION 11
parental sense of responsibility (Willson & Hughes, 2005; Wu, West & Hughes, 2008). Students
who are retained are more likely to continue to struggle academically throughout their school
career, drop out of school later, and suffer negative emotional affects from retention (Jimerson &
Kauffman, 2003).
Within every community anecdotal records about the success or failure of retention
abound. Recently, researchers are questioning whether the methodology of prior studies showing
the ill effects of retention was so flawed as to render the results useless (Lorence, 2006). And
while the debate continues, educators and parents are still left with the need to answer the
question now: should this particular child be retained or socially promoted?
The practice of retention is not new. From the early 1900s on, students have repeated a
grade level in school when they have fallen behind (Homes & Matthews, 1984). In many
countries, particularly developing countries, grade retention is the national policy. In sub-
Saharan Africa, the rate of retention can run as high as 22%, compared with 12% in North Africa
and the Middle East, 9% in Latin America, and 8% in the Caribbean (Brophy, 2006; Taylor et
al., 2010). In contrast, the retention rates range from one to five percent (Brophy, 2006) in
developed countries, most of whom have policies endorsing social promotion.
Some countries have no national policy on retention or social promotion, but local culture
accepts and endorses the practice of retention. For example, in the Flemish section of Belgium,
there is no set of mastery criteria for grade levels: each school decides if a child is ready to
promote, and parents and teachers make the decision jointly to retain a child in first grade. The
negative overtone of retention is not as prevalent as it is in the United States (Goos, Van Damme,
Onghena, & Petry, K., 2011).
EFFECTS OF RETENTION 12
In the United States, President Clinton called for the end of Social Promotion in his State
of the Union speeches (Clinton, 1998). He stressed that if a student did not demonstrate
appropriate competence to be promoted, he should be retained. The California legislature
followed the national trend and introduced State Standards and an end to social promotion by
adopting AB 1626 in 1999. This law required school districts to have a detailed promotion /
retention policy approved by the local school board. SB 1370 was a companion bill that provided
guidelines and funding for schools to offer intensive intervention for students at risk of being
retained. As a result of the new laws, districts throughout California instituted policies that
detailed how to identify students who were at risk of being retained, what intervention
opportunities they should be given, and what criteria should be met before the decision to
promote or retain was made. The State of California has not kept data on how many students
have been retained as a result of the law, or whether the law was effective in furthering student
success (California Department of Education, n.d.)
Statement of the Problem
The Regent Union School District (RUSD) has retained 832 students over the past twelve
years, representing approximately 5.8% of the students enrolled annually in the district. The
percentage of students being retained peaked in June 2000 when the district instituted its new
retention policies based on State legislation prohibiting social promotion. At the beginning of the
2011-2012 school year the district had fifteen schools with 190 students that had been retained
between their kindergarten and fifth grade years.
RUSD has never conducted a formal study to measure the academic effects of student
retention on the students they have retained. They have not conducted an analysis of the reasons
for retention; whether retentions are parent, teacher, or administrator initiated; whether the
district policy of early identification for risk of retention was followed; and whether
EFFECTS OF RETENTION 13
interventions preceded retention. There has been no analysis of whether students who were
retained were later found to have a learning disability, or whether students are more likely to be
retained at some district schools but not others. No analysis of language, ethnicity, age, or gender
of retained students has been conducted. It is therefore unknown whether RUSD retention policy
is implemented as written, or if retention results in their students’ academic success.
Purpose of the Study
The purpose of this research is two-fold: (1) to study the effects of retention on student
achievement as measured by teacher grades and California Student Testing and Reporting
(STAR) scores; and (2) to analyze the characteristics and academic performance of students who
were retained from 2005-2011.
Research Questions
This study will seek to answer the following questions:
1. How does retention of a RUSD student in kindergarten through second grade affect the
student’s future academic achievement?
2. What is the correlation between selected demographics and retention?
3. What are the policy implications of this study?
Importance of the Study
Given the fact that RUSD students are retained each year; that nationally, retention has
been shown to have adverse effects on some retained students; that students of color, poverty,
and second language acquisition are retained in larger proportions; and that some retained
students later demonstrate learning disabilities; it behooves RUSD personnel and parents to
understand the effects of retention on their students so that wise decisions related to promotion
and retention can be made. A detailed analysis of retained RUSD students’ performance data
compared to at-risk non-retained RUSD students’ performance data will allow district personnel
EFFECTS OF RETENTION 14
to identify the circumstances under which retention benefitted the student. The knowledge that is
gained from this local study will assist RUSD educators in implementing a retention/promotion
policy that is aligned both with state law and with what is best for students in their unique
context.
Definition of Terms
Content Standards (also referred to as “standards”):
Content and a set of prescribed skills to be mastered at each grade level as adopted by the
California Department of Education.
Individualized Education Plan (IEP):
A legal document describing the results of academic, cognitive, and other related testing
that demonstrates a child’s eligibility for receiving special education services. The plan
sets annual learning and performance goals and describes the conditions under which the
student will receive additional assistance. This federal protection is part of the Individuals
with Disability Education Act (IDEA), which was reauthorized in 2004.
Intervention:
Providing students with targeted, specific, additional assistance in the curricular areas
where they are not demonstrating mastery.
Resource Services Program (RSP):
Limited, specific, targeted instruction received by special education students in their
classroom, or as a pull-out class, for a certain number of minutes each day.
Retention:
The practice of requiring students to repeat a year in the same grade level they have just
completed.

EFFECTS OF RETENTION 15
Social Promotion:
The practice of promoting students to the next grade even if they have not demonstrated
mastery of the content and skills at their current grade level.
Special Day Class:
Special education placement in a regular school’s classroom that has fewer students and
more adult assistance in order to meet the needs of students with severe learning
disabilities.
Special Education:
The services a child receives as a result of being identified as having a learning disability.
Standardized Testing and Results (STAR):
The comprehensive annual state tests for students in grades two through twelve based on
the California Content Standards.

EFFECTS OF RETENTION 16
Chapter Two: Literature Review
The latter part of the twentieth century produced three major studies that became the gold
standard for the argument against grade level retention despite the laws ending social promotion
that were being enacted over the same period of time (Lorence, 2006). Jackson (1975), Holmes
and Matthews (1984), and Jimerson (2001) all conducted meta-analyses of existing studies and
concluded that the evidence overwhelmingly supported promotion over retention for ensuring
student success over time.
Early Landmark Meta-Analyses
Jackson (1975) analyzed 17 naturalistic studies comparing retained students with
promoted students for achievement and socioemotional adjustment; 12 studies of performance
and adjustment of retained students using a pre- and post-test design; and three studies
comparing only the academic factors of struggling students who were randomly assigned to be
retained or promoted. All of the original studies were published between 1911 and 1973.
Jackson’s analysis (1975) of these studies showed mixed results: statistically significant
results in the17 naturalistic studies favored promoted students in 24 achievement analyses and 27
socioemotional analyses, compared with two achievement analyses and three socioemotional
analyses favoring the retained students. The remaining non-significant analyses favored
promoted students 85 times, and retained students 63 times.
For the 12 pre/post design studies, 69 analyses demonstrated academic gains and 29
analyses demonstrated socioemotional gains for promoted students, while one showed a
statistical socioemotional decrease and two showed no difference for retained students. All other
analyses were statistically non-significant (Jackson, 1975). Jimerson (2001) suggests that one
caution in interpreting these results is the absence of a control group: students may be
EFFECTS OF RETENTION 17
demonstrating gains, but it is not clear if these gains are greater for retained students than for
students who were promoted.
Jackson’s final group of three studies (1975) demonstrated that in the random assignment
for retention, only one analysis demonstrated promotion as statistically beneficial, and no
analyses demonstrated retention as statistically beneficial. All other analyses were statistically
insignificant. Based on these results, Jackson concluded that retaining students with academic or
adjustment difficulties was not validated by the research. Although retention might provide some
benefit for students, it would have no benefit over promoting the student to the next grade
(Jackson, 1975).
The next statistically comprehensive meta-analysis of retention was conducted by
Holmes and Matthews (1984), and was based on effect sizes. Their analysis reviewed 44 studies
from 1929 to 1981 encompassing 4,208 retained students, compared to 6,924 promoted students.
Of the 44 studies, 18 were matched comparisons based on selected demographic factors and
achievement. In the areas of academic achievement and socioemotional adjustment of both
matched and non- matched cohort studies, Holmes and Matthews found statistical benefit for
promoted students overall, with retained students showing lower achievement and
socioemotional adjustment (Holmes & Matthews, 1984). They concluded that their meta-analysis
demonstrated that the negative effects of retention consistently outweighed any potential positive
effects of retention.
Holmes (1989) increased his meta-analysis by including 19 additional studies published
between 1925 and 1989 so that the meta-analysis now included 25 cohorts matched on a variety
of demographic, achievement, and socioemotional factors. He reported that nine studies
demonstrated positive results for retained students but that the benefits waned over the years.
EFFECTS OF RETENTION 18
Because 54 studies indicated negative effects of retention, he concluded again that promotion
was to be preferred over retention, due to the overwhelming evidence of the positive educational
and socioemotional effects of promotion.
Jimerson (2001) cited concerns with the methods employed in the Holmes & Matthews
(1984) and Holmes (1989) retention studies due to the inclusion of data that had no matched
comparison groups. Jimerson’s meta-analysis (2001) sought to rectify these shortcomings, and
he reviewed data that was more current, published between 1990 and 1999.
Jimerson (2001) analyzed only the 20 studies that included a comparison group of
promoted students that were designed to address the efficacy of retention using academic or
socioemotional factors along with demographic features. Academic growth was measured by
achievement tests or classroom grades. The majority of the students were retained in
kindergarten through third grade, although six studies included students retained as high as
eighth grade. Statistical analysis of the studies he reviewed was also based on effect sizes
(Jimerson, 2001.)
Out of 175 statistically significant academic achievement outcomes, 84 showed no
statistical difference between the groups, 82 favored the promoted comparison group, and nine
favored the retained cohort. Of those nine, only three favored retained students beyond the
repeated year. With regard to socioemotional outcomes, 127 analyses showed no difference
between the two groups, while eight favored the retained group of students and 13 favored the
promoted cohort (Jimerson, 2001).
Included in Jimerson’s meta-analysis (2001) was a review of the individual authors’
conclusions related to grade level retention efficacy: Although four of the 20 authors concluded
that retention was favorable, they also concluded that retention alone was not sufficient to
EFFECTS OF RETENTION 19
guarantee student success. Jimerson (2001) also indicated that consistent with Holmes’ findings
(1989), the longitudinal effect of retention showed that initial benefits dissipated or reversed in
later years, and increased the likelihood of dropping out in high school (Jimerson, 2001;
Jimerson, Anderson & Whipple, 2002). Regardless of Jimerson’s concerns about Holmes and
Matthews (1984) and Holmes (1989) methodology, his conclusion is consistent with their
conclusions: research does not indicate that grade level retention provides more academic or
social benefits than promotion (Jimerson, 2001).
Current Responses to the Early Meta-Analyses of Retention Studies
Lorence (2006) thoroughly reviewed both the Holmes (1989) and Jimerson (2001) meta-
analyses and, based on his examination of their methodology, he has decried their use in guiding
policy. His primary concern is that the authors effectively created a scorecard that gave each
study equal weight, even though some of the included studies were mediocre and some were
sound. He also argues that Jimerson’s evaluation (2001) is flawed because he included several
reports based on two single studies showing negative outcomes for retention as if they were
individual studies, thus lending additional weight to the number of studies showing negative
effects for retention (Lorence, 2006). This practice created the impression that more studies had
actually shown negative effects for retention than the actual number of studies that demonstrated
that conclusion.
Lorence argues that only studies that meet rigorous research criteria should have been
included in both the meta-analyses (2006). These criteria include the composition and size of the
groups, the measures that were used, and the inclusion of only peer-reviewed studies in the meta-
analyses.
One of the inherent issues in retention studies is the lack of randomized control groups
(Hong & Raudenbush, 2005; Lorence, 2006; Wu, West, & Hughes, 2008). The practice of
EFFECTS OF RETENTION 20
creating matched groups based on equivalent scores can create the illusion of equivalence, but
may be biased against retained students because it does not account for the difficulties in
unspecified areas in which retained students differ from their promoted peers; for example,
degree of hyperactivity, parent educational level, and so forth. (Jackson, 1975; Wu et al., 2008).
Hong and Rudenbush (2005) point out that even statistical adjustments for background variables
do not remove bias when the two groups are hardly comparable: promoted students simply do
not have the same academic risk factors to begin with that retained students have.
Another design decision for creating matched groups is to make statistical adjustments
for some of the pre-retention variables through use of multiple regression or an analysis of
covariance (Wu et al., 2008). However, this process assumes that a limited number of covariates
are adequate; that the linear relationship between the covariates and the outcome has been
specified correctly; and that the lines for retained and promoted groups are parallel (Wu et al.
2008; Lorence, 2006). It also assumes that all of the confounding variables have been measured
(Hong & Raudenbush, 2005).
An additional issue related to comparison groups is the actual basis of comparison. Some
studies compare retained and promoted students as “same-age” or “same-year” peers: students
are the same age, but are in different grades when being compared (Holmes, 1989; Karweit,
1992; Hong & Raudenbush, 2005; Wu et al., 2008). Other studies use a “same grade”
comparison which involves comparing performance means of promoted and retained students
when they are in the same grade, but during different years (Holmes, 1989; Wu et al., 2008), or
comparing outcomes of retained students with their current non-retained peers (Hong &
Raudenbush, 2005; Lorence, 2006). Holmes meta-analysis (1989) reported that same-age studies
tend to show that retention has a negative effect, possibly because the promoted peers have had
EFFECTS OF RETENTION 21
an additional year of new learning opportunities. Some same-grade studies show a positive initial
result for retention, probably because retained students have had an extra year to master grade
level material. Lorence (2006) concludes that same-grade comparisons are a more worthwhile
comparison because they compare retained students to their new peers, especially if the measure
used is a grade-equivalent score, while Hong and Raudenbush view this as a drawback because
they do not show what the retained student may have attained if he had been promoted (Hong &
Raudenbush, 2005).
Critique of Holmes, (1989)
In his quest to provide a rigorous review of Holmes(1989), Lorence included only the 10
studies out of 63 that were peer reviewed (2006). Of these, six studies had inadequately matched
groups and two were questionable based on whether the retained and promoted students were
comparable at the time of retention. The degree of similarity and difference between retained and
promoted students was uncertain. None of the studies met current standards of a sample size of
several hundred for a non-experimental research study, and only four studies met the criterion of
a properly controlled study (Lorence, 2006). Of the four, two of these studies reported negative
outcomes for retention, and two reported positive outcomes for retention.
However, both of the studies that reported negative outcomes for retention used same age
comparisons as well as grade equivalent units. This is problematic because retained students are
being compared to promoted students on tests that methodologically favor students exposed to
next level curriculum. (Lorence, 2006). Thus Lorence concludes that Holmes meta-analysis
(1989) does not provide overwhelming support for the notion that retention is “…an ineffective
or harmful educational remediation strategy” (Lorence, 2006, p. 748).
EFFECTS OF RETENTION 22
Critique of Jimerson, (2001)
Although nine of the 18studies contained in Jimerson’s meta-analyses (2001) showed
what he considered an adequate design, Lorence (2006) maintains that most suffered from small
sample sizes and limited locations. Only four of the best designed studies had large enough
sample sizes to be considered generalizable, and of the four, two reported negative effects of
retention and two reported positive effects of retention. However, all four studies contained
measurment flaws that left them being less than authoritative in deciding whether retention or
promotion is superior (Lorence, 2006).
A further criticism of Jimerson’s meta-analysis (2001) is that his conclusions included
one tally of how many analyses were favorable compared to how many were not, without
disaggregating them based on same-age or same-grade study design. When Lorence (2006)
provided a closer examination of the study by Alexander et al. (1994), reviewed by Jimerson
(2001), he demonstrated that although the same-age comparisons yielded 23 negative effects, a
comparison of the same-grade data showed another result: the scores for math and reading pre-
retention were significantly lower for the retained group than the control group, but post
retention, the mean of the retained group was higher than the mean of the socially promoted
control group (Lorence, 2006). If Jimerson (2001) had disaggregated the data based on same-age
or same-grade design, his conclusions may have been less supportive of social promotion
(Lorence, 2006.)
One of the more unusual studies that Jimerson analyzed (2001) is highlighted by Lorence
(2006) because of a unique finding based on its design. The study (Pierson & Connell, 1992)
compared 74 retained students with 69 promoted peers in a same-grade promoted group matched
by IQ and sex. Additionally, 35 students who were promoted, although they should have been
EFFECTS OF RETENTION 23
retained, were matched with the retained students. The significant result of this particular study is
that the retained students’ academic achievement scores were similar to their peers who had
compatible IQ scores, but they exceeded their socially promoted peers (Lorence, 2006). Since
scores were an aggregate of grades and test scores over several years, it is unknown when the
positive effect of retention for these students occurred. Lorence (2006) believes more weight
should be accorded to these results because this study involves students who should have been
retained, but were promoted, as the matched group: Although not random, they were clearly
more closely related at the time of the retention study.
Based on this methodological analysis of Jimersons’ meta-analysis (2001), Lorence
(2006) again concludes that the results should not be touted as a ringing endorsement of
promotion over retention.
Methodological Considerations for New Retention Research
Lorence (2006) raises some important research design questions. Although the authors of
individual studies and meta-analyses make conclusions about the value of retention or
promotion, there is little agreement about what constitutes a successful retention. He suggests
that researchers agree on the goal of retention in order to measure its effectiveness: Is a
successful retention an increase in the student’s performance; or performance that is equal to his
promoted peers; or performance that exceeds his grade peers; or performance compared to
socially promoted low-achieving peers? Further, in order to be more accurate, studies of
retention should include the specific instructional practices that were provided to retained
students in order to examine whether it is the retention or the instruction post retention that
improves student outcomes (Lorence, 2006).
EFFECTS OF RETENTION 24
Recent Methods to Correct Design Deficiencies
In order to circumvent some of the pitfalls related to non-randomized retention studies,
Wu et al. (2008) recently created matched groups using propensity score analysis in their study
of first grade students who had been retained. In order to form matched groups, students who
were retained were matched with promoted students based on a difference in propensity scores of
no more than .025.
Propensity scores are based on a wide variety of baseline measures from the participants
related to the outcome variable and the treatment measure. A statistic then assigns each
participant a single propensity score that predicts the probability that each student will be in the
treatment group. The larger the propensity score, the larger the predicted probability that the
child will be retained. This process reduces group selection bias (Hong & Raudenbush, 2005;
Lorence, 2006; Wu et al., 2008; Peterson & Hughes, 2011).
The other methodological problem related to retention studies, comparing same-age
versus same-grade growth, was addressed by collecting four years of data on both grade level
standard scores and Woodcock-Johnson III W scores, which are nationally standardized and
comparable to age norms. Data included measures that were given .5 years prior to retention and
up to 2.5 years after retention (Wu et al., 2008).
Using growth curve modeling, a process that estimates each child’s learning growth
trajectory, the researchers were able to examine the short term and long term effects of retention
and found mixed results. In the area of reading, retained students grew more slowly than
matched promoted peers in the first year after retention, but grew at a faster rate than their
matched group so that there was no significant difference in reading ability by third grade for
retained students, or fourth grade for promoted students (Wu et al., 2008). Similar effects were
found for math, but the gains for retained students were not as great. When allowing for the
EFFECTS OF RETENTION 25
moderating factors that were measured, Wu et al. (2008) found that the short-term benefit of
retention was greater for students who did not have strong parental involvement in the schools,
or for students who had fewer pro-social skills.
Also using propensity scores, Hong & Raudenbush (2005) decided to approach the
problem of measuring the effect of retention on students from a slightly different approach. They
compared groups of students at schools whose policies changed to allow for retention and those
whose policies did not allow for retention. Comparing 1,080 schools with retention policies that
included 471 retained students and 10,255 promoted students, and 141 schools that did not retain
students with a total of 1,117 promoted students, they collected student performance data four
times over two years. Propensity scores based on 207 pre-treatment covariates were used to
balance the final group of 407 retained students with promoted students at schools that allowed
retention (Hong & Raudenbush 2005).
The results of their study showed that students attending a school with a retention policy
did not perform better in reading or math when compared with schools that did not have a
retention policy. In analyzing data against growth trajectories, they found that retained students
fared worse in both reading and math than non-retained students over time. The growth model
also indicated that the students would have achieved more had they been promoted, although
they still would not have achieved at as high a level as those who had never been at risk for
retention (Hong & Raudenbush, 2005).
International Findings
In a monograph for The International Academy of Education, Brophy (2006) describes
five types of retention: two of a voluntary nature most often occurring in developing nations, and
three common practices in more developed nations. Students in developing nations may not have
local access to schools that have more than a few grades, or they attend school sporadically
EFFECTS OF RETENTION 26
because they work so much to provide for their family’s basic needs or provide childcare while
their parents work. In both of these situations, parents may voluntarily have their child repeat
grades so that they can either continue to learn or to make up what they have missed due to
absences.
In developed nations, student retention is more common where the student lacks language
ability in the native language of the school; where students have to pass high stakes exams to
qualify for secondary or post-secondary schooling; and finally, where students are failing and the
school initiates the retention (Brophy, 2006). Unlike the United States, where most retentions
happen in the earlier grades, Brophy (2006) notes that nations such as Burundi and Kenya that
have competitive exams for secondary school see more retention in the later years of elementary
school. These retentions are not based on low performance; rather, students who have a higher
academic potential are retained to give them more opportunity to prepare to compete for the
limited openings in secondary institutions.
Taylor et al. (2010) have described in detail how absences due to children providing day
care to siblings have greatly increased the number of over-age students in the classroom, so that
one-third of fourth grade South African students are over-age due to retention or lost years of
schooling. They find a pattern of low performance in their overage students similar to retained
students, when compared to grade and age peers in South African schools. In their nationwide
study of fourth grade literacy, nine- and ten-year-old students outperformed both the 11-year-old
and 12-year-old students significantly (t=3.86).
Studies of first grade retainees in Flanders, using propensity scores on a same-grade
analysis, indicate that students still show a performance disadvantage at the end of elementary
EFFECTS OF RETENTION 27
school. A same-age comparison shows the retained students would have performed better on
both a short term and long term basis had they not been retained (Goos et al., 2011).
Martin (2011) studied 3261 retained high school students in Great Britain, using
structural equation modeling in order to review both the academic and social-emotional effects
of retention. He found a significant main effect of retention (p<.001) for predicting both negative
academic factors and positive maladaptive motivation factors.
Policy Considerations
California Assembly Bill (AB) 1626, the Pupil Promotion and Retention Law, calls for
schools to have retention polices for students who do not master grade level expectations based
on specific criteria (California, n.d.). California Education Codes (EC) flesh out the requirements
that Local Education Agencies (LEA) must include (California, n.d).
EC 48070.5(a) specifies that the policy applies to students between second and third
grade; third and fourth grade; fourth and fifth grade; at the end of the intermediate grades and the
beginning of middle school grades; and at the end of middle school grades and the beginning of
high school. Criteria for retention or risk of retention should be based on the results of the State
Testing and Reporting (STAR) results and district designated academic achievement measures
(EC 48070.5 (b)). Decisions related to retention at the end of second and third grade should be
made primarily on the basis of reading proficiency; for the other designated grades, the decision
should be based primarily on proficiency levels in reading, English language arts, and
mathematics (California, n.d.).
Furthermore, EC 48070.5(d) gives teachers (who are deciding not to retain a child who
qualifies for retention based on state and LEA policy) the responsibility to identify which
interventions are necessary to assist the student to attain proficiency levels. It also puts the onus
EFFECTS OF RETENTION 28
on the teacher to provide this evaluation to the parent and the principal prior to the child being
promoted.
Parents have the right to be notified early in the year that their child is at risk for retention
(EC 48070.5 (e)), and may appeal that decision to the School Board with the burden on the
parent to show why the teacher’s decision should be overruled (EC 48070.5 (f)). Since the
intention of AB1626 is to ensure that students are well-prepared for the rigor of the new grade
level, EC 48070.5 (h) also mandates that the LEA is responsible for providing opportunities for
remedial instruction for students who are at risk for retention or who are retained (California,
n.d.).
It is interesting to note that California AB 1626 and EC 48070.5 do not list social or age
considerations as a reason for designating who is at risk for retention or who should be retained,
when both of those factors are commonly cited by school personnel and parents as a reason for
retention (Smith & Shepard, 1988; Jimerson & Kauffman, 2003; Hong & Raudenbush, 2005;
Willson & Hughes, 2005; Wu, West & Hughes, 2008). EC 48070.5 (d) does give the teacher
discretion to specify in writing that retention is not the appropriate intervention for a student, but
it does not specify any other grounds for the teacher to recommend retention on any grounds
other than meeting the adopted academic performance standards. However, a loophole is
provided by EC 48070.5 (j), that states that:
Nothing in this section shall be construed to prohibit the retention of a pupil not
included in grade levels identified pursuant to subdivision (a), or for reasons other
than those specified in subdivision (b), if such retention is determined to be
appropriate for that pupil.
EFFECTS OF RETENTION 29
No guideline is given for how the appropriateness of retention should be decided, and as
we have seen above, the California Department of Education has conducted no studies to
determine if retentions result in academic success, and there is much debate in the research as to
the appropriateness of retention.
Regent Union School District (RUSD) has complied with AB 1626 and EC 48070.5 by
creating and adopting Administrative Regulation (AR) 5123 in 1999 (see Attachment C). Their
policy has clear guidelines for identifying, in all grade levels, which students are moderately at
risk for retention, significantly at risk for retention, and which students should be retained. AR
5123 provides timelines and forms for notifying parents of retention risk as well as opportunities
for teacher and schoolwide intervention.
Intervention Services and Retention
Teachers are often the gateway to intervention and/or retention (EC 48070.5; Beebe-
Frankenberger, Bocian, MacMillan & Gresham 2004; Peterson & Hughes, 2011). However, if
teachers value retention because they believe it will help a child emotionally and academically,
and will increase student confidence, they often do not recommend retained students for summer
school intervention programs because they believe the child has the entire next year to catch up
(Peterson & Hughes, 2011.).
Peterson & Hughes (2011) have conducted research to investigate the level of educational
services offered to promoted low-performing students, as well as retained students, in the years
pre- and post-retention. They contend that since research by D’Agostino (2004) and Ehri,
Dreyer, Flugman, & Gross (2007) have demonstrated the success of small group and
individualized, structured reading programs for improving student achievement, such
interventions could stem the need for retention (Peterson & Hughes, 2011).
EFFECTS OF RETENTION 30
They conducted a longitudinal study of 760 promoted and retained first grade students in
Texas. The students were administered the Woodcock-Johnson III Broad Reading and Math
assessments. The researchers surveyed teachers regarding the provision of the following
academic services: reduced class size, one-to-one tutoring by an adult during the school day,
remedial instruction outside of the classroom during the school day, tutoring by a peer or older
student during the day, remedial instruction before or after school, small group intensive
tutoring, and one-to-one tutoring by an adult before or after school. These services were not
provided to all students (Peterson & Hughes, 2011). Additionally, students were assigned
propensity scores based on 72 predictor variables for academic achievement in order to provide a
more rigorous test of association between retention status and services that were received.
Peterson and Hughes (2011) found that when students were in first grade for the first
time, the students who were subsequently retained or promoted did not differ in the number of
remedial services they received. However, during the second year, students who were retained in
first grade received fewer remedial services than their promoted at-risk peers who were now in
second grade. The authors note that the effect of this practice was that retained students may
have been deprived of remedial services, despite the fact that their retention was based on the
acknowledgement that they were in need of remediation. Retention functioned as the
intervention for academic risk, even though that may not have been the intention (Peterson &
Hughes, 2011).
Beebe-Frankenberger et al. (2004) have studied the ways in which retention serves as an
intervention or a holding pattern for students who evidence academic risk. In their study, over
half of the students who later qualified for special education services were initially retained.
EFFECTS OF RETENTION 31
Retention, in the lives of these students, served as a first step intervention before they were
assessed for a learning disability.
Hypothesis
Given that earlier research on the effects of retention may have had serious flaws, one
might discount researchers’ conclusions that retention is not beneficial (Hong & Raudenbush,
2005; Lorence, 2006; Wu et al., 2008). However, more recent studies that have corrected those
design flaws still show limited value, of a short-term nature only, in retaining students (Hong and
Ruadenbush, 2005; Wu et al., 2008; Peterson & Hughes, 2011). The results showing the lack of
benefit of retention are found elsewhere internationally, also. Based on these findings, I believe
that a study of the academic effects of retention on students in the Regent Union School District
will show that retention is not academically beneficial. The study will not encompass the
socioemotional effects of retention on the student.
EFFECTS OF RETENTION 32
Chapter Three: Methodology
The California Promotion/Retention Law changed as of January 1999. Education Code
48070 required School Boards to adopt policies that would end social promotion (see Appendix
A). Schools were to identify students who were not meeting grade level proficiency
requirements, or were at risk of not meeting grade level proficiency requirements, and offer
additional intervention. If students still did not meet proficiency, they would be retained. Regent
Union School District (RUSD) has maintained extensive data on students who were identified at
risk and retained, but to date no extensive formal analysis of that data had been performed: it was
unknown whether retention benefitted the students or not. The purpose of this research was two-
fold:
Study 1: To conduct a Retrospective Control Group Study examining the effects of
retention on student academic achievement.
Study 2: To analyze the demographics and academic achievement of students who were
retained from 2005-2011.
A detailed analysis of retained students’ performance data compared to non-retained
students’ performance data allowed us to verify that retention did not increase student
achievement. This study examined historical data in order to answer the following research
questions:
1. How did K-2 retention of an RUSD student affect the student’s future academic
achievement?
2. What was the correlation between selected demographics and retention?
3. What are the policy implications of this study?
EFFECTS OF RETENTION 33
Quantitative Research Design
Study 1: Pre/Post Independent Retrospective Control Group Design
The change in California law in 1999, which brought social promotion to an end,
presented the opportunity to create a retrospective control group. There were students in
Kindergarten through second grade in 1998-1999 who would have been retained under the new
law had they been students in those same classrooms in 1999-2000. By applying the retention
criteria instituted in 1999-2000 to the students in 1998-1999, we created a matched retrospective
control group for the students who were retained in 1999-2000 (our experimental group). A
retrospective quasi-experimental design was used to analyze the effect of retention on academic
and effort performance for K- second grade students in the district.
The Pupil Record Card (PRC) is a permanent record of a student’s progress during their
years in the district (see Appendix B). At the end of each year, the classroom teacher recorded
each student’s reading and math level as Above Grade (Ab), At Grade (At), or Below Grade
(Bel). They also recorded effort in both areas as Outstanding (O), Good (G), Satisfactory (S),
Needs to Improve (N), or Unsatisfactory (U).
To develop the academic and effort profile of the retained students group, I first analyzed
the PRC data for the 1999-2000 retained Kindergarten, first and second grade students (N=129):
These students comprised the Intervention Group. Next I reviewed the PRC of all non-retained
Kindergarten, first, and second grade students from 1998-1999 to select the students who
matched the profile of the retained group. All 1998-1999 Kindergarten, first and second grade
students who matched the 1999-2000 retained group profile comprised the retrospective control
group. Matched pairs were created using propensity scores based on the 1998-99 and 1999-2000
PRC grades.
EFFECTS OF RETENTION 34
The PRC scores for reading, math, and effort for both groups as well as STAR reading
and Math scores were compared in their sixth grade year in order to study the effects of
retention. A Dependent Groups t-test was used to assess for statistical significance (p < .05). This
is a graphic representation of the non-equivalent retrospective control group design, where X
represents retention as the intervention:
Retrospective Control Group Intervention Group
1998-1999 1999-2000

