1703 North Beauregard St.
Alexandria, VA 22311-1714
Tel: 1-800-933-ASCD (2723)
8:00 a.m. to 6:00 p.m. eastern time, Monday through Friday
Local to the D.C. area: 1-703-578-9600, press 2
Toll-free from U.S. and Canada: 1-800-933-ASCD (2723), press 2
All other countries: (International Access Code) + 1-703-578-9600, press 2
November 28, 2005 | Volume 3 | Number 18
The Effect of Retaining Kindergarten Students on Reading and Mathematics Achievement
Does retaining low-performing students in kindergarten lead to improved achievement?
Despite a wealth of research on the effects of retention and social promotion on students’ personal and academic development, a significant degree of uncertainty remains regarding the use of these strategies. Prior research suggests that neither strategy adequately addresses the academic needs of low-achieving students, although studies do suggest possible benefits to social promotion (such as lower dropout rates for promoted students). In addition, some researchers—including the authors of the study highlighted in this issue of ResearchBrief—have hypothesized that retaining low-achieving students increases the academic homogeneity of both the promoted and retained classes, making instruction easier and more effective. Although retention policies in general remain a source of debate, retention in kindergarten has long been viewed in a somewhat different light, largely because discussion of student promotion at the kindergarten level focuses more on the developmental readiness of the learner than on using retention as a behaviorist motivation mechanism. This issue of ResearchBrief looks at promotion and retention and the effect these strategies have on kindergarten students.
Guanglei Hong and Stephen Raudenbush conducted the study highlighted in this issue of ResearchBrief (see below for full citation). The authors examined three questions related to retention policies:
In attempting to answer these questions, the authors looked at data from the Early Childhood Longitudinal Study Kindergarten cohort (ECLS-K). The full data set consists of more than 20,000 students, but this study looked at 11,843 children for whom appropriate test score data and school retention information was available. The group consisted of 471 retainees and 10,255 promoted students in 1,080 schools with a retention policy and 1,117 promoted students in 141 schools without such policies. The authors compared the students in the sample group with the full population and found that the sample contained fewer poor and minority children as well as fewer children from non-English-speaking homes.
Most of the students involved in ECLS-K (and making up the data set in this study) were observed three times—in the fall and spring of 1998–99 and again in the spring of 2000. A random sample of students was also observed in the fall of 1999, and all students were tested up to four times over the course of the study.
The authors structured their data analysis so as to create the equivalent of an experimental model. They defined two treatment conditions: (1) the presence of a retention policy and (2) the application of retention (or promotion) on individual students. Because the schools were not randomly assigned into treatment groups, the authors attempted to balance the groups based on 238 pretreatment covariates (relevant school-level characteristics).
To identify the effect of retention on students who were promoted, the authors looked at the characteristics of students promoted in retention schools and identified 207 covariates that enabled them to specify a group of students whose chance of being retained was essentially zero. They then used these covariates to identify similar students in the nonretention schools, allowing an examination of the effect of retaining students on the students who were promoted. In other words, if these “unretainable” students in schools with retention policies did better than their peers in the nonretention schools, then it would be concluded that retention was likely the cause for their success.
On average, schools with retention policies had significantly better learning conditions than schools without such policies. Retention schools had smaller class sizes, greater parental involvement, more order in classrooms, and were more likely to be private, located in the suburbs, and have low minority enrollments. In creating their comparison groups, the authors noted that retained students were more likely to be retained if, among other factors, they were non-Hispanic, male, from a low-SES background, from a single parent-multiple sibling family, had never received center-based child care, had a learning disability, and entered kindergarten at a young age.
The authors also found a number of classroom and school-based predictors for retention. The chance of retention was greater for students in classrooms with, among other characteristics, a high proportion of boys, 3–
4 year olds, retained students, students with disabilities, or a low number of students recognizing letters at the start of the year. Teachers in such classes tended to teach lower-level skills, have less experience teaching at the grade level, and lack an advanced degree. At the school level, students more likely to be retained attended schools that were smaller, were nonpublic, lacked adequate instructional resources, had lower paid teachers, had lower attendance, had no services for disabled students, used increased security measures, and had a lower safety rating. Principals of these schools had less training in administration and placed more emphasis on student behavior than academics.
When the researchers analyzed their data, they found that students in retention schools generally scored higher on reading and mathematics; once controls for prior achievement and other pretreatment advantages were applied, however, the differences disappeared. Ultimately, the estimated effect of retention policies on student achievement was -0.24 for reading, with a standard error of 0.86; and -0.14 for mathematics, with a standard error of 0.55. When the researchers analyzed the effect of retention on promoted students (did retention of low-achieving students result in higher achievement for their promoted peers), no significant effects were found. Within the retention schools, retained students generally scored significantly lower on both math and literacy assessments than their at-risk peers who were promoted. The authors estimated that the likely effect not being promoted would have had on the promoted students was similarly negative.
Retention policies have significant negative effects on retained students and little or no significant effects on their promoted peers. Estimates suggest that promoted students would show lower growth if they had been retained, whereas retained students would experience higher growth if promoted.
This research focused on children in kindergarten and 1st grade.
Although no detectable difference in the scores between retention and nonretention schools could be found, the authors suggest that this is because only a very small percentage of students are held back, obscuring the negative effects of retention on the schoolwide averages. Although the authors noted that their data sample included fewer poor and minority children, as well as fewer children from non-English-speaking backgrounds, than the full ECLS-K data set, little consideration was given to what, if any, implications this might have for the study findings. As highlighted by the authors, the long-term implications of retention in kindergarten have yet to be explored using ECLS-K data and may provide useful information as the cohort ages.
Hong, G., & Raudenbush, S. (2005). Effects of kindergarten retention policy on children’s cognitive growth in reading and mathematics. Educational Evaluation and Policy Analysis 27(3), 205–224
Retention, Social Promotion, and Student Outcomes: A Meta-Analysis of Recent Research, Research Brief 3(1)
Retention and Student Achievement,
School Dropouts: Home and School Effects,
Early Childhood Longitudinal Study.
National Center for Education Statistics
____________All comments regarding ReseachBrief should be sent to RBfeedback@ascd.org. To speak directly with an ASCD staff member, please Contact Us.
Dan Laitsch serves as ASCD's consultant editor for ResearchBrief. Laitsch is an assistant professor in the Faculty of Education at Simon Fraser University in British Columbia, Canada, and is coeditor of the International Journal for Education Policy and Leadership.
Copyright © 2005 by Association for Supervision and Curriculum Development
Subscribe to ASCD Express, our twice-monthly e-mail newsletter, to have practical, actionable strategies and information delivered to your e-mail inbox twice a month.
ASCD respects intellectual property rights and adheres to the laws governing them. Learn more about our permissions policy and submit your request online.