0_____________X______________0

PRC academic and effort scores in reading and math served as the pre-test data, and PRC
academic and effort scores in reading and math as well as STAR scores in Reading and Math
from 6
th
grade served as the post-test data.
Retrospective Control Group Intervention Group
Pre (1998-1999) X Post (1999-2005)
PRC scores: PRC scores:
Reading Reading
Reading Effort Reading Effort
Math Math
Math Effort Math Effort
6
th
grade STAR
Reading
Math

Study 2: Correlation Design
For the purposes of conducting a robust correlation study of demographics and retention
of students enrolled in 2011-12 who had been retained (N=179), district databases and student
cumulative files were gleaned for student demographic, program, and achievement data. Each
separate category of data was analyzed in order to ascertain correlation to retention.
EFFECTS OF RETENTION 35
Participants and Setting
RUSD is a K-6 school district located in Northern Los Angeles County in a city of
176,971 people. Within the city, 28% of the population is under the age of 18, and 30% are
between the ages of 25 and 44. It is a growing city: the population increased 17.5% from 2000-
2010. Average income is $82,602 and 73% of the population owns their homes. The median
home price in March 2011 was $330, 000. Of adults 25 years or older, 65% have had some
college. Ethnically, 76% are white, 14.2% Hispanic, 1.7% Black, 4.8% are Asian, and 2.7%
multiracial.
During the 1998-1999 school year RUSD had an enrollment of 8559 students, including
1217 kindergarten through second grade students. Fifty-four kindergarten through second grade
students were retained that year. In 1999-2000, under the first year RUSD implemented the
Promotion and Retention law, total district enrollment was 8959, and 129 kindergarten through
second grade students, out of a total of 1267 students, were retained. Total k-6 student retention
in 1998-99 was 0.74% of district enrollment. Retention increased and peaked at 3.32% of the
kindergarten through second grade enrollment and 1.85% of the district enrollment in 1999-
2000. Retention has decreased slightly since then: 2010-2011 statistics show a district enrollment
of 10,423 with a 1.82% retention rate.
Table 1

Comparison of RUSD Retention Rates

Year Number of
students
Kindergarten-
second grade
Number of students
retained in
Kindergarten-second
grade (Percentage)
Total
number of
students in
all grades
Total number of
students retained
in all grades K-6
(Percentage)
1998-1999 3822 54 (1.41%) 8559 63 (0.74%)

EFFECTS OF RETENTION 36
1999-2000 3875 129 (3.32%) 8959 166 (1.85%)

2010-2011 4380 23 (0.53%) 10,423 190 (1.82%)

Academic Intervention
Many teachers include regular academic or behavioral classroom interventions for
students who are not meeting grade level expectations. However, retention is often times
recommended when sufficient progress is not made even after classroom interventions have been
tried. In both Study 1 and Study 2, retaining a student for a second year in their current grade
level was an intervention.
Prior to 2000, district personnel made the decision to retain a child based on the parent’s
request or the recommendation of the teacher. It was the custom to hold several parent-teacher
meetings throughout the year to discuss the needs of the child when they were not meeting grade
level expectations as measured by teacher-made or publisher-made tests. Often teachers felt that
a child’s immaturity was a reason to retain, even if the child was meeting grade level
requirements. Parents who agreed to have their child retained in Kindergarten signed a
continuation form, since law required parental permission. At other grade levels, the decision to
retain was legally the school’s decision, but staff considered parents’ preferences when
recommending retention. Not meeting grade level expectations or emotional immaturity
continued to be the main reasons for retention. Parental agreement was not required for retention
in first through sixth grades.
After Education Code 48070 was passed in 1999, the district created
Promotion/Retention Guidelines (see Appendix C) in order to comply with the law. The intention
of the guide was to standardize the way teachers identified at-risk students and to identify
EFFECTS OF RETENTION 37
resources they could use for intervention in order to prevent retention. Initially, however, it
resulted in more students being retained as teachers and administrators implemented the new
protocol.
The first RUSD Promotion/Retention Guide used California Achievement Test-6 scores,
which was the State Accountability Measure at the time. The updated version of the RUSD
Promotion/Retention Guide uses STAR scores, with Far Below Basic in Math and Reading as
the criteria for retention in second through sixth grade. The Intervention group participants in
Study 1 and all of the participants in Study 2 were retained using the updated criteria.
Table 2
2000 RUSD Guidelines for Retention by Grade Level
Grade Level Criteria for Retention
Kindergarten
through
Third Grade
Identified as at-risk for retention:

 Below basic on Reading/Language Arts multiple measures:
 Report Card “Requires Additional Support” checked for both areas in
Reading
 Writing Sample – rubric of 0
 CAT 6 – below 16th percentile on Total Reading and Language Arts
 More than 25 days absence
Fourth
Through
Sixth Grade
Identified as at risk for retention:

 Below basic on Reading/Language Arts and Math multiple measures
 Report Card Grade N and checks in Reading Word Attack and
Comprehension and Math Computation, Problem Solving/Application
and Mathematical Communication
 Writing Sample – rubric of 0
 Math CRTs – below 50 percent
 CAT 6 – below 16th percentile on Total Reading, Language Arts and
Math
 More than 25 days absence

EFFECTS OF RETENTION 38
Instrumentation and Procedures
Study 1
Study 1 is a Retrospective Control Group study comparing achievement of students who
were retained in Kindergarten through second grade in 1999-2000 with a control group of
Kindergarten through second grade students who would have been retained in 1998-1999 if the
same criteria for retention had been applied. Data from this study was collected from the PRC
maintained by the district (Appendix B).
Pupil Record Card and STAR scores were coded as shown in Table 3.
Table 3
Pupil Record Card and STAR Variables Coding
Reading and Math Achievement Scores  Below Grade Level=1;
 At Grade Level=2;
 Above Grade Level=3
Reading and Math Effort Scores  Needs Improvement =1;
 Satisfactory =2;
 Good=3;
 Outstanding=4.
STAR Language Arts and Math Scores  Far Below Basic=1
 Below Basic=2
 Basic=3
 Proficient=4
 Advanced=5

Study 2
Study 2 is a correlation study of specific demographic information of students retained
each year from 2005 through 2011. Demographic details were taken from the enrollment forms
(Appendix D) completed by parents when the student was first enrolled in an RUSD school,
regardless of the year they enrolled. These variables were coded as listed in Table 4.
EFFECTS OF RETENTION 39
Table 4
Coding of Enrollment Form Variables
Variable Code
Gender Female=1; Male=2
Birth Quarter
(Based on State
Kindergarten
Enrollment Cut-off
Date)
December 3-February = 1;
March-May = 2;
June-August = 3;
September-December 2 = 4.
Family
Composition
Child lives
With neither parent = 0;
With mother and father =1;
With mother only = 2;
With father only = 3;
With parent and other adult =4.
Birth Order Only or oldest child= 1,
Second eldest = 2,
and so on.
Number of Children
in the Home
Only child = 1;
If one of two = 2;
If one of three = 3,
and so on.
Parent’s Education
Level
(Recorded
separately for each
parent)
Not a High School Graduate = 0;
High School Graduate =1;
Some College (includes AA) =2;
College Graduate =3;
Graduate School =4.
Parental
Employment
If neither parent is employed = 0;
If mother is employed =1;
If father is employed =2;
If both parents are employed =3.

EFFECTS OF RETENTION 40
Table 4
Coding of Enrollment Form Variables (continued)
Variable Code
Race Alaskan Native =1; American Indian = 2; Black or African
American = 3; White= 4; Guamanian =5; Hawaiian =6;
Samoan =7: Tahitian =8; Other Pacific Islander = 9; Asian
Indian =10; Cambodian= 11; Chinese =12; Filipino= 13:
Hmong =14; Japanese = 14; Korean = 16; Laotian =17;
Vietnamese = 18; Other Asian =19; Hispanic=20.
English Language
Learner Status
English Language Learner =0;
Reclassified Fluent English Proficient=2;
Initial Fluent English Proficient=3;
English Only=4
Prior District If a child has only been enrolled at an RUSD school =0;
If a child was previously enrolled at another K-6 school =1.

Information from the students’ cumulative files was coded in accordance with Table 5.
Table 5
Coding for Variables from the Cumulative File
Variable Code
Student Study
Team (SST)
The parent did not participate in an SST the year the child was retained =0;
A parent participated in an SST the year the child was retained =1;
A parent participated in two =2
and so on
Previous SST The parent did not participate in any SST meetings prior to the year the
child was retained = 0;
A parent participated in one =1;
A parent participated in two =2,
and so on.
At Risk
Notification
The parent was not notified of the risk for retention prior to the retention
decision being made=0;
The parent was notified =1.

EFFECTS OF RETENTION 41
Table 5
Coding for Variables from the Cumulative File (continued)
Variable Code
Intervention No evidence that the child was offered school based
intervention the year he was retained =0;
If it was offered=1.
Prior Intervention No evidence that the child was offered school based
intervention the year prior to which he was retained =0;
If it was offered=1.
Post-retention Intervention No evidence that the child was offered school based
intervention the year following retention=0;
If it was offered=1.
Individualized Education Plan
(IEP) testing:
If the child has an IEP, list the date he was tested and qualified.
Did Not Qualify If the child was tested for an IEP but did not qualify, list the
date of the testing.
Title I School If the Child does not attend a Title I school=0;
If he does=1.

The data in Table 6 was collected from the last reporting period of the student’s Progress
Report Card the year they were retained and again at the end of the 2011-12 school year.

EFFECTS OF RETENTION 42
Table 6
Progress Report Code Variable Coding
Variable Code
Reading
Composite Score
K-3 grade:

Phonics , Sight Vocabulary (K only), and Comprehension
(each coded separately as follows):
 Requires Additional Support=1;
 Works Within Grade Level Standards =2;
 Demonstrated Strengths=3.

Effort in Reading:
 Needs Improvement =1;
 Satisfactory =2;
 Good =3;
 Outstanding=4.

4-6
th
Grade:

Reading Achievement:
 Needs Improvement =1;
 C=2;
 B=3;
 A=4

Reading Effort:
 Needs Improvement =1;
 Satisfactory =2;
 Good =3;
 Outstanding=4.
Work/Study
Habits
Composite Score
Listens and follows Instructions; Stays on Task; Completes Work
Accurately; Completes Class Work on Time; Completes Homework on
Time; Uses Materials Correctly; Uses Organizational Skills; Completes
Work Neatly; and Participates in Classroom Activities
(each coded separately as):
 Needs Improvement =1;
 Satisfactory =2;
 Good =3;
 Outstanding=4.

EFFECTS OF RETENTION 43
Table 6
Progress Report Code Variable Coding (continued)
Variable Code
Citizenship
Composite Score
Exercises Self Control; Respects Authority; Respects Others’ Rights and
Property; Observes Rules; Cooperates with Others; Accepts Responsibility
(each coded separately as):
 Needs Improvement =1;
 Satisfactory =2;
 Good =3;
 Outstanding=4.
Absences and
Tardies
Actual Number of Absences and Tardies recorded separately

Number of Suspensions data is kept in an RUSD database. Suspensions were coded using
the date of the suspension and the reason for the suspension.
STAR scores for Reading were coded as Far Below Basic =1, Below Basic =2, Basic =3.
Proficient =4, and Advanced =5, using the scores for second through sixth grade.
CELDT scores were coded Beginner=1, Early Intermediate=2, Intermediate=3, Early
Advanced=4, and Advanced =5.
National Subsidized Lunch Plan (NSLP) status was used as the indicator for
economically disadvantaged. RUSD keeps this information in a confidential file, separate from
all other student information. Not a participant =0; Participant=1.
The University of Southern California’s Institutional Review Board and the RUSD Board
of Trustees approved both Study 1 and Study 2 in March, 2012. Data collection began in May,
2012. For Study 1, I first coded the PRC for each Kindergarten through Second grade student
retained in 1999-2000, and for each non-retained Kindergarten through second grade student in
1998-1999. I analyzed the 1999-2000 PRC student data to create a profile of the students (the
EFFECTS OF RETENTION 44
experimental group) that were retained. Using that profile, I analyzed the PRC of non-retained
Kindergarten through second grade 1998-1999 students and selected students using propensity
scores to match the experimental group. This matched comparison group served as the
retrospective control group. In addition to the 6
th
grade PRC achievement data, I coded data
related to performance on 6
th
grade STAR Reading and Math scores. Using SPSS, I ran an
dependent group t-test to analyze the effect of retention on academic performance and effort for
K- sixth grade students where p<.05.
For Study 2, data from the enrollment forms, report cards, and SST forms were retrieved
from the cumulative files of each current RUSD student who had been retained. Information
concerning their STAR scores, NSLP participation, CELDT levels, IEP testing or current IEP,
and number of suspensions were queried from the RUSD data base for each retained child. All of
the information listed above was coded for each child, and data was run through the Statistical
Package for the Social Sciences Software (SPSS) in order to find the correlations related to
retention.
Analysis
Study 1 is a Retrospective Control Group study where the pre/post test comparison used
two different sets of students and their achievement scores. The appropriate test to run was the
dependent groups t-test to compare the control and experiment groups. The confidence level I
looked for on this test was p<.05.
Study 2 is a correlation study of twenty-seven categories of continuous data related to
students enrolled in RUSD in the 2011-2012 school year, who had been retained in Kindergarten
through sixth grade. I ran the frequencies, crosstabulations, and correlation coefficient statistical
tests to measure the statistical significance of each category, where p< .05.

EFFECTS OF RETENTION 45
Limitations (Threats to Validity)
Internal Validity
Study 1
One threat to internal validity was selection bias in the control group. Because it was
impossible to do a randomized trial in an area as socially significant as a child’s promotion or
retention, we were not able to use the same group of students for our pre and post -tests. Since
our intervention group had already been retained, our best means of creating a control group was
to retrospectively assign students to a control group based on propensity scores. If we had been
able to randomly choose who to retain, and then randomly chose a control group, we would not
have to question the internal validity of the study.
Study 2
The correlation of the selected demographic data for the study may indicate that there is a
relationship between a particular variable and retention, when in fact they may be correlated but
there is no implied causation. For example, although there is a high correlation between birth
order and retention, we cannot conclude that birth order causes retention, when in fact another
variable that was not studied could have been responsible for the correlation between birth order
and retention.
External Validity
Study 1
The results of this study are delimited to the sample of students, this particular school
district, and the measures that were chosen; therefore the demonstrated results may not hold true
for other populations. The study uses measurement tools that are specific to the culture of
assessment that exists in this district. Pupil Record Card grades are based on scores from
assignments that the teacher chooses to use for the achievement grades. Although there may be
EFFECTS OF RETENTION 46
some agreement across a grade level of what grades should be used, there is still wide variation,
and the individual teacher makes the decision of which and how many assignments or tests to
include. The data is also over ten years old, and we now live in a time that is much more focused
on grade level standards. Although the PRC could be completed using any in-class measurement
chosen by the teacher, teachers are currently more likely to use academic Content Standards as a
measure of grade level ability than they were in 1999 or 2000. The use of STAR scores is highly
generalizable in California, but would not be generalizable in other states. Since another district
could not implement the study in the same way, we cannot generalize beyond the treatment.
The Effort grades on the PRC are somewhat subjective: it is the teacher’s perception of
the amount of effort the child expended in that subject area. More experienced teachers may be
looking at similar student behaviors but assigning a different effort grade. Another researcher
using the same categories to measure student achievement and effort might very well be
interacting with teachers who hold very different views on how to rate the students in each of
these subjective areas.
Study 2
This study presents similar threats to external validity. The Report Card grades are based
on the teacher’s measure of the child’s ability and effort relative to her understanding of what
constitutes grade level performance. Because they could be completed using any in-class
measurement chosen by the teacher, there is little standardization in assigning the report card
grades. Other districts could thus not implement the study in the same way, and we cannot
generalize beyond the treatment.
EFFECTS OF RETENTION 47
Statistical Conclusion Validity
Study 1
In order to create a retrospective control group, teacher measures of achievement and
effort were used. This may affect the reliability of the measurements. The study had a large
intervention sample of 92 retained students and a matched retrospective control group of 1099
students. Due to missing scores on some of the measures, the final matched groups (based on
propensity scores) ranged from 68to 72, depending on the measure used for the t-tests. Although
statistical validity may have been affected by sample size, power was increased due to the
matched design using propensity scores.
Study 2
The two areas of concern for statistical validity in Study 2 are the reliability of the
measures and the sample size. As discussed previously, some of the measurements in this study
are based on teacher perspective of effort and achievement. Because these may vary from teacher
to teacher, they do not have the same statistical validity as CST scores, which are more
established. However, the study does have a sufficiently large sample size (n=10,423) and
intervention group (n=172) to ensure statistical validity.
EFFECTS OF RETENTION 48
Chapter Four: Results
During conferences to decide whether or not to retain a child, usually two questions are
asked. The first is, “If we retain him, will he perform better in school than if we don’t?” The
answer to this question is difficult to come by, since it requires analyzing the child’s performance
against the performance of students with a similar academic profile who were not retained. Study
1 addresses that question by identifying the group of 1999-2000 RUSD k-2 students who were
retained under the 1999 California law that banned social promotion, and used propensity score
to assign a retrospective control group of k-2 students from 1998-99 who would have been
retained if the law had been adopted a year earlier.
The other question often asked at the retention conference is, “If we retain this child, will
he then be successful in school?” In terms of academic success, the parent uses the report card
that comes home three times a year as their measure of academic success. The teacher uses three
measures: the report card reflecting classroom performance; the annual California Standards Test
(CST) where students are expected to perform at a Proficient or Advanced level in order to be
considered at grade level; and if the child is an English Language Learner (ELL), the student’s
performance on the annual California English Language Development Test (CELDT). A review
of the student’s report card or CST and CELDT scores at any point in their elementary academic
career can address the current level of academic success. These tests can also be used to compare
the student’s performance against non-retained students who took the same test. Study 2 was
designed to address the 2
nd
question by examining the academic success of currently enrolled
retained k-6 RUSD students through this prism. The combined results of the two studies will
provide guidance in retention discussions by answering the following research questions:
EFFECTS OF RETENTION 49
1) How does retention of an RUSD student in kindergarten through second grade affect the
student’s future academic achievement?
2) What is the correlation between selected demographics and retention?
3) What are the policy implications of this study?
Research Question 1: How does retention of an RUSD student in kindergarten through
second grade affect the student’s future academic achievement?
Study 1
Six sets of grades were collected from the Pupil Record Cards (PRC) of RUSD students
who had been retained in kindergarten, first grade, or second grade at the end of the 1999-2000
school year. These scores included Reading Achievement, Reading Effort, Math Achievement
and Math Effort. Their sixth grade CST scores in Language Arts and Math were also gathered.
Collectively, these scores represented the experimental group’s levels of academic performance.
The same data were gathered for the comparison group, which was made up of non-retained
students who attended RUSD kindergarten, first grade, and second grade in 1998-1999.
A linear regression of the end of year Kindergarten through 2
nd
grade PRC scores of 92
retained students and 1099 comparison group students was used to assign each student a
propensity score. The retained students were then matched to a comparison student using the
closest propensity score. A dependent t-test was run comparing all sixth grade PRC scores and
CST scores. Final n=68, 70, or 72 depending on which outcome was analyzed.
Results are seen in Table 7. The PRC score is the teacher’s subjective view of the
student’s academic ability and effort. In the academic areas of Reading and Mathematics, the
mean of the retained group is less than the mean of the comparison group, but p>.05, indicating
that retention did not benefit the students’ academic performance. The mean of the retained
EFFECTS OF RETENTION 50
students in both Reading Effort and Math Effort is greater than the mean of the comparison
group, p>.05, indicating that it is not statistically significant.
Table 7

Statistical Results of the t-Test of Pupil Record Card (PRC) Scores

Mean n Standard
Deviation
Correlation t df Significance
6th Grade PRC
Reading Achievement
Grade
(Retained Group)

6th Grade PRC
Reading Achievement
Grade
(Comparison Group
1.81

1.92
36

36
.58

.55
-.23 -.75 35 .457

6th grade PRC Reading
Effort Score
(Retained Group)

6th grade PRC Reading
Effort Score
(Comparison Group)
2.40

2.37
35

35
.78

1.03
-.231 .12 34 .906
6th grade PRC Math
Achievement
Grade (Retained
Group)

6
th
grade PRC Math
Achievement Grade
(Comparison
Group)
1.64

1.86
36

36
.54

.54
-.175 -.1.6 35 .118
6th grade PRC Math
Effort Score
(Retained Group)

6th grade PRC Math
Effort Score
(Comparison Group)
2.39

2.33
36

36
.84

.96
-.91 .25 35 .804

EFFECTS OF RETENTION 51
The CST scores are obtained from a standardized test that students took at the end of their
elementary school experience, (6
th
grade). Outcomes of the t-test comparison are reported in
Table 8. The mean of the retained group on the Language CST is less than the mean of the
comparison group, indicating that retention did not increase the retained group’s Language Arts
performance on a standardized test (p>.05). Although the retained group’s mean on the Math
CST is slightly higher than the comparison group’s mean, it is not statistically significant
(p>.05), indicating that retention did not increase academic performance in math.
Table 8

Statistical Results of the t-Test of California Standards Test (CST) Scores

Mean

N Standard
Deviation
Correlation t df Significance
6th grade CST
Language
Arts (Retained
Group)

6th grade CST
Language
Arts
(Comparison
Group)
3.31

3.54
35

35
.93

1.04
-.60 -.94 34 .353
6th grade CST
Math
(Retained
Group)

6th grade CST
Math
(Comparison
Group)
3.26

3.24
34

34
1.02

.96
-.04 .12 33 .904

Study 2
Study 2 examined the performance of currently retained K-6
th
grade students on the CST,
the CELDT, and classroom report card grades. The purpose here was threefold: (1) to compare
EFFECTS OF RETENTION 52
the retained students’ post-retention performance on CST with their non-retained peers; (2) to
compare retained ELL students’ performance on CELDT against themselves, and also against
their non-retained ELL peers; and (3) to see how teachers rated the retained students’ classroom
performance during the retention year and again during the most recent year.
Findings in Regards to CST Performance
Students retained in second through fourth grade took the CST at the end of both their
retained year and their repeated year. The CST Language Arts scores from their repeated year
were compared in a matched design against the mean scores of their current non-retained peers
using a dependent t-test. Table 9 presents the results.
Table 9

Statistical Results of the t-Test of Language Arts CST Scores

Mean N Std.
Deviation
Correlation t df Significance
Repeated Year
Language Arts
CST scores

Non-retained Peer
Group Language
Arts CST scores
2.92

3.98
24

24
1.34

.24
.04 -3.83 23 .001

The academic performance on the CST Language Arts, measured by the t-test, indicates
that the mean of the retained group was less than the mean of their non-retained peers with a
significance level of p=.001. The Effect Size (ES) equals .48. This indicates that even though the
retained group had had an additional year of instruction at that grade level compared to their non-
retained peers, their performance did not meet the level of their peers. Thus, retention did not
benefit them academically when compared to their peers.

EFFECTS OF RETENTION 53
Another way of viewing retained students’ Language Arts performance post- retention is
to compare the frequency with which they scored in each performance band of the CST
compared to all RUSD students (Table 10).
Table 10

A Comparison of Retained Students 2011-12 CSTs with All Performance Levels with All Non-
Retained RUSD Students

Retained Students All RUSD Students
Far Below Basic 6(4.8%) 135(1.9%)
Below Basic 17(13.6%) 265(3.82%)
Basic 37(29.6%) 1091(15.7%)
Proficient 33(26.4%) 2223(32%)
Advanced 32(25.6%) 3215(46%)

Post-retention, retained students were more than twice as likely as their peers to score in
the Far Below and Below Basic levels than their non-retained peers. They were less likely to
score Proficient or Advanced than their non-retained peers. Only 52% of the retained students
met the State and Federal cut-off for grade level competency (Proficient or Advanced), while
78% of their non-retained peers met that standard of academic performance.
Findings in Regards to CELDT Performance
Given that the CELDT is a standardized test that can compare year over year growth, two
separate tests were run. The first was a t-test that measured the growth of the individual student
in a pre-post design using their scores from the Fall at the beginning of their repeated year and
their own scores exactly one year later. Table 11 summarizes the results.

EFFECTS OF RETENTION 54
Table 11

Statistical Results of the t-Test of Individual CELDT Scores

Mean N Std.
Deviation
Correlation t df Significance
Repeated Year
CELDT(the pre-
test)

CELDT score one
year later (the
post-test)
2.69

3.03
36

36
1.14

.97
.45 -1.78 35 .083

Although the mean score increased, p=.083, which indicates the increase was not statistically
significant. The ES equals .15, considered a small effect. Retention did not offer significant
benefit for the ELL students as measured by the CELDT.
A second t-test was run using the CELDT scores of the retained students and comparing
the mean of their growth in the year after their repeated year with the mean of their non-retained
current peers in a matched design. The results of this t-test (Table 12) reveal that the mean
performance of the retained students is greater than that of their non-retained peers. However,
p=.517 and ES= .08, indicating that their growth does not meet the level of significance to
suggest that retention was the reason for their performance level exceeding that of their peers.
Table 12

Statistical Results of the t-Test of Retained and Non-Retained Students’ CELDT Scores

Mean N Std.
Deviation
Correlation t df Significance
Retained students’
CELDT scores

Non-retained
students’ CELDT
scores
3.03

2.89
40

40
.974

.51
-.40 .66 39 .517
EFFECTS OF RETENTION 55
Sample Characteristics in Regards to t-Tests using Report Card Reading Grades
A review of the reading grades on the report card gives us the opportunity to find out
whether teachers perceive retained students as working within grade level prior to retention and
post retention. Since there is no control group, we cannot ascribe any improvement to retention.
All we can measure is whether students made progress against the teachers’ expectations post
retention. In other words, does the retained student meet teacher expectations for performance at
that grade level post-retention?
A t-test comparison of the composite reading scores based on the sum of the means of
students’ reading grade, reading effort grade, phonics, vocabulary, and comprehension grades is
found in Table 13. Although the mean post- retention was higher, the ES was .47, only
approaching a medium effect.
Table 13

Results of the t-Test of Composite Report Card Reading Scores: Retention Year and 2011-12

N Minimum Maximum Mean Std.
Deviation
Year End Composite Reading Score
(Retained Year)
136 1.00 3.00 1.53 .44
Year End Composite Reading Score
(2011-12)
136 1.00 4.00 2.02 .72

The Kindergarten and first through third grade report cards provide information about the
reading achievement sub-skills of phonics and comprehension. Achievement is reported at three
levels: Requires Additional Support, Works Within Grade Level Standards, and Demonstrates
Strength. Results of the t-Tests of report card phonics and comprehension grades in the retention
year compared to 2011-12 are reported in Table14. These results indicate that the mean increased
at a significant level, with an ES approaching a medium effect. Since these scores were not
EFFECTS OF RETENTION 56
compared to a control group, all we can note is that the teachers’ perception of the increase of
performance in phonics and comprehension was slight, not necessarily due to retention.
Table 14

Results of the t-Test comparing Phonics and Comprehension Report Card Grades from the
Retention Year and 2011-2012

Paired Samples Mean N Std.
Dev.
Correlation t df Sig. ES
Retained Year
Phonics Grade
2011-12 Phonics
Grade

Retained Year
Comprehension
Grade
2011-12
Comprehension
Grade
1.26

1.78

1.38

1.75
86

86

88

88
.46

.66

.55

.57
.30

.08
-7.12

-4.61
85

87
.000

.000
.42

.32

Sample Characteristics in Regards to Frequencies Using Report Card Reading Grades
Tables 13 and 14 provided t-test information indicating teachers perceived some students
benefitting from retention, as shown by the increased group means. In order to examine the
number of students who were rated higher by teachers post-retention on the phonics and
comprehension report card grades, frequencies were viewed by cross tabulation. Table 15
presents the results of the cross tabulation of report card phonics scores from the retained year
and the 2011-12 school year.

EFFECTS OF RETENTION 57
Table 15

Crosstabulation of Report Card Phonics Scores: Retention Year and 2011-2012

2011-12 Phonics Report Card Grade
Retained Year Phonics Report
Card Grade
Requires
Additional
Support
Works
Within Grade
Level
Standards
Demonstrates
Strengths
Total
Requires Additional Support
Works Within Grade Level
Standards
Demonstrates Strengths
28(93%)
2(7%)

0
30(67%)
15(33%)

0
7 (64%)
3(27%)

1(9%)
65(76%)
20(24%)

1(1%)
Total 30(34%) 45(52%) 11(13%) 86(100%)

Teachers graded 24% (n=20) of the retained students at grade level or above in phonics
during the retention year. Although their perception was that seven per cent (n=2) of the retained
students benefitted from retention in the area of phonics, 51% (n=44) remained at their retention
year level of performance, and 47% (n=40) performed at a lower level post-retention.
A crosstabulation of report card comprehension scores from the retention year and 2011-
12 (Table 16) indicated that teachers believed 82% (n=23) of the 28 students who required
additional support in comprehension prior to retention continued to need the additional support
post-retention. Teachers also scored 34% (n=30) of the students who had performed within grade
level in comprehension prior to retention as requiring additional support post-retention. Of the
7% (n=6) of the students who were graded as demonstrating strength in comprehension prior to
retention, 5.7% (n=5) were seen as requiring additional support post-retention; and only .3%
(n=1) were seen as working within grade level post-retention. Again, it needs to be restated that
these scores had no comparison group, and so the only conclusion that can be drawn from them
is that according to teachers’ subjective scores on the report cards, retained students, by and
EFFECTS OF RETENTION 58
large, did not meet grade level expectations in comprehension post–retention. From the
perspective of the teacher, the majority of the students did not benefit from retention.
Table 16

Results of the Crosstabulation of Report Card Comprehension Grades comparing the Retained
year and 2011-12

2011-12 Comprehension Grade
Retained Year Comprehension
Grade
Requires
Additional
Support
Works
Within
Grade Level
Standards
Demonstrat
es Strengths
Total
Requires Additional Support
Works Within Grade Level
Standards
Demonstrates Strengths
23(82%)
3(11%)

2(7%)
30(56%)
23(42%)

1(2%)
5(83%)
1(17%)

0
58(66%)
27(31%)

3(3%)
Total 28(32%) 54(61%) 6(7%) 88(100%)

Sample Characteristics in Regards to Frequencies Using Report Card Reading Grades
Compared with CST scores
When teachers retain a child, their goal is that students will be successful academically in
future years. A crosstabulation of teacher scores in phonics and comprehension during the
retained year and the CST scores in 2011-12 provide an opportunity to compare whether students
whom the teachers graded as requiring additional support were able to perform within grade
level, post-retention, on a standardized test, the CST language arts test.
A cross tabulation of phonics report card grades (retained year) and CST language arts
scores (2011-12) indicate that 58% (n=41) of the students whom teachers viewed as requiring
additional support in phonics during the retention year performed below proficient or advanced
on the CST, compared to 25% (n=5) who were scored as within grade level, and none who were
scored as demonstrating strength. (See Table 17) By way of contrast, all of the students (n=3)
whom teachers indicated were demonstrating strength in phonics prior to retention performed at
the proficient or advanced level post-retention.
EFFECTS OF RETENTION 59
Table 17
2011-12 CST Language Arts Proficiency Level and Retained Year Phonics Grade
Crosstabulation

Retained Year Phonics Report Card Score
2012 CST
Language Arts
Scores
Requires
Additional
Support
Works Within
Grade Level
Demonstrates
Strengths
Total
Far Below Basic
Below Basic
Basic
Proficient
Advanced
3(4%)
12(17%)
26(37%)
16(23%)
13(19%)
0
1(5%)
4(21%)
6(31%)
10(43%)
0
0
0
2(67%)
1(33%)
3 (3%)
13 (14%)
30(32%)
24(26%)
24(26%)
Total 70(74%) 21(23%) 3(3%) 94(100%)

A crosstabulation comparison of retained year comprehension grades and 2011-12 CST
language arts scores (Table18) shows that 57% (n=31) of the students who were rated below
grade level in comprehension their retained year were not able to score proficient or advanced on
CST post-retention, compared to 35% (n=9) who were rated at grade level, or 33% (n=2) who
were rated demonstrates strengths in their retained year.
Table 18

2011-12 CST Language Arts Proficiency Level and Retained Year Comprehension Grade
Crosstabulation

Retained Year Comprehension Report Card Score
2012 CST Language
Arts Scores
Requires
Additional Support
Works Within
Grade Level
Demonstrates
Strengths
Total
Far Below Basic
Below Basic
Basic
Proficient
Advanced
2(4%)
10(18%)
19(35%)
15(27%)
9(16%)
0
2(8%)
7(27%)
6(23%)
11(42%)
1(17%)
0
1(17%)
3(50%)
1(17%)
3(3%)
12(14%)
27(32%)
24(28%)
21(24%)
Total 55(63%) 26(30%) 6(7%) 87(100%)

EFFECTS OF RETENTION 60
Sample Characteristics in Regards to the Report Card Work/Study and Citizenship
Grades
Kindergarten through fifth grade students received grades in a variety of areas related to
their habits as a student and their citizenship. A composite Work/Study Habits score was created
from the sum of the means of the sub-skills of Listening, Stays on Task, Completes Work
Accurately, Completes Classwork on Time, Completes Homework on Time, Completes Work
Neatly, and Participates in Classroom Activities. Table 19 presents a comparison of the
composite scores, showing an increase in the students’ means post-retention. The ES was .30, a
small effect size.
Table 19

A Comparison of Composite Work/Study Habits Scores: Retention Year and 2011-12

N Minimum Maximum Mean Std.
Deviation
Work/Study
Habits
(Retained
Year)
140 1.00 3.88 2.24 .57
Work/Study
Habits
(2011-12)
154 1.29 4.00 2.64 .79

A t-test comparing each sub category of the Work/Study Habits of retained students
compared with 2011-12 report card grades (Table20) demonstrates that the teachers’ subjective
measures of student growth resulted in an increased mean at a significant level in each sub-
category. However, the ES for each sub-category is small.

EFFECTS OF RETENTION 61
Table 20
Results of the Work/Study Habits t-Test Comparing Retained Year and 2011-12

Paired Samples Mean n Std.
Dev.
Correlation t df Sig. ES
Retained Year Listening
2011-12 Listening

Retained Year Stays on
Task
2011-12 Stays on Task

Retained Year Completes
Work Accurately
2011-12 Completes Work
Accurately

Retained Year Completes
Classwork on Time
2011-12 Completes
Classwork on Time

Retained Year Completes
Homework on Time
2011-12 Completes
Homework on Time

Retained Year Completes
Work Neatly
2011-12 Completes Work
Neatly

Retained Year Participates
in Classroom Activities
2011-12 Participates in
Classroom Activities
2.19
2.56

2.01

2.63

1.95

2.43

2.01

2.75

2.64

2.97

2.27

2.70

2.30

2.70

138
138

135

135

131

131

136

136

139

139

132

132

135

135
.82
.99

.85

1.06

.76

.92.

.76

1.04

.97

1.08

.79

.93

.74

.86
.44

.35

.25

.28

.34

.38

.20
-4.45

-6.54

-12.04

-7.77

-3.30

-5.13

-4.62
137

134

130

135

138

131

134
.000

.000

.000

.000

.000

.000

.000
.2

.31

.27

.37

.16

.24

.24

Teachers also subjectively rated students’ performance in citizenship categories. A
composite score is based on the sum of the means of the sub-scores for Exercises Self Control,
Respects Authority, Observes Others’ Rights and Property, Observes Rules, Cooperates With
Others, and Accepts Responsibility. A comparison of the Report Card Citizenship grades is
EFFECTS OF RETENTION 62
reported in Table 21, indicating that the mean of the retained students’ citizenship scores
increased post- retention; however, the ES was small, .2.
Table 21
A Comparison of Composite Citizenship Scores: Retention Year and 2011-12

N Minimum Maximum Mean Std.
Deviation
Composite
Citizenship
(Retained
Year)
139 1.00 4.00 2.6508 .80655
Composite
Citizenship
(2011-12)
154 1.00 4.00 2.9708 .89360

A t-test comparing each sub-category of the Citizenship scores of retained students
compared with 2011-12 report card grades (Table 22) demonstrates that the teachers’ subjective
measures of student growth did not result in an increased mean at a significant level in any sub
category. Additionally, the ES for each subcategory is small, suggesting that teachers did not
perceive any benefits in Citizenship due to retention.
Table 22
Results of the Citizenship t-Test Comparing the Retained Year and 2011-12 Report Card Scores

Paired Samples Mean n Std. Dev. Correlation t df Sig. ES
Retained Year
Exercises Self Control
2011-12 Exercises Self
Control

Retained Year Respects
Authority
2011-12 Respects
Authority
2.29

2.63

2.87

3.23
139

134
.92

1.05

.88

.91
.44

.38
-3.89

-4.15
138

133
.000

.000
.17

.2

EFFECTS OF RETENTION 63

Table 22
Results of the Citizenship t-Test Comparing the Retained Year and 2011-12 Report Card Scores
(continued)

Paired Samples Mean n Std. Dev. Correlation t df Sig. ES
Retained Year
Observes Others’
Rights and Property
2011-12 Observes
Others’ Rights and
Property

Retained Year
Observes Rules
2011-12 Observes
Rules

Retained Year
Cooperates With Others
2011-12 Cooperates
With Others

Retained Year Accepts
Responsibility
2011-12 Accepts
Responsibility
2.76

3.02

2.58

2.97

2.76

3.02

2.70

3.08
135

135

135

135

137

137

130

130
.88

.97

.96

.98

.89

.98

.86

.96
.30

.45

.37

.38
-2.83

-.4.50

-2.92

-4.13
134

134

136

129
.005

.000

.004

.000
.04

.2

.14

.02

Research Question 2: What is the correlation between selected demographics and
retention?
Study 2
The demographic and family information relating to the retained students who were
enrolled K-6 in RUSD during the 2011-12 school year was gleaned from the enrollment form
parents completed, and program participation information was obtained from RUSD databases.
Sample Characteristics Related to Education Setting.
During the 2011-2012 school year, RUSD had a K-6 enrollment of 10,200 students, 184
of whom had been retained at some point (K-5) during their school career. Due to student
EFFECTS OF RETENTION 64
transfers out of the district, the final retention group that was studied consisted of 179 students.
Within the study group, 85.5% (n=153) of the students had been retained at an RUSD school,
14.5% (n=26) had been retained at another district prior to enrollment in RUSD, and 26.4%
(n=47) of the students had some of their K-6 schooling in another district. Eighty-six percent
(n=154) of the students had been retained in Kindergarten or first grade, and one student had
been retained as late as 5
th
grade. Table 23 shows the grade level in which the students were
originally retained compared to their 2011-12 grade level.
Table 23

Summary of Retained Student Enrollment by Grade Level

Grade Grade Level student was retained Student’s Grade Level in 2011-12
Kindergarten 106(59.2%) 23 (12.8%)
First 48 (26.8%) 30 (16.8%)
Second 13 (7.3%) 17 (9.5%)
Third 6 (3.4%) 23 (12.8%)
Fourth 5 (2.8%) 23 (12.8%)
Fifth 1(.6%) 39 (21.8%)
Sixth 0 24 13.4%)
Total 179 (100%) 179 (100%)

Sample Characteristics in Regards to Gender, Ethnicity, Age and Program Participation
The pattern with which students are most likely to be retained nationally was echoed in
the RUSD: a greater proportion of males (62%, n=111), ethnic minorities (51.4%, n=92), and the
youngest students in the class (55.3%, n=99) were retained. Disproportionally retained based on
total district enrollment were Limited English Proficient students (LEP) 25.7%, (n=46) and
Reclassified English Proficient students (RFEP) 4.5%, (n=8); Economically Disadvantaged
Students (35.8%, n=64); Students with Disabilities (33.5%, n=60); Students enrolled in a Special
Day Class (11.7%, n=21); and students attending a Title 1 school (20.1%, n=36) (See Table 24).
EFFECTS OF RETENTION 65
When reviewing data on Parental Education Level, it is clear that students whose parents
were not high school graduates are more than three times as likely to be retained. Students whose
parents have a college or graduate degree are the least likely to be retained: only 36% of the
retained students had parents with college or advanced degrees compared with 72.28% of the
non-retained students’ parents.
Table 24
Comparison of Study Group and District Demographics

Study 1
Participants
RUSD Total Enrollment
(Minus the Study Group)
Gender:
Males
Females
Total

111 (62%)
68 (38%)
179 (100%)

5108 (52%)
4811(48%)
9919 (100%)
Ethnicity:
White, non-Hispanic
Hispanic
African American
Filipino
Japanese
All Others

87 (48.6%)
76 (42.5%)
2 (1.1%)
3 (1.7%)
3 (1.7%)
8 (4.4%)

5132 (51.7)
2828(28.5%)
482 (4.87%)
567 (5.7%)
95 (.96%)
816 (8.2%)
Language Proficiency:
Limited English Proficient
Reclassified English Proficient
Initial English Proficient
English Only

46 (25.7%)
8 (4.5%)
4 (2.2%)
121 (67.6%)

1589 (15.7%)
289(2.84%)
249(2.44%)
8517(83.59%)
Economically Disadvantaged 64 (35.8%) 1908(19%)
Students With Disability 60 (33.5%)
Student attends a Special Day Class 21 (11.7%)
Student attends a Title I school 36 (20.1%) 1248(12.58%)
Identified as GATE (Gifted and Talented
Education)
1 (.6%) 442 (4.4%)
Student has a 504 Plan 2 (1.1%) 182(1.79%)
Student has a Health Plan 2 (1.1%) 320(3.14%)
Parent Educational Level
Not High School Graduate
High School Graduate
Some College (Includes AA)
College Graduate
Graduate School/Post Graduate Training

16 (11%)
34 (23.2%)
44 (30.4%)
39 (27.5%)
13 (8.5%)

327(3.3%)
2051(20.7%)
2678(26.28%)
4703(46.16%)
2661(26.12%)
EFFECTS OF RETENTION 66

Sample Characteristics in Regards to Family
At the time of enrollment, most students lived with both parents (77.5%, n=138), either
as an intact family or in a shared parenting arrangement, while 15.2% (n=27) lived with their
mother only and 1.1% (n=2) lived with their father only. A small group of students (3.4%, n=6)
did not live with either parent, and 2.8% (n=5) lived with a parent and another adult.
Most students lived in homes where the parents were employed: 43.9% (n=75) had both
parents employed; 33.5% (n=60) had only their father employed; 14% (n=25) had only their
mother employed, and 6.4% (n=11) came from a home where neither parent was employed.
Table 25 presents a crosstabulation of the student’s birth order and the number of
children in the family. It is obvious that in every multiple-child family, except for the five-child
family, the retained children in this study group are most often the youngest child in the family,
54.8% (n=86).
Table 25

Crosstabulation of the Student’s Birth Order in the Family and Number of Children in the
Family

Only
Child
Two
Children
Three
Children
Four
Children
Five
Children
Seven
Children
Total
Firstborn 18 29 10 3 60(34.68%)
Second
Child
41 13 3 57(32.95%)
Third
Child
31 6 1 38(21.9%)
Fourth
Child
12 4 16(9.2%)
Fifth
Child
1 1(.6%)
Seventh
Child
1 1(.6%)
Total 18(10%) 70(40.87%) 54(31.21%) 24(13.87%) 6(3.47%) 1(.6%) 173

EFFECTS OF RETENTION 67
The scope of this research did not specifically include gathering data to see if other
siblings had also been retained at some point. However, within this retention group under study
(which spans seven years of data) there were six families who were represented by more than
one retained child in their family. There were three sets of twins: two sets being the only children
in the household, and one set of twins with one additional sibling for a total of three students in
their home. Two other families had two of their four children in the retention group (birth orders
one and four out of four, and one and two out of four); one family had three out of five of their
children in the retention group (birth orders three, four, and five).
Sample Characteristics in Regards to Special Education
An analysis of IEP qualification of RUSD retained students shows that 39.7% (n=71)
were tested for a learning disability: 33.5% (n=60) qualified for an IEP and 6.1% (n=11) did not.
The majority of retained students with an IEP qualified during their pre-Kindergarten (17.3%,
n=31) or Kindergarten years (6.1%, n=11). Table 26 summarizes the grade levels at which each
student was tested. An interesting observation is that in the group of retained students with an
IEP, the majority (66.7%; n=40) qualified for it prior to the year the retention decision was made.
A small group (6.1%; n=11) qualified up to five years post-retention.
Table 26

Grade Levels At Which Students Were Tested for a Disability

Retention
Grade
IEP Testing Grade Number Tested
That Qualified
for an IEP
Number Tested
That Did Not
Qualify for an IEP
K Pre K 22 3
K K 3
K K (Repeated year) 2
K 1 1
K 2 2 1
K 3 5

EFFECTS OF RETENTION 68
Table 26

Grade Levels At Which Students Were Tested for a Disability (continued)

Retention
Grade
IEP Testing Grade Number Tested
That Qualified
for an IEP
Number Tested
That Did Not
Qualify for an IEP
K 4 1
1 Pre K 6
1 K 6 1
1 1 (Repeated year) 2 1
1 2 1 1
1 3 1
2 Pre K 3
2 2 (Repeated year) 1
2 4 1
3 2 2
3 3 (Repeated year) 1
3 4 1
4 3 1 2

Only 8.3% (n=5) students who had IEPs were exited from their IEP after being retained, and all
five had disabilities in the area of Speech Language Impairment (SLI).
Sample Characteristics in Regards to Attendance
Absence rates for retainees were highest during the retention year. Although lower in
2011-12, it still exceeds the absence rate of non-retained students (see Table 27).
Table 27
Comparison of Attendance rates Pre-Retention and Post-Retention

Minimum Maximum Mean
Retained Students’ Retention
Year Absences
0 35 8.84
2011-12 Retained Student
Absences
0 33 6.45
Non-Retained Students 2011-12
Absences
0 Unknown 5.6

EFFECTS OF RETENTION 69
Sample Characteristics in Regards to Suspensions
Eleven percent (n=20) of the retained group were suspended during the period under
study. Suspensions for this sample reached a peak at three years post-retention: 9 of the 10
children suspended had been retained in kindergarten; the tenth was retained in first grade. The
most frequent reason for suspension was “General Disruptive Behavior”. Table 28 presents the
number of suspensions compared to the retention year.
Table 28
Numbers of Suspensions Compared to Retention Year

Retention Year 2 Students (1 each)
1 Student (2 Suspensions)
1 Student (3 Suspensions)
Total Students: 4
Total Suspensions: 7
Repeated Year 3 Students (1 each)
1 Student (4 Suspensions)
Total Students: 4
Total Suspensions: 7
2 Years Post Retention 4 Students (1 each)
Total Students: 4
Total Suspensions: 4
3 Years Post Retention 5 Students (1 each)
1 Student (2 Suspensions)
1 Student (3 Suspensions)
Total Students: 7
Total Suspensions: 10
4 Years Post Retention 4 Students (1 Suspension)
1 Student (2 Suspensions)
Total Students: 5
Total Suspensions: 6
5 Years Post Retention 3 Students (1 each)
1 Student (2 Suspensions)
Total Students: 4
Total Suspensions: 5
6 Years Post Retention 1 Student (3 Suspensions)
Total Students: 1
Total Suspensions: 3

EFFECTS OF RETENTION 70
A comparison of the number of suspensions for retained versus non-retained students in the
2011-12 school year (Table 29) indicates that retained students were much more likely to be
suspended than their non-retained peers.
Table 29

Comparison of Suspension Rates During the 2011-2012

Average Number of Suspensions for
Retained Students
17(9.5%)
Average Number of Suspensions for
Non-Retained Students
127(1.25%)

In addition to information related to academic success or lack of success related to
retention, California State law and RUSD policy require other procedures to be in place. Among
those procedures are participation in school-based academic intervention programs, parent
notification of at-risk status, and parent participation in the SST process. The information that
follows was obtained during the review of student cumulative files.
Findings in Regards to Academic Risk Notification and Intervention
Although RUSD retention policy prescribes that parents be notified of their child’s risk of
retention at the earliest possible date, the cumulative file review indicated that only 38% (n=68)
had received formal notification. Policy also requires that students who are retained participate in
academic intervention programs provided by the school before and after retention, processes that
are to be noted in the cumulative file. Results of intervention findings are noted in Table 30.

EFFECTS OF RETENTION 71
Table 30

Intervention Participation as Noted in the Cumulative File
No Yes
Intervention offered in the year prior to the retention year
(K students excluded)
46
(63.1%)
27
(36.9%)
Intervention offered in the retention year 90
(50.3%)
89
(49.7%)
Intervention offered post-retention in the repeated year 107
(69%)
48
(31%)

There was insufficient documentation in the cumulative files regarding whether intervention took
place, and if it did take place, what type, what duration, and what the effectiveness of the
intervention was. This lack of detail prevented us from running any statistical analysis of the
effectiveness of the intervention clause of the retention policy.
Sample Characteristics in Regards to the Student Study Team (SST) Process
When students have difficulty in school due to academics or for social reasons, the
RUSD process includes holding an SST meeting to discuss the concerns, brainstorm ideas, and
create an academic or behavioral improvement plan. Parents are invited to be a part of the
meeting. Copies of the plan are given to the parent and teacher, and a copy is placed in the
cumulative file. Table 31 outlines the results of the cumulative file review for SST information.
Table 31
Parent Participation in the SST Process as Noted in the Cumulative File

No One SST Two SSTs Three SSTs
Parent participated in an SST in
the retention year
102
(57%)

70
(39.1%)

6
(3.4%)

1
(.6%)

Parent participated in an SST in
the year prior to retention
(Kindergarten retainees not
included)
52
(71.2%)

16
(21.9%)

4
(5.5%)

1
(1.6%)

EFFECTS OF RETENTION 72
During the cumulative file review, many instances were found where the SST that was
held prior to retention served only to formalize the decision to retain. In those cases there was no
brainstorming or planning of alternate interventions, nor was there time given to see if any
interventions could provide an alternative to retention. As the purpose of the SST requirement in
the policy is to give alternative practices an opportunity to increase student performance, any
correlation tests of SST participation with student achievement would have rendered false
information. All that can be said for certain is that the purpose of the SST in the RUSD retention
policy has not been followed with fidelity.
EFFECTS OF RETENTION 73
Chapter Five: Discussion
Two separate studies were conducted to measure the academic effects of retention on
students. One was a study of a population of K-2 students from the 1998-99 and 1999-2000
school years to conduct a Retrospective Control Group comparison of retained and non-retained
students’ performance. The goal of this study was to see if retained students would make enough
academic growth due to retention that they would be able to catch up to their peers, or even
surpass them as a result of being retained. Based on the literature review, the hypothesis was that
retention would not be academically beneficial, i.e., the performance of retained students would
not meet or exceed their non-retained peers’ academic performance.
The second study focused on a group of retained students from the 2011-2012 school
year in order to provide a more current analysis of retained students’ performance relative to
grade level expectations and to review selected demographics correlated to retention. The goal of
this study was to see if retained students would benefit from retention so that they could now
perform within the teachers’ subjective grade level expectancies in the classroom day to day, and
whether they could also perform within the grade level competencies expected by the State on
the California Standards Test (CST), and show annual growth on the California English
Language Development Test (CELDT). Together both studies shed light on different aspects of
the effects of retention.
Academic Achievement as Measured by a Retrospective Control Group
Study 1
Due to the immense consequences of retaining students for the purposes of creating a
matched control group to study the effects of retention, this study instead used a quasi-
experimental Retrospective Control Group model. Changes in California law ending social
promotion in 1999 provided the opportunity to compare students who were retained in 1999-
EFFECTS OF RETENTION 74
2000 under the new law with students who were not retained in1998-99, but would have been
retained had the law been in effect in 1998-99. The experimental group was the group of students
retained in 1999-2000, and the retrospective control group (a comparison group) was the 1998-
1999 cohort.
Whether viewing Language Arts performance through the perspective of teacher assigned
grades on the Pupil Record Cards (PRC) at the end of sixth grade, or through the perspective of
Language Arts performance on the sixth grade CST, the results were the same: after retention,
the retained students did not perform academically at the same level as their non-retained peers.
On the measures of math achievement, the retained group scored lower on sixth grade teachers’
classroom Math scores, and slightly higher, but not to a level of significance, on the CST sixth
grade Math test.
Although it is interesting to note that teachers rated the retained students’ classroom
effort in Math and Language Arts slightly higher than non-retained students, the results were not
statistically significant. In other words, retaining the students did not lift their academic
performance or effort to the level of their non-retained peers. Retention did not benefit these
students academically as compared to their non-retained peers.
Methodological Considerations
Internal validity.
Hong and Raudenbush(2005), Lorence (2006), and Wu, West, and Hughes (2008) have
all pointed out the inherent difficulties in matching students in non-random tests based on
equivalent scores, since it does not necessarily account for the unknown ways in which peers
differ from each other. Some of the variables that they suggest may be hidden during the
matching process are parent educational levels and social competencies. They suggest this
presents a hidden bias in the matching process when creating a control group, which is a threat to
EFFECTS OF RETENTION 75
internal validity. One weakness of the current study is that no attempt was made to match
students on any variables in the propensity scores other than academic classroom performance
and effort in Language Arts and Math. However, given that the retrospective control group was
comprised of all of the students who would have been retained had they entered school a year
later, it is probable that the study’s design partly compensates for those hidden variables that
cannot be measured in a non-random study.
The outcome of prior studies (Holmes (1989), Hong and Raudenbusch (2005), Lorence
(2006),) suggested that there was no benefit to analyzing students’ progress by comparing “same
age” or “same year” comparisons. All retrospective control group tests in this study were run
using a “same grade” comparison, thus it is unknown whether comparing retained students with
their promoted peers would have resulted in other outcomes.
Another potential weakness of this retrospective study is that there was no comparison of
repeated year versus retained year progress. Since there are multiple life events that occur in the
span of five to seven years as well as individual maturity rates that vary over time, it is unknown
what weight those factors carried in the final sixth grade outcomes. It could be argued, however,
that those life events occur for both the retained group and the non-retained group and that both
groups were selected using the same initial academic factors. These tests do not show whether
one group is subjected to more naturally occurring life events than the other group, and if so,
what affect those events had on the academic outcomes, since those factors were outside of the
scope of this research. However, one experience unique to the retained group could be the social-
emotional effects of retention on academic performance, which was not measured in this study
and could have affected the academic outcomes. If retention does increase social-emotional
EFFECTS OF RETENTION 76
burdens, and that extra burden decreases academic performance, that would constitute another
argument against retention.
External validity.
This study was limited to students in Kindergarten through 2
nd
grade in a single Southern
California suburban school district from 1998 through 2005.One of the measurements used was a
local protocol for rating students— Above, At, or Below grade level—in Language Arts and
Math, as well as effort in both of those curricular areas. The other measure was the CST scores
for Language Arts and Math. Schools outside of this district would not have the same local
protocol, and schools outside of California would not have student results tied to CST. Both of
these factors would make it difficult to replicate this study.
Statistical validity.
Although statistical validity may have been affected by sample size, power was increased
due to the matched design using propensity scores. Given that four of the outcomes were based
on teacher scores, the subjectivity of the scores could call into question the reliability of the
results. However, the outcomes of the teacher scores and the standardized tests were similar,
both showing no evidence that the academic gains by retained students showed a statistical
advantage for retention.
Academic Achievement Measured by Teachers’ Classroom Expectations and State Testing
Study 2
The study of 179 Kindergarten through sixth grade students who had previously been
retained and were enrolled in RUSD during the 2011-2012 school year provided the opportunity
to see whether retention benefitted students as measured by teacher grades and state testing.

EFFECTS OF RETENTION 77
Report Card Reading Grades
An analysis of the reading report card grades by t-tests comparing Composite Reading
scores, Phonics scores, and Comprehension scores for retained students showed increased means
at a level of significance on all three tests. However, the Effect Sizes (ES) were small to medium.
An additional concern in interpreting these scores is that 31% of the retained group was still
below grade level on the composite reading score post-retention. Even if internal and external
validity questions were not an issue, this would not be a ringing endorsement for the
effectiveness of retention.
Students’ Reading Progress as Measured by Grade Level Expectations
A more helpful analysis of the teachers’ report card grades to see whether retention was
beneficial for students was to see whether the teacher rated the students as performing at grade
level post-retention. In both of the reading sub-skills of phonics and comprehension, a majority
of students who required Additional Support prior to retention, continued to be rated as requiring
Additional Support post-retention (phonics, 93%; comprehension, 82%). The majority of
students who had been rated as within grade level at the retention year, fell back to require
Additional Support post-retention (phonics, 67%; comprehension, 56%). Even the students who
demonstrated strengths in these areas at retention no longer demonstrated strength post-retention
(phonics, 91% rated as requiring Additional Support or Works Within Grade Level;
comprehension, 100% rated as requiring Additional Support or Working Within Grade Level
post-retention).
Teachers’ report card scores are subjective in nature and vary from one teacher to
another, one school to another. This creates both external and internal validity concerns in using
the reading report card scores. However, what is valuable from reviewing these scores is that
these measures are viewed by parents, school personnel, and the general public as a significant
EFFECTS OF RETENTION 78
way to determine whether a student is meeting grade level expectations: Many a reward has been
given a school child based on report card scores that demonstrate he is working at least at grade
level. Parents and teachers also use report card scores to validate that a retained child has “caught
up” to his peers or is still lagging. For this reason, it is troubling to see that even when using
report card grades, there is overwhelming data that retention does not result in retained students
meeting teachers’ expectations of grade level performance.
Methodological Concerns
Internal validity.
Conclusions based on these report card scores showing that retention made children more
academically successful are problematic. Teachers’ report card grades are subjective and call
into question the internal validity of the assumption that retention resulted in increased
performance. Since there is no concrete measure of what constitutes “Requires Additional
Support” from one year to the next, it could be a comparison of apples to oranges. More
experienced teachers might rate a student’s classroom performance quite differently than a less
experienced teacher. Teachers at a grade level at one school might weigh factors that are
included in their grade differently than teachers at the same grade level at another school within
the district.
The other argument against concluding that teacher scores indicate retention had a
positive impact on academic performance is that there are other factors such as additional
academic intervention at school or at home that may have affected a child’s academic
performance, and those factors, not retention, may in fact be responsible for the improvement in
the scores. This concern about internal validity urges caution in interpreting the results as an
endorsement of retention for academic growth purposes.
EFFECTS OF RETENTION 79
External validity.
The scores used for the report card reading grade were based on one particular district’s
practice where each teacher decides which achievement tests, classroom assignments, and
subjective effort ratings should be included in the report card grade. There may or may not have
been agreement across a grade level as to which scores should be included in the report card
grade. Therefore, these results are not generalizable beyond this district.
State Testing
An external and non-subjective measure of whether students perform at grade level post-
retention was provided by disaggregating their CST scores and CELDT scores. For CST, the
measure of grade level competence is Proficient or Advanced. Only 52% of the retained students
were able to meet this standard post-retention as compared to 78% of their non-retained peers. If
retention were an effective means of remediating poor academic performance, it would be
expected that more retained students would have performed at the Proficient or Advanced level.
The matched design t-test CST Language Arts scores of retained students at the end of
their repeated year and their non-retained peers, demonstrated that the mean of the retained
group did not meet the mean of the non-retained group. Every test that compared retained and
non-retained students using CST scores showed statistically that retention did not benefit the
retained students compared to his or her non-retained peers.
School personnel and the general public often assume and assert that ELL students
should be retained in order to increase their language acquisition. However, a review of their
year- over-year CELDT test scores showed a small ES of .30, and an insignificant increase in the
mean. Retained students made about the same English language growth in two years that their
non-retained peers made in one year. In other words, retention does not significantly affect
positive language growth for ELL students.
EFFECTS OF RETENTION 80
Methodological Concerns
Internal validity.
Due to the ethical considerations of creating a randomized control group for this study,
student scores were analyzed using a matched control group, and using the individual students’
pre-post scores. The CELDT is highly reliable, and the statistical insignificance and low ES of
the tests that used CELDT scores lends weight to the results.
External validity.
CELDT is a statewide California test, and the results from this population of ELL
students should be highly generalizable to other ELL students in California.
Statistical conclusion validity.
Although statistical validity may have been affected by sample size, power was increased
by use of a matched design.
Demographics and Retention
Age and Maturity
RUSD mirrors the national practice of disproportionately retaining males and the
youngest students in the class. During the review of the students’ cumulative files, it was not
uncommon to see the reason for retention stated as the child being the youngest in the class,
needing more time to mature, or specifically stating that he was a boy who needed more time to
mature, even when the child was marked as “Works Within Grade Level Standards” or
“Demonstrates Strengths” for both academic areas of Phonics and Comprehension.
Although the scope of this study did not include the social or emotional causes or effects
of retention, it is interesting to note that if immaturity was the reason for retention; giving
students an extra year of time to mature did not benefit them academically as measured by
EFFECTS OF RETENTION 81
standardized test scores. In other words, the gift of time may have increased maturity, but that
increase in maturity did not lead to greater academic gains.
Work/Study Habits and Citizenship scores measured some behaviors that are associated
with maturity. Both composite scores for these areas showed a slight increase in the mean post
retention, but the ES were too small to attribute the change to retention. Similarly, when t-Tests
were performed on the seven Work/Study Habits sub-groups of Listening, Stays on Task,
Completes Work Accurately, Completes Classwork on Time, Completes Homework on Time,
Completes Work Neatly, and Participates in Classroom Activities, small increases in the mean
were significant but associated with ES too small to indicate that retention was responsible for
the increase.
Individual t-Tests performed on the subgroup categories of Citizenship, i.e., Exercises
Self Control, Respects Authority, Observes Others’ Rights and Property, Observes Rules, and
Accepts Responsibility also show statistically validated increased means, but again, the ES is so
small in each case that retention cannot be credited with the increase.
Methodological Concerns
Internal validity.
No attempt was made to validate the construct of maturity consisting of Work/Study
Habits and Citizenship grades. Although some of the sub categories may be understood by
teachers and the general public as indicators of maturation, no tests validating the construct were
run. The value of including them in this study is that parents, teachers, and the general public
view report card scores as a measure of student growth and grade level performance. As noted
earlier, using subjective teacher scores presents both internal and external validity problems.
Differing teacher expectations may mean that the scores assigned in one class are on a different
scale than those of her neighboring teacher.
EFFECTS OF RETENTION 82
The role that age plays in maturity is well known. Ferreting out the maturity due to age
and maturity due to retention has not been done. Additionally, no comparison between the
changes in scores from the retention year to the current year against non-retained peers has been
calculated. So although some may view the increase in means as an endorsement of the benefit
of retention as related to maturity, those scores must be viewed with caution. All that can be said
with certainty is that the teachers generally viewed the students as having more maturity-related
skills post retention than they did prior to retention. It does not constitute a causal relationship
between retention and the increase in those maturity-related sub-skills.
External validity.
Because the report scores for Work/Study Habits and Citizenship are local measures used
by one district, these results are not generalizable.
There is one final issue related to the student’s age and the disproportionality of retention
for students born in September through December. During the years that this study was
conducted, the cut-off date for Kindergarten enrollment was December 2. The three months of
additional growth and maturity afforded students born before September 1 has often been given
as the reason for retaining the student who was born in September, October, or November.
During the 2012-13 school year, California instituted Transitional Kindergarten, moving
the enrollment date for kindergarten up by one month each year until 2014, when the new
Kindergarten enrollment cut-off date will become September 1. Students whose fifth birthday
falls between September 1 and December 2 will not be required to have two years of
Kindergarten prior to first grade. The first year, Transitional Kindergarten, is intended to be more
developmental than Kindergarten but more academic than Preschool. It will be interesting to see
EFFECTS OF RETENTION 83
if the current disproportionality of Fall births in the retention group moves to become the
summer birthdates in future years, and if retention numbers drop.
Policy Considerations
The RUSD Promotion/Retention Guidelines outlines the district policy and intended
practices based on Education Code 48070. Each of the requirements of the law is addressed with
corresponding district practice expectations. The law requires that each district create academic
levels a child must meet in order to promote at the second, third, fifth, and sixth grade. The
district has done so, and has also created guidelines for retention at Kindergarten, first, and
fourth grades. However, there are several areas this research has identified where practice has not
matched district policy and procedures.
Parental Involvement
Per district policy and procedures, schools are to notify parents at the earliest opportunity
of their child being at risk; to provide the SST process for parents and schools to work together
to create a plan to provide additional assistance; and to notify parents of retention by May 1
st
.
Cumulative file review found that parents were formally notified in only 38% (n=68) of the
cases. Only 43% (n=77) of the parents participated in an SST during the retention year, and
many of them were called only for the purpose of formalizing the retention in late May or June.
This is problematic for several reasons, one being that parents are responsible for their
children and they should have the right to be informed of the child’s lack of progress and to work
with the district to create a plan for meeting their child’s needs. The profile of retained students
compared to non-retained students demonstrates that retained students’ families are
disadvantaged on several fronts: Their parents are more likely not to have graduated high school
or are only high school graduates (33%, n=49); they are twice as likely to be economically
disadvantaged (36%, n=60); almost twice as likely to speak a different language (30%, n=54);
EFFECTS OF RETENTION 84
and more likely to belong to a minority ethnic group (51%, n=92). It behooves educators to make
a concerted effort to include them in the decision-making process because these families may
have a limited understanding of what is at stake, and they may be intimidated by the power
differential that exists due to the cluster of variables noted above.
Academic Performance Guidelines for Retention
District Guidelines require that K-2 students should be graded as “Requires Additional
Support” in two areas of reading on their progress report in addition to being listed as “At Risk
for Retention” and “Performing at Below Basic” on district multiple measures. No records were
found that indicated multiple measures had been taken into account, and only 26% (n=23) of the
students were graded as “Requires Additional Support” in two areas on the reading progress
report. Only 43% (n=24) of the students who were graded as “Requiring Additional Support” in
reading comprehension during the retention year were able to score Proficient or Advanced on
the Language Arts CST, so the current practice of using the K and 1
st
grade academic guidelines
for retention is not working, if the goal of retention is remediation.
The fact that so few of the retained students fell within the district’s retention guidelines
may be an indication that the majority of the students are actually retained for reasons other than
academic ability. One weakness in this study is that no data were gathered for analysis regarding
retention reasons other than academic performance.
Academic Intervention
Academic intervention pre-and post–retention is a requirement of the education code and
district policy and procedures. However, only 50% (n=90) of the students received academic
intervention during the retention year, and 69% (n=107) were offered intervention post-retention.
It was not clear how many students took advantage of the offer of academic intervention post-
retention, since the offer was in the form of summer school, which parents were not obligated to
EFFECTS OF RETENTION 85
accept. Participation in academic intervention opportunities during the school year post-retention
was not well documented. However, it is possible that the dearth of intervention information
found in the cumulative files was due to the fact that the majority of the students who were
retained performed at an academic level that was higher than what would trigger the need for
intervention during their retention year. Because intervention services are based on academic
performance, and the majority of the retained students were performing within or above grade
level standards, it stands to reason that academic intervention was not offered: they just didn’t
need it.
One weakness of this study is that no data were collected to shed light on the type,
duration, or quality of academic intervention that was offered. There was also no attempt made
to track whether the academic intervention that was offered prevented more students from being
retained. In other words, were the retained students who did receive academic intervention a
subset of a larger group who received intervention successfully and thus were not retained? It is
also not known how many students availed themselves of the offer of academic intervention
post-retention, so it is unclear for how many students retention was the de facto intervention
(Peterson & Hughes, 2011).
Special Education
District policy does not provide clear guidelines as to whether students with an IEP
should be retained. It states, “It is possible that special education students can be retained, but
only under certain circumstances” (Appendix C; p. 14) Those circumstances are not delineated,
other than that students who are not making 80% of their IEP goals are at risk. One prism for
discussing the results of this study has been presented by Beebe-Frankenberger et. al (2004), who
raised the specter of retention being used as a holding pattern for special education testing. Their
EFFECTS OF RETENTION 86
research showed evidence of struggling students being retained, and when they still did not
improve, being tested and qualified to receive Special Education services.
A review of the data in this current study indicates that 72% (n=43) of the retained
students who had an IEP qualified for it prior to retention. An additional 3% (n=6) of the entire
retention group were tested prior to retention but did not qualify. For all of these students,
retention was not a holding pattern for receiving special education services.
Of those who were tested for a learning disability post-retention, 10% (n=6) qualified for
an IEP during their repeated year, an indication that their struggles continued to be monitored
immediately post retention, and the remaining 18% (n=11) qualified up to five years post-
retention. These last-identified students may have been in the holding pattern for testing that
Beebe-Frankenberger (2004) describes, or it may be that the academic gap between them and
their peers continued to grow until the gap between their performance and grade level norms and
the quality of their classwork became factors that were used to qualify them for special education
services.
Additionally, 2.7% (n=5) of the entire retention population were tested but did not qualify
up to two years post-retention. For these students and the rest of the group that were tested for an
IEP post-retention, it is a clear indication that school personnel and parents were not seeing
academic benefits to the retention.
The numbers related to IEP qualification represent an interesting dilemma. Although it is
good that 72% of those with an IEP were qualified prior to retention, it raises the question of
why retention was considered and implemented, given that an IEP is supposed to be an
individualized plan to implement all of the supports necessary for ensuring academic growth for
EFFECTS OF RETENTION 87
the child, and district procedures are silent on retention for Special Education students. For these
students, retention became an intervention in the suite of interventions in their IEP.
Further Study
Given that the majority of retained students did not meet the district’s academic criteria
for retention, the district should study the real reasons behind the large number of K and first
grade retentions. Intervention targeting those factors should be implemented and studied for their
effectiveness in decreasing the need for retention. As Transitional Kindergarten is fully
implemented, the performance of students enrolled in that program should be analyzed to see if
the disadvantage of age is ameliorated.
Birth Order
The findings related to the preponderance of youngest children in the family being
retained are worth further investigation. What makes the youngest child more likely to be
retained? Family dynamics; being perceived as the baby; the additional stress of meeting older
siblings’ needs; the amount of parental time available to the youngest; family resources available
for outside of school enrichment opportunities; and the possibility of prior difficult school
experiences with older siblings are factors that might be studied.
Behavior
Suspension in an elementary school setting is rare. However, retained students are almost
eight times as likely to be suspended as non-retained students. Further research should be
conducted to investigate the relationship between behavior and retention. This study did not
demonstrate whether behavior impacted learning and thus led to retention, or if retention
exacerbated behavior and thus led to a higher rate of suspensions. Because fidelity to the
district’s academic guidelines for retention was so low, behavior may in fact have been one
EFFECTS OF RETENTION 88
masked factor that was the real reason for retention to begin with for the students who later had a
disproportionate number of suspensions.
Maturity as a Reason for Retention
Additional research that would add to our understanding of the effects of retention is a
comparison study on the social/emotional/ maturity factors that precede and follow retention.
Since it is clear that factors other than academic performance played a role in these students
being retained, it would be helpful to pursue this line of study.
Finally, the numbers and percentages of any study are enriched by the stories the players
tell. A qualitative investigation that parallels this study would provide additional insight in to
why retention does not increase student achievement.
Policy Dilemmas
This research has clearly demonstrated that retention does not benefit RUSD students
academically. Even if the extra year does provide time for the student to mature (which was not
measured as a part of this study), that possible advantage of an extra year to increase maturity
does not result in greater academic success. The California Department of Education (CDE) has
not conducted any study of the benefits of retention (California Department of Education, n.d),
and this introduces a quandary for RUSD: State law requires a practice that does not benefit
district students at a cost of millions of dollars each year.
However, state law (for elementary schools) only requires policies for retention or
promotion at the end of second, third, fifth and sixth grades. It allows that kindergarten students
may be retained by parent request with school agreement and parent signed consent. The current
study found that the majority of the retentions took place in K and first grade. Only 11.3%
(n=20) took place at grade levels where retention connected to academic performance is State
mandated.
EFFECTS OF RETENTION 89
Policy Recommendations
Lorence (2006) has drawn attention to the fact that there is little agreement about what
constitutes a successful retention. This study has demonstrated that retention does not result in
increased academic success for retained students compared to their non-retained peers. The
district should decide what the purpose and goal of retention should be, and use that as a basis
for writing the policy. Should retention success be measured by any increase in the students’
scores, by the student meeting grade level requirements post-retention, or by the student
exceeding other low performing non-retained peers? The present research demonstrated that
some retained students increased their own academic performance and met grade level
requirements post-retention, but in neither case did the data endorse retention as the cause for the
increase in performance.
The stated goal of the guidelines in the district’s Promotion/Retention Handbook is to
improve student learning and support students who are at risk for failure. This research has
demonstrated that the policy is not meeting success (improving student learning), nor is it being
implemented with fidelity (supporting students who are at risk for failure). One possibility that
would allow the district to act within the law and still be responsive to the research within their
own district is to rewrite the guidelines for K and 1
st
grade retentions.
Current academic guidelines are mostly not being followed, and when they are, the
results are not long term success as measured by CST scores and teacher report card grades. The
district should examine the other reasons K and first grade students are being retained(e.g.
maturity, age, language ability, birth order, and IEP status) , and write a policy that addresses
those reasons. Education Code 68070.5(j) makes allowances for that.
Additionally, the district should begin implementing and measuring the other procedures
that have been addressed already in this chapter. The district should investigate whether
EFFECTS OF RETENTION 90
increased, targeted, high quality, consistent intervention, and parent involvement in the SST
process results in remediating poor student performance. If it does, when K and 1
st
grade
students promote to 2
nd
grade, they would not meet state requirements for retention.
Conclusion
A Retrospective Control Group study investigated the effect of retention on student
academic performance using both teacher classroom grades and standardized CST scores in
order to ascertain whether retained students benefitted academically from retention. The study
investigated the question whether retained students would perform at the same level or even
better than a comparison group of students who had not been retained. A second study comprised
of currently retained students examined the classroom performance of retained students to
investigate whether they met grade level standards post-retention as measured by teacher
classroom scores, CST, and CELDT.
The results of both studies demonstrate that post retention:
1.) Students in a retrospective control group study did not perform as well as their
comparison group of low achieving non-retained students;
2) Almost half (48%) of the retained students did not show grade level proficiency on CST;
3) ELL students did not gain an advantage in language learning due to retention;
4) Teacher perception of student grade level performance indicated that only 24% of the
retained students were working within grade level.
Thus, the practice of retaining a student in this district does not benefit the student
academically.
EFFECTS OF RETENTION 91
References
Alexander, K.L., Entwistle, D.R., and Dabner, S.L. (1994). On the Success of Failure: A
Reassessment of the Effects of Retention in the Primary Grades. New York: Cambridge
University Press.

Beebe-Frankenberger, M., Bocian, K.M., MacMillan, D.L., & Gresham, F.M. (2004). Sorting
second grade students: Differentiating those retained from those promoted. Journal of
Educational Psychology, 96, 204-215.

Brophy, J. (2006). Grade repetition. Education policy series, 6, 420-437.

California Department of Education, (n.d.). FAQs Promotion and retention. Retrieved February
13, 2012 from http://www.cde.ca.gov/re/lr/pr/faqppr.asp

California Education Code 48070. Laws and regulations. Retrieved February 13, 2012 from
http://www.cde.ca.gov/re/lr/

Clinton, W. (1998). State of the union address: Full text. Retrieved May 12, 2012,
fromhttp://www.washingtonpost.com/wpsrv/politics/special/states/docs/sou 98.htm

Frey, N. (2005). Retention, social promotion, and academic redshirting: What do we know and
need to know? Remedial and Special Education, 26, 6. 332-346.

EFFECTS OF RETENTION 92
Goos, M., Van Damme, J., Onghena, P., & Petry, K. (2011). First-grade retention:
Effects on children's actual and perceived performance throughout elementary education.
Society for Research on Educational Effectiveness.

Holmes, C.T.(1989) Grade level retention effects: A meta-analysis of research studies . In L.A.
Shepard & M.I. Smith (Eds.), Flunking grades: Research and policies on retention (pp.
16-33). London: Falmer.

Holmes, C.T., & Matthews, K.M. (1984). The effects of nonpromotion on elementary and junior
high school pupil: A meta-analysis. Review of Educational Research, 54. 225-236.

Hong, G. & Raudenbush, S.W. (2005). Effects of kindergarten retention policy on children’s
cognitive growth in reading and mathematics. Educational Evaluation and Policy
Analysis, 27, 3, 205-224.

Jackson, G. (1975). The research evidence on the effects of grade retention. Review of
Educational Research, 45, 613-635.

Jimerson, S., Anderson, G., & Whipple, A. (2002). Winning the battle and losing the war:
Examining the relationship between grade retention and dropping out of high school.
Psychology in the Schools, 39, (4), 441-457.

EFFECTS OF RETENTION 93
Jimerson, S.R. (2001). A synthesis of grade retention research: Looking backward and moving
forward. The California School Psychologist, 6, 46-59.

Jimerson, S.R. & Kaufman, A.M. (2003) Reading, writing, and retention: A primer on grade
retention research. The Reading Teacher, 56, (7), 622-635.

Karweit, N.L. (1992). Retention Policy. In M.C. Alkin (ed) Encyclopedia of Educational
Research (6
th
ed.), New York: Macmillan.

Martin, A.J. (2011) Holding back and holding behind: grade retention and students’ non-
academic and academic outcomes. British Educational Research Journal, 37,(5), 739-
763.

Peterson, L.S. & Hughes, J.N. (2011). The differences between retained and promoted children
in educational services received. Psychology in the Schools, 48(2), 156-165.

Pierson, L. and Connell, J.P. (1992). Effect of grade retention on self-system processes, social
engagement, and academic performance. Journal of Educational Psychology, 84(3), 300-
307.

Plummer, D.L., & Graziano, W.G. (1987). Impact of grade retention on the social development
of elementary school children. Developmental Psychology, 23 (2), 267-275.

EFFECTS OF RETENTION 94
Smith, M.L. & Shepard, L.A. (1988). Kindergarten readiness and retention: A Qualitative study
of teachers’ beliefs and practices. American Educational Research Journal, 25 (3), 307-
333.

Taylor, N., Mabogane, T., Shindler, J., & Akoobhai, B. (2010). Seeds of their struggle: The
features of under-and overage enrollment among grade 4 learners in South Africa.
Consortium for Research on Educational Access, Transitions, and Equity.

Willson, V.L. & Hughes, J.N. (20050. Retention of Hispanic/Latina students in first grade:
Child, parent, teacher, school and peer predictors. Journal of School Psychology 44, 31-
49.

Wu, W.; West, S.G.; & Hughes, J.N. (2008). Effect of retention in first grade on children’s
achievement trajectories over 4 years: A Piecewise growth analysis using propensity
score matching. Journal of Educational Psychology, 100(4) 727-740.

EFFECTS OF RETENTION 95
Appendix A
Social Promotion Retention Laws

California Department of Education (http://www.cde.ca.gov/re/lr/pr/index.asp)
Page Generated: 2/8/2012 2:57:00 PM.
Display version
Pupil Promotion & Retention
Current law and information on pupil promotion and retention policies and related supplemental
instruction programs.

Background Information on Pupil Promotion and Retention and Related Supplemental
Instruction
Until 1998, California law only required school districts to have policies pertaining to Pupil
Promotion and Retention (PPR) (see California Education Code [EC] Section 48070), without
specifying any requirements for those policies. With a growing concern about the "social
promotion" of students who do not meet grade-level standards, in 1998 the California Legislature
passed, and the Governor signed, a new set of laws which transformed the way students move
from grade level to grade level . The three-bill package, Assembly Bill (AB) 1626 (Chapter 742,
Statutes of 1998), AB 1639 (Chapter 743, Statutes of 1998) and Senate Bill (SB) 1370 (Chapter
942, Statutes of 1998), outlined new pupil promotion and retention policy requirements and
provided for related supplemental instruction programs. These laws became effective in January
1999. As part of the larger standards-based school reform movement, this legislation outlined a
promotion process based on student attainment of grade-level content standards (EC Section
EFFECTS OF RETENTION 96
48070.5 (a) through (j)), and at the same time provided funding for intensive instructional
support for those students who are at risk of being retained (in grades two through six) or who
are recommended for retention or retained (in grades two through nine) (see EC sections 37252.2
and 37252.8).
The following links are pertinent California Education Code sections:
California Education Code sections 37252-37253.5 (Outside Source)
Summer school and other supplemental instruction.
California Education Code sections 48070-48070.5 (Outside Source)
Promotion and retention.
California Education Code sections 48010-48011 (Outside Source)
Parental permission for retention in kindergarten.
The next two links provide information and answers to specific questions regarding Pupil
Promotion and Retention and PPR-related and other Supplemental Instruction (SI) Programs.
Frequently Asked Questions — Pupil Promotion and Retention (Updated 7-Apr-2010)
Frequently Asked Questions — Supplemental Instruction Programs (Updated 7-Apr-2010)
Questions: Literacy, History, and Arts Leadership Office mautry@cde.ca.gov |916-323-
6269
Last Reviewed: Friday, February 04, 2011

EFFECTS OF RETENTION 97
Appendix B
Pupil Record Card

EFFECTS OF RETENTION 98
Appendix C
Academic Intervention

ACADEMIC
INTERVENTION

Guidelines for
Identifying and Assisting
Students at Risk of Retention
and
Intervention Strategies

Revised June 2003

EFFECTS OF RETENTION 99
INTRODUCTION

The intent of promotion/retention standards is to focus resources toward improving student
learning and supporting students who are at-risk of failure. Expectations are to rethink how we
use supplemental services and funding sources to provide optimal support for students who are
retained, at risk of being retained, or who perform below grade level in math, reading and
language arts.

The information in this handbook is intended to furnish teachers with district policy, a
timeline, data collection forms and guidelines for identifying students at risk of retention.
Once a student is determined to be a candidate for retention, periodic re-evaluation will
proceed with interventions developed specific to student need.
DISTRICT POLICY 5123
STUDENTS

5123 Promotion/Acceleration/Retention

The Governing Board expects students to progress through each grade within
one school year. To accomplish this, instruction should accommodate the
varying interests and growth patterns of individual students and include
strategies for addressing academic deficiencies when needed.

Students shall progress through the grade levels by demonstrating growth in
EFFECTS OF RETENTION 100
learning and meeting grade-level standards of expected student achievement.

When high academic achievement is evident, the Principal and school staff may
recommend a student for acceleration into a higher grade level. The student’s
social and emotional growth shall be taken into consideration in making a
determination to accelerate a student.

Students shall be considered for promotion and retention in specific subject areas
and between certain grade levels:

1. Kindergarten and first grade in reading
2. First grade and second grade in reading
3. Second grade and third grade in reading
4. Third grade and fourth grade in reading
5. Fourth grade and fifth grade in reading, English language arts and math
6. Fifth grade and sixth grade in reading, English language arts and math

The end of sixth grade and the beginning of junior high school grades in reading,
English language arts and math

As early as possible in the school year and in students’ school careers, the
Principal or school staff shall identify students who should be retained and who
EFFECTS OF RETENTION 101
are at risk of being retained in accordance with law, Board policy, administrative
regulation and the following criteria:

Students shall be identified on, but not limited to the following criteria:
• Report Card Grades
• District CRTs
• Writing Samples
• Running Records
• Scores on nationally normed standardized test (such as CAT 6)
• Student Portfolios

DISTRICT POLICY 5123

STUDENTS

5123 Promotion/Acceleration/Retention(continued)

It is the intent of the District to provide intervention programs for any student at
risk of not meeting grade level standards. If a previously retained student still
does not meet grade level standards, that student shall participate in the
intervention programs. A second retention of that student requires approval of
the superintendent. A notification for a second retention in the District requires
approval of the Superintendent.
EFFECTS OF RETENTION 102
When a student is recommended for retention or is identified as being at risk for
retention, the Principal or school staff shall provide opportunities for remedial
instruction to assist the student in overcoming his/her academic deficiencies.
Such opportunities may include but are not limited to tutorial programs, after-
school programs, summer school programs and/or consultation with a student
study team.
A process shall be established to enable the parent or guardian to appeal the
recommendation of retention of their child. It is the responsibility of the appealing party to
show why the child should not be retained. The District has a process for appealing a
decision to retain a student.

Legal Reference:
EDUCATION CODE
37252-37253 Summer school
46300 Method of computing ADA
48011 Admission on completing kindergarten; grade placement of pupils coming from other
districts
48070-48070.5 Promotion and retention
48431.6 Required systematic review of students and grading
51215 Proficiency standards in basic skills
51216 Assessment of pupil proficiency
51217 Withholding diploma (high school)
51218 Separate proficiency standards
56345 Elements of individualized education plan
60641-60647 Standardized Testing and Reporting Program
EFFECTS OF RETENTION 103
60648 Minimum performance levels
CODE OF REGULATIONS, TITLE 5
200-202 Admission and exclusion of students
Management Resources:
CDE PROGRAM ADVISORIES
06121.89 - 06123.89 Educating Young Children: Next Steps in Implementing the School
Readiness Task Force Report
0916.91 Retention of Students in Elementary and Middle Grades, CIL 91/92-02
CDE MANAGEMENT ADVISORIES
0900.90 Changes in Law concerning Eligibility for Admission to Kindergarten 90-10
LEGISLATIVE COUNSEL OPINION
1001.98 Promotion and Retention #21610
CSBA ADVISORIES
1112.98 Student Promotion/Retention Advisory
WEB SITES
CSBA: http://www.csba.org
CDE: http://www.cde.ca.gov

DISTRICT ADMINISTRATIVE REGULATION 5123

STUDENTS

5123 Promotion/Acceleration/Retention

EFFECTS OF RETENTION 104
Kindergarten

Children five years or older who have completed one year of kindergarten shall be
admitted to first grade regardless of age, unless the parent/guardian and the district agree
that the student shall continue in kindergarten for not more than one additional school
year. (Education Code 48011)

Whenever a student continues in kindergarten for an additional year, the Principal shall
secure an agreement, signed by the parent/guardian, stating that the student shall continue
in kindergarten for not more than one additional school year. (Education Code 46300)

Acceleration from Kindergarten to First Grade

A student enrolled in kindergarten may be admitted to the first grade at the discretion of
the principal and upon determination that the student is ready for first grade work.
(Education Code 48011) Admission shall be subject to the following minimum criteria: (5
CCR 200)

1. The student is at least five years of age.

2. The student has attended a public school kindergarten for a long enough time to
enable school personnel to evaluate his/her ability.

EFFECTS OF RETENTION 105
3. The student is in the upper five percent of his/her age group in terms of general
mental ability.

4. The physical development and social maturity of the student are consistent with
his/her advanced mental ability.

5. The parent/guardian of the student has filed a written statement with the school
district approving the placement in first grade.

Continuation in Kindergarten

Students who have completed one year of kindergarten shall be admitted to first grade
unless the parent/guardian and the district agree that the student shall continue in
kindergarten for not more than one additional school year. (Education Code 48011)

Whenever a student continues in kindergarten for an additional year, the Principal shall
secure an agreement, signed by the parent/guardian, stating that the student shall continue
in kindergarten for not more than one additional school year. (Education Code 46300)

5123 Promotion/Acceleration/Retention (continued)

Retention at Other Grade Levels

EFFECTS OF RETENTION 106
The Principal and school staff shall identify students who should be retained or who are at
risk of being retained at the following grade levels: (Education Code 48070.5)

1. Between grades 1 and 2

2. Between grades 2 and 3

3. Between grades 3 and 4

4. Between grades 4 and 5

5. Between grades 5 and 6

6. Between the end of grades 6 and the beginning of junior high school grades

Students between kindergarten and grade 1 shall be identified primarily on the basis of
their reading readiness. Students between grades 1 and 2, grades 2 and 3 and grades 3 and
4 shall be identified primarily on the basis of their level of proficiency in reading.
Proficiency in reading, English language arts and mathematics shall be the basis for
identifying students between grades 4 and 5, grades 5 and 6 and grade 6 and junior high
school. (Education Code 48070.5)

Factors for consideration: English speaking students shall be identified on, but not limited
EFFECTS OF RETENTION 107
to, the following criteria:

• Report Card Grades
• District CRTs
• Writing Samples
• Running Records
• Scores on nationally normed standardized test (such as CAT 6)
• Student Portfolios

If a student is identified as performing below the minimum standard for promotion, the
student shall be retained in his/her current grade level unless the student’s regular
classroom teacher determines, in writing, that retention is not the appropriate intervention
for the student’s academic deficiencies. This determination shall specify the reasons that
retention is not appropriate for the student and shall include recommendations for
interventions other than retention that, in the opinion of the teacher, are necessary to assist
the student in attaining acceptable levels of academic achievement. (Education Code
48070.5)

The teacher’s evaluation shall be provided to and discussed with the student’s
parents/guardians and the principal or designee by April 1 before any final determination
of retention or promotion. (Education Code 48070.5) If the teacher’s recommendation to
promote is contingent on the student’s participation in a summer school or interim session
remediation program, the student’s academic performance shall be reassessed at the end of
EFFECTS OF RETENTION 108
the remediation program, and the decision to retain or promote the student shall be
reevaluated at that time.

If the student does not have a single regular classroom teacher, the principal or designee
shall specify the teacher(s) responsible for the decision to promote or retain the student.
(Education Code 48070.5)

When a student is identified as being at risk of retention, Principal and school staff shall so
notify the student’s parents/guardians as early in the school year as practicable. The
student’s parent/guardian shall be provided an opportunity to consult with the teacher(s)
responsible for the decision to promote or retain the student. (Education Code 48070.5)

Criteria for Promotion/Retention
Grades 1-3
Criteria for moderately at risk for retention:
• Basic on Reading/Language Arts multiple measures
• Report Card shows “Requires Additional Support” for both areas in Reading
• Writing Sample – rubric of 1 or 2
• CAT 6 – 26th–40th percentile on Total Reading and Language Arts
• More than 25 days absence

Criteria for significantly at risk for retention
EFFECTS OF RETENTION 109
• Below basic on Reading/Language Arts multiple measures:
• Report Card shows “Requires Additional Support” for both areas in Reading
• Writing Sample – rubric of 0-1
• CAT 6 – 17th–25th percentile on Total Reading and Language Arts
• More than 25 days absence

Criteria for Promotion/Retention
Grades 1-3
Criteria for retention:
• Identified as at risk for retention
• Below basic on Reading/Language Arts multiple measures:
• Report Card has ”Requires Additional Support” checked for both areas in Reading
• Writing Sample – rubric of 0
• CAT 6 – below 16th percentile on Total Reading and Language Arts
• More than 25 days absence

Grades 4-6
Criteria for moderately at risk for retention
• Basic on Reading/Language Arts or Math multiple measures:
• Report Card Grade N and checks in Reading Word Attack and Math Computation
• Writing Sample – rubric of 1-2
• Math CRTs – 60–69 percent
• CAT 6 – 26th–40th percentile on Total Reading, Language Arts and Math
EFFECTS OF RETENTION 110
• More than 25 days absence

Criteria for significantly at risk for retention:
• Below basic on Reading/Language Arts and Math multiple measures
• Report Card Grade N and checks in Reading Word Attack and Comprehension and
• Math Computation and Problem Solving/Application
• Writing Sample – rubric of 0
• Math CRTs – 50–59 percent
• CAT 6 – 17th–25th percentile on Total Reading, Language Arts and Math
• More than 25 days absence

Criteria for retention:
• Identified as at risk of retention:
• Below basic on Reading/Language Arts and Math multiple measures
• Report Card Grade N and checks in Reading Word Attack and Comprehension and
Math Computation, Problem Solving/Application and Mathematical
Communication
• Writing Sample – rubric of 0
• Math CRTs – below 50 percent
• CAT 6 – below 16th percentile on Total Reading, Language Arts and Math
• More than 25 days absence

English Language Learners
EFFECTS OF RETENTION 111

English Language Learners shall be monitored for their progress in the acquisition of
English. Upon redesignation, an English Language Learner shall be required to meet the
minimum standard for promotion. Intervention programs should be provided to all
English Language Learners. After four years, if not redesignated, student will be evaluated
for FEP status. Those students not redesignated would enter a mandatory at-risk
intervention program.

Special Education

Special education students are expected to meet 80 percent of their IEP goals for promotion
to the next grade level and demonstrate progress on standardized or alternative assessment
measures. The IEP team makes decisions regarding retention consideration.

Parent Notification

The Principal shall also provide a copy of the district’s promotion/retention policy and
administrative regulation to those parents/guardians who have been notified that his/her
child is at risk of retention.

The teacher’s decision to promote or retain a student may be appealed consistent with
Board policy, administrative regulation and law.

EFFECTS OF RETENTION 112
The burden shall be on the appealing party to show why the teacher’s decision should be
overruled. (Education Code 48070.5)

To appeal a teacher’s decision, the appealing party shall submit a written request to the
Superintendent or designee specifying the reasons why the teacher’s decision should be
overruled. The appeal must be initiated within 10 school days of the determination of
retention or promotion. The teacher shall be provided an opportunity to state orally
and/or in writing the criteria on which his/her decision was based.

Within 30 days of receiving the request, the Superintendent or designee shall determine
whether or not to overrule the teacher’s decision. Prior to making this determination, the
Superintendent or designee may meet with the appealing party and the teacher. If the
Superintendent or designee determines that the appealing party has overwhelmingly
proven that the teacher’s decision should be overruled, he/she shall overrule the teacher’s
decision.

The Superintendent or designee’s determination may be appealed by submitting a written
appeal to the Board within 15 school days. Within 30 days of receipt of a written appeal, the
Board shall meet in closed session to decide the appeal. The Board’s decision may be made
on the basis of documentation prepared as part of the appeal process or, at the discretion of
the Board, the Board may also meet with the appealing party, the teacher and the
Superintendent/designee to decide the appeal. The decision of the Board shall be final.

EFFECTS OF RETENTION 113
If the decision of the Board is unfavorable to the appealing party, he/she shall have the
right to submit a written statement of objections which shall become part of the student’s
record.

Remedial Instruction

With the parent/guardian’s consent, the Principal may require a student who has been
recommended for retention or has been identified as being at risk of retention to participate
in a supplemental instructional program. Such programs shall be offered during the
summer, after school, on Saturdays and/or during intersessions. Services shall not be
provided during the regular instructional day if it would result in the student being
removed from classroom instruction in the core curriculum. (Education Code 37252.5)

An intensive remedial program in reading or written expression should include:
- Phoneme awareness
- Systematic explicit phonics and decoding
- Word attack skills
- Spelling and vocabulary
- Explicit instruction of reading comprehension
- Writing
- Study skills

These services shall be provided to students in the following priority order: (Education
EFFECTS OF RETENTION 114
Code 37252.5)

1. Students who have been recommended for retention or who have been identified as
being at risk of retention pursuant to Education Code 48070.5

2. Students who have been identified as having a deficiency in mathematics, reading or
written expression based on the results of the tests administered under the State
testing program

This supplemental instruction program shall be developed in accordance with the
requirements of Education Code 37252.5.

Recommended Time Line
At any stage in this process, a student may be determined to display sufficient progress to justify removal
from the “at risk” designation.

Interventions
Documentation of supplemental service efforts is critical. Not only will funding be
based on these data, but parents should be informed of all intervention efforts. The
following list suggests options for developing the Academic Intervention Plan.

Tutoring
EFFECTS OF RETENTION 115
Cross-age tutoring
Summer School
Intersession
Home/School Contract
Saturday School
Extend the school day-before or after school

ADJUSTED CURRICULUM
Make sure materials are at student’s instructional level
Present lessons in small increments allowing practice time and intermittent reinforcement
Give clear, concrete directions
Team with another teacher who has appropriate level groups
Use techniques of reading lab, math lab, aides and other support personnel
Tape record lessons in content areas for students who have difficulty with reading or need repetition
Use multi-sensory approach in directed lessons
Let student do extra credit according to interest
Establish very consistent routine and explain any changes
Establish temporary groups for specific skills
Use cooperative learning techniques
Use concrete teaching devices
Set-up learning centers for study skills, i.e. following directions
Consult specialists and principal for ideas

MODIFIED ASSIGNMENTS
Shorten assignments
Adjust level of difficulty; cut paper in half with scissors
EFFECTS OF RETENTION 116
Present follow-up pages one at a time and number them
Be sure student can verbalize directions to teacher, aid or “study buddy”
Provide folder for routine work and label it “Work to be Done” or “Work That is Complete”
Provide student with individual list of work to be done
Change schedules or adjust time spent in each activity. Vary sedentary and physical activities
Allow responses to demonstrate knowledge required by using tape recorder, Systems 80, language
master, person acting as a secretary, art project, etc.
Use timer to increase on-task behavior
Use visual examples on papers
Send positive notes /make positive calls to parents

LEARNING CONTRACTS FOR BEHAVIOR
Define for student the behavior expected and have him/her role play it. Then make a contract with the
student
Take a Polaroid picture of student behaving as you want (working on task, etc.) and tape the picture to
the desk
Time out
Immediate reward after expected behavior
Establish token economy. A simple method is passing out paper squares and giving an award for a
certain number of squares
Give puzzle pieces to reward on-task behavior. The student can put the puzzle together after pieces are
earned
Establish a reward system with parents
Chart and graph appropriate behaviors
Find out student’s interests (art, helping principal, etc.) and let him earn time to do related activity
Take home reward sheets
EFFECTS OF RETENTION 117

INSTRUCTIONAL ASSISTANT
Provide one-on-one assistance for needed subjects for short period each day.
Manage contracts for on-task behavior
Help determine interest
Provide immediate feedback for answers
Check with student to determine if directions are clear
Review concept taught at group during seat work time

ALTERED ENVIRONMENT
Group according to sociogram
Seat student next to achiever or appropriate role model
Move student closer to board or teacher
Lessen distractions with a study carrel as student feels need
Clear desk of all activities except current task and pencil
Help student to organize desk and materials
Participate in another classroom activity when applicable
Teach student relaxation techniques as needed

PARENT CONTRACT
Explain student’s strengths and weaknesses
Give concise directions for helping student
Set-up reward system based on positive notes
Call and give positive reports to parents
Keep open communication at all times
Explain Student Study Team if student is referred
EFFECTS OF RETENTION 118
Sample of Letter to be sent to home with all students

Dear Parents,

As the new school year begins, I would like to let you know about the district’s
promotion and retention policy for kindergarten through sixth grade students. This policy
establishes criteria for promotion from one grade level to the next. It also provides criteria
for identifying a student as “at-risk for retention.” I know parents hold high academic
expectations for every child, and we want to provide you with all of the support and
encouragement possible to insure your child’s success.

Students will be identified as at-risk of retention based on the following:
Reading/Language Arts
• Report Card Grade N and “Requires Additional Support” checked in the area
of Reading
• Writing Sample score of 1 or zero
• Standardized test (CAT 6) results below the 40th percentile
Mathematics
• Report Card Grade N and “Requires Additional Support” checked in the area
of Math
• District Mathematics test results below 70%
EFFECTS OF RETENTION 119
• Standardized test (CAT 6) results below the 40th percentile

If a student is identified as at-risk, the teacher will notify the parent and an academic
intervention plan will be developed. The teacher will continue to closely monitor student
progress. The promotion/retention process will include the teacher, parents, and school
principal.
Students enrolled in a Special Education program will be monitored through their
Individual Education Plan (IEP) and will meet all criteria upon exit of the Special Education
program. Students enrolled in an English Language Development Program (ELD) will be
considered at-risk and will meet all criteria upon redesignation to Fluent English Proficient
(FEP).
I know your child’s principal and teacher will be providing you with additional
information. Thank you for your support as we work together to insure high academic
achievement for every student.

Sincerely,

Joan M. Lucid, Ed.D.
Assistant Superintendent, Instruction

EFFECTS OF RETENTION 120

Intervention Planning Conference

To the Parent(s) of ________________________________ Date _____________________

This letter is to inform you that at this time your child has been identified as being at risk of
retention based on the following:

Reading/Language Arts
• Progress Report grade “N” and the notation ”Requires Additional Support “ in
Reading
• Writing Sample rubric score of 1 or zero
• Standardized test (CAT 6) score below the 40th percentile
Mathematics
EFFECTS OF RETENTION 121
• District Math assessment score of 69% or less
• Progress Report grade “N” and the notation “requires additional support” in Math
• Standardized test (CAT 6) score below the 40th percentile

According to Regent Union School District’s Promotion and Retention Policy, students
unable to attain minimum grade-level standards may be recommended for retention.
Working with you, we will discuss options and develop an Intervention Plan.

Please plan to attend this conference on _____________at _______________ a.m./p.m. in

___________________________.

If you are not available at this time please indicate options which would work for you on
the tear-off at the bottom of this page. Also, note the *telephone number where I may call
you to confirm a new appointment.

Sincerely,

Intervention Planning Conference

Yes, I will attend on __________ as you have scheduled.
EFFECTS OF RETENTION 122

No, I cannot make the scheduled meeting, but I can come at these times:

Date _______________________ Time ___________________, or

Date _______________________ Time ___________________

Student’s Name ____________________________

Parent’s Signature___________________________ *Telephone number
________________________

Form 354M (4/03)

Procedures for Supplemental Instruction Intervention

A student referred to an extended school day, Saturday or intersession intervention
must be tracked accordingly. The attendance forms must be completed completely and
accurately in order for the district to receive the appropriate funding and the intervention
efforts to continue.

Each school is allotted Pupil Promotion/Retention funding based on 5% of the
EFFECTS OF RETENTION 123
previous year’s CBEDS enrollment for grades 2-6. The 5% are then multiplied times 120
hours of intervention time - times the state rate. Students may be enrolled in the program
for the number of hours that are needed for the appropriate intervention in
reading/language arts and math. One student may need 40 hours, whereas another may
require 160 additional hours.

Student attendance sheets and teacher time sheets must be prepared and reviewed
carefully. The number of student hours times the state rate should be sufficient to cover the
cost of the teacher’s time. Hours must only be reported in hour increments (not partial
hours).

If you have any questions or need additional information, please contact the
Instruction Office.

Special Considerations

SPECIAL EDUCATION

Districts receiving special education funding are required to comply with both the
EFFECTS OF RETENTION 124
Individuals with Disabilities Education Act (20 USC Section 1400 et seq.) and Section 504 of the
Rehabilitation Act (29 USC Section 794 et seq.). Students enrolled in Special Education receive
an Individualized Education Plan that should include the following:

The Individualized Education Plan (IEP) is a written document that includes:
• The present levels of the student’s education performance
• The measurable annual goals, including benchmarks, or short term objectives related to
meeting the student’s needs that result from the student’s disability to enable the student to
be involved in and progress in the general curriculum and,
• Meeting each of the student’s other education needs that result from the student’s disability;
• The specific special education instruction and related services and supplementary aids and
services to be provided to the student, or on behalf of the student, and a statement of the
program modification or supports for school personnel that will be provided for the student
to do the following:
-To advance appropriately toward attaining the annual goals
-To be involved and progress in the general curriculum

Also included in the IEP are appropriate objective criteria, evaluation procedures and
schedules for determining, on at least an annual basis, whether the annual goals are being
achieved. The IEP shall also include the determination of the IEP team as to whether
differential proficiency shall be developed for the student. All information should be noted in
the IEP.
It is possible that special education students can be retained, but only under certain
circumstances. The previous information should be used as a guidepost when considering the
EFFECTS OF RETENTION 125
decision about promotion and retention for a student in Special Education. Students who exit
the Special Education program are held to the same standards as all other students.
Decisions regarding the promotion or retention of English Language Learners (ELL)
require special considerations. It is inappropriate to retain English Language Learners who
have failed to meet academic standards in the areas in which they have been provided only
limited instruction. Students who have been enrolled less than one year may be identified as at-
risk up to four years.

ENGLISH LANGUAGE LEARNERS (ELL)

All ELL students in District will be assessed with the appropriate ELD assessments
based on the ELD Content Standards. Many will also be assessed by existing English
Assessments that are based on the Language Arts Content Standards, depending on English
proficiency. When an ELL student is redesignated to FEP, all other promotion/retention
standards apply.
Any ELL student with special needs and so identified should have records reviewed
periodically as his/her IEP dictates. These students are considered at-risk.

CALIFORNIA DEPARTMENT OF EDUCATION
Key Elements of AB 1626, Pupil Promotion and Retention
Education Code
Key Elements Section
I. Local Policy on Pupil Promotion and Retention
In addition to existing local policies for pupil retention and promotion (adopted pursuant to Education
EFFECTS OF RETENTION 126
Code § 48070), the governing board of each school district and each county board of education shall
approve a policy regarding the promotion and retention of pupils between the following grades:
(1) Second grade and third grade.
(2) Third grade and fourth grade.
(3) Fourth grade and fifth grade.
(4) The end of the intermediate grades and the beginning of middle school grades.
(5) The end of middle school grades and the beginning of high school.

• The policy shall base the identification of pupils “(1) Between second grade and third grade” and
“(2) Between third grade and fourth grade” primarily on the basis of the pupils’ levels of proficiency in
reading.
• The policy shall base the identification of pupils “(3) Between fourth grade and fifth grade,” “(4)
Between the end of the intermediate grades and the beginning of middle school grades,” and “(5)
Between the end of middle school grades and the beginning of high school” primarily on the basis of
the pupils’ levels of proficiency in reading, English language arts, and mathematics.

II. Minimum Levels of Pupil Performance on STAR
The Superintendent of Public Instruction shall recommend and the State Board of Education shall adopt
the levels of pupil performance for the achievement tests administered under the STAR Program
(Education Code § 60640-60647) in reading, English language arts, and mathematics. The performance
levels shall be the minimum levels required for satisfactory performance in the next grade and shall be
adopted only after the STAR tests have been aligned [pursuant to Education Code § 60643(a)(3)] to the State
Board of Education adopted content and performance standards [pursuant to Education Code § 60605(a)].

Ill. Criteria for Retention
The local policy approved for Education Code § 48070.5(a) (see I. above) shall identify pupils who should
be retained and who are at risk of being retained in their current grades on the basis of either of the
following:
(1) Results of the STAR test and the minimum levels of proficiency recommended by the State Board of
EFFECTS OF RETENTION 127
Education pursuant to Education Code § 60648 (see II. above).
(2) Pupils’ grades and other indicators of academic achievement designated by the district.

IV. Exception to Retention Criteria
If either measure (1) or (2), pursuant to Education Code § 48070.5(b) (see III. above), identifies that a pupil is
performing below the minimum standard for promotion, the pupil shall be retained unless the pupil’s
regular classroom teacher specifies in writing that retention is not the appropriate intervention. This
written determination shall specify:
• The reasons that retention is not appropriate for the pupil and
• Recommendations for interventions other than retention that in the opinion of the teacher are necessary
to assist the pupil to attain acceptable levels of academic achievement.

If the teacher’s recommendation to promote is contingent upon the pupil’s participation in a remediation
program, the pupil’s academic performance shall be reassessed at the end of the remediation program and
the decision to retain or promote shall be reevaluated at that time.

The teacher’s evaluation shall be provided to and discussed with the pupil’s parent or guardian and the
school principal before any final determination of pupil retention or promotion.

V. What Shall Be Included in New Local Policy
The policy shall:

• Provide for parental notification when a pupil is identified as being at risk of retention. This
notice shall be provided as early in the school year as practicable. The policy shall provide a
pupil’s parent or guardian the opportunity to consult with the teacher or teachers responsible for
the decision to promote or retain the pupil.

• Provide a process through which the decision of the teacher to retain or promote a pupil may be
EFFECTS OF RETENTION 128
appealed. If an appeal is made, the burden shall be on the appealing party to show why the
decision of the teacher should be overruled.

• Provide that pupils who are at risk of being retained in their current grades be identified as early
in the school year, and as early in their school careers, as practicable.

• Indicate the manner in which opportunities for remedial instruction will be provided to pupils
who are recommended for retention or who are identified as being at risk for retention.

• Specify the teacher or teachers responsible for the promotion or retention decision if the pupil
does not have a single regular classroom teacher.

• Be adopted at a public meeting of the governing board.

VI. Exceeding Retention Criteria
Nothing in AB 1626 prohibits the retention of a pupil not included in grade levels identified
pursuant to Education Code § 48070.5(a) (see I. above), or for reasons other than those specified
in Education Code § 48070.5(b) (see 111. above), if such retention is determined to be appropriate
for that pupil. Nothing in AB 1626 shall be construed to prohibit a governing board from
adopting promotion and retention policies that exceed the criteria established in AB 1626.

Vll. Mandated Costs
If the Commission on State Mandates determines that this act contains costs mandated by the
state, reimbursement to local agencies and school districts for those costs shall be made pursuant
to Part 7 (commencing with Section 17500) of Division 4 of Title 2 of the Government Code. If the
statewide cost of the claim for reimbursement does not exceed $1,000,000, reimbursement shall be
EFFECTS OF RETENTION 129
made from the State Mandates Claims Fund.

VII. Effective Date of Provisions
The provisions of AB 1626 shall become operative on the same date that the act takes effect
pursuant to the California Constitution (January 1, 1999).

CALIFORNIA DEPARTMENT OF EDUCATION
Key Elements of AB 1639, Mandatory Summer School:
Required Intensive Instructional Programs
Education Code
Key Elements Section
I. Supplemental Instruction for Pupils in Grades 2 through 9 Who Have
Been Retained
In addition to existing requirements to offer summer school programs (adopted pursuant to
Education Code § 37252), the governing board of each district maintaining any or all of grades
2 to 9, inclusive, shall offer programs of direct, systematic, and intensive supplemental
instruction to pupils enrolled in grades 2 to 9, inclusive, who have been retained pursuant to
Education Code § 48070.5 (see AB 1626).

A school district may require a pupil who has been retained to participate in supplemental
instructional programs. The district shall provide a mechanism for a parent or guardian to
decline to enroll his or her child in the program. Attendance in supplemental instructional
programs shall not be compulsory within the meaning of Education Code § 48200 (compulsory,
full-time education).
EFFECTS OF RETENTION 130

II. Programs of Supplemental Instruction for Pupils in Grades 2 through 6 Who Have Low Scores
The governing board of each district maintaining any or all of grades 2 to 6, inclusive, may
offer programs of direct, systematic, and intensive supplemental instruction to pupils enrolled
in grades 2 to 6, inclusive, who have low mathematics, reading, or written expression scores to
allow those pupils to achieve proficiency in standards adopted by the State Board of Education.

Services offered shall be provided to pupils in the following priority order:
(1) Pupils who have been recommended for retention or who have been identified as being at
risk of retention pursuant to Education Code § 48070.5 or school district policies.
(2) Pupils who have been identified as having a deficiency in mathematics, reading, or written
expression based on the results of STAR testing (Education Code § 60640-60647).

Each school district shall use results from tests administered under the STAR Program
(Education Code § 60640-60647) or other evaluative criteria to identify eligible pupils pursuant
to Education Code § 37252.5 (b).

III. Pupils Completing Grade 6 or Grade 9
For purposes of this section, a pupil shall be considered to be enrolled in a grade immediately
upon completion of the preceding grade. Summer school instruction may also be offered to
pupils who were enrolled in grade 6 or grade 9 during the prior fiscal year after the completion
of grade 6 or grade 9, respectively.

IV. Legislative Intent
It is the intent of the Legislature that pupils who are at risk of failing to meet state adopted
standards, or who are at risk of retention, be identified as early in the school year, and as early
EFFECTS OF RETENTION 131
in their school careers as possible and be provided the opportunity for supplemental
instruction sufficient to assist them in attaining expected levels of academic achievement.

V. Program Implementation
Supplemental educational services pursuant to Education Code § 37252.5 (a) and (b) shall be
provided during the summer, after school, on Saturdays, or during intersession, or in a
combination of summer school, after school, Saturday, or intersession instruction. Services
shall not be provided during the pupil’s regular instructional day if doing so would result in the
pupil’s being removed from classroom instruction in the core curriculum.

An intensive remedial program in reading or written expression offered pursuant to AB 1639
shall, as needed, include instruction in phoneme awareness, systematic explicit phonics and
decoding, word attack skills, spelling and vocabulary, explicit instruction in reading
comprehension, writing, and study skills.

Each school district shall seek the active involvement of parents and classroom teachers in the
development and implementation of supplemental instructional programs provided pursuant to
AB 1639.
VI. Funding Provisions
(1) The maximum amount of funding for the purposes of programs offered pursuant to AB 1639 to serve
pupils in grades 2 to 6, inclusive, shall not exceed 10 percent of the statewide total enrollment in
grades 2 to 6, inclusive, for the prior fiscal year multiplied by 120 hours, multiplied by the hourly
rate for the current fiscal year determined pursuant to Education Code § 42239(c) (see SB 1370). Any
funding provided for the purposes of AB 1639 shall first be used by the district to provide services
required pursuant to Education Code § 37252.5 (a) (retained students, see 1. above).
The funding shall be allocated in the following manner.
EFFECTS OF RETENTION 132
(A) Notwithstanding Education Code § 42239(e), a school district that offers instruction
pursuant to Education Code § 37252.5 (a) and (b) (see I. and II. above) to serve pupils in
grades 2 to 6, inclusive, shall be entitled to receive an additional reimbursement in an
amount up to 5 percent of the district’s total enrollment in grades 2 to 6, inclusive, for the
prior fiscal year multiplied by 120 hours, multiplied by the hourly rate for the current fiscal
year determined pursuant to Education Code § 42239(c). (see SB 1370)
(B) The balance of the appropriation made for the purposes of funding programs offered pursuant to
this section to serve pupils in grades 2 to 6, inclusive, shall be allocated for reimbursement for
pupil attendance in instruction pursuant to Education Code § 37252.5 (a) and (b) that is in excess
of 5 percent of the district’s enrollment for the prior year in grades 2 to 6, inclusive, multiplied by
120 hours, multiplied by the hourly rate for the current fiscal year determined pursuant to
Education Code § 42239(c). (see SB 1370)
(2) If the funds claimed by school districts pursuant to the above subparagraph (B) exceed the
available balance of the appropriation made for the purposes of funding programs offered
pursuant to the above paragraph (1) after the minimum allocation to eligible districts has been
made pursuant to the above subparagraph (A), the allocation of the balance shall be prorated
based on each district’s share of the total additional hours of instruction offered pursuant to
the above subparagraph (B).
VII. Mandated Costs
If the Commission on State Mandates determines that this act contains costs mandated by the
state, reimbursement to local agencies and school districts for those costs shall be made pursuant
to Part 7 (commencing with Section 17500) of Division 4 of Title 2 of the Government Code. If the
statewide cost of the claim for reimbursement does not exceed $1,000,000, reimbursement shall be
made from the State Mandates Claims Fund.
VIII. Urgency Clause
EFFECTS OF RETENTION 133
The provisions of AB 1639 took effect as of September 23, 1998.Appendix D
Enrollment Form

Abstract (if available)

Abstract During conferences to decide whether or not to retain a child, usually two questions are asked. The first is, “Will retention bring success in school?” The other is, “If we retain him, will he perform better in school than if we don’t retain him?” The goal of this research was to see if retained students would make enough academic growth due to retention that they would to catch up to their peers, or even surpass them as a result of being retained. Two separate studies were conducted to measure the academic effects of retention. The first study identified the group of 1999-2000 students who were retained under the changing California law that banned social promotion. Propensity scores were used to create a retrospective control group of K-2 students from 1998-99 who would have been retained if the law had been adopted a year earlier. The second study examined records of retained students from the 2011-2012 school year. Performance pre- and post-retention was measured using California Standards Test (CST) scores, California English Development Test (CELDT) scores, and teachers’ scores. Results indicate that retention does not benefit students academically. Retained students did not meet the level of CST performance of their non-retained peers

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Academic achievement among Hmong students in California: a quantitative and comparative analysis

PDF

A longitudinal study on the performance of English language learners in English language arts in California from 2003-2012

PDF

The impact of work study on the integration and retention of undergraduate students at the University of California

PDF

The effects of open enrollment, curriculum alignment, and data-driven instruction on the test performance of English language learners (ELLS) and re-designated fluent English proficient students ...

PDF

The impact of learning communities on the retention and academic integration of Latino students at a highly selective private four-year institution

PDF

Examining the effectiveness of the intervention programs for English learners at MFC intermediate school

PDF

Acceptance, belonging, and capital: the impact of socioeconomic status at a highly selective, private, university

PDF

The effect of a language arts intervention on the academic achievement of English language learners

PDF

The effectiveness of a district's social skills curriculum for students with disabilities

PDF

Use of accountability indicators to evaluate elementary school principal performance

PDF

CST performance of English language learners in two neighboring districts from 2002-03 to 2012-13

PDF

Assessing and addressing random and systematic measurement error in performance indicators of institutional effectiveness in the community college

PDF

An evaluation of the School Assistance and Intervention Team process in California public schools: lessons learned and indications for policy change

PDF

Raising student achievement at Eberman Elementary School with effective teaching strategies

PDF

Examining the effectiveness of teacher training on the improvement of California standardized test scores at Eva B. Elementary School

PDF

The impact of the Norton High School early college program on the academic performance of students at Norton High School

PDF

School connectedness: a comparison of students' and staff school connectedness perceptions

PDF

How does the evidence-based method of training impact learning transfer, motivation, self-efficacy, and mastery goal orientation compared to the traditional method of training in Brazilian Jiu-Jitsu?

PDF

Why should they stay? a social network analysis of teacher retention

PDF

A comparison of value-added, orginary least squares regression, and the California Star accountability indicators

Asset Metadata

Creator Hamlin, Christine A. (author)

Core Title The practice and effects of a school district's retention policies

School Rossier School of Education

Degree Doctor of Education

Degree Program Education (Leadership)

Publication Date 04/12/2013

Defense Date 02/20/2013

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

Tag academic effects of retention,grade level retention,OAI-PMH Harvest,social promotion

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Hocevar, Dennis (committee chair), Lucid, Joan (committee member), Shakhbagova, Julietta (committee member)

Creator Email cahamlin@gmail.com,cahamlin@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-236161

Unique identifier UC11292725

Identifier etd-HamlinChri-1545.pdf (filename),usctheses-c3-236161 (legacy record id)

Legacy Identifier etd-HamlinChri-1545.pdf

Dmrecord 236161

Document Type Dissertation

Format application/pdf (imt)

Rights Hamlin, Christine A.

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 a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA