Publications>
Research Bulletin

Phi Delta Kappa Center for Evaluation, Development, and Research
December 1995, No. 15

Grade Retention and School Dropout:
Policy Debate and Research Questions
By Melissa Roderick

Over 30% of 14-year-olds were enrolled in a grade below ninth grade, their modal grade level, in 1992. Grade retention rates and, as a result, the proportion of students overage for grade by the time they reach high school have risen nearly 40% over the past two decades. Many teachers believe that retention, particularly in the early grades, is an effective strategy to remediate poor school performance and may reduce the likelihood of later school failure. Research on grade retention, however, concludes that repeating a grade provides few remediational benefits and may, in the long run, place students at a higher risk of dropping out of school. Should school systems be alarmed by these increases in retention rates? And what, if anything, would concern over retention mean for policy? This research bulletin looks at the evidence regarding whether and how grade retention may influence the chances that a student would drop out in the context of the policy trends that have contributed to rising rates of grade retention over the past two decades.

RETENTION RATES AND SCHOOL POLICY

There is no precise national estimate of the proportion of youths who experience grade retention. A rough estimate of the incidence of grade retention can be obtained by examining the percentage of a cohort enrolled below its modal grade level in a given year.(1) In 1984, for example, the Current Population Survey estimated that 11.1% of six-year-olds were enrolled below their modal grade level, first grade. Eight years later, at 14 years of age, 31.4% of this cohort was enrolled below its modal grade level, ninth grade. Thus, in 1992, approximately 20% of 14-year-olds may have experienced a grade retention between first and eighth grades. Most of these retentions occur early in students' school careers. Figure 1 shows the proportion of students who were enrolled below their modal grade level by age in 1992. These cross-sectional rates demonstrate that the largest increase across cohorts in the proportion of students enrolled below their modal grade level occurs between six and nine years of age, or between the modal grade levels of first to fourth grade.

Retention rates and the proportion of students who are overage for grade vary significantly by race and gender. Table 1 shows the proportion of the 1984 cohort of sixth-graders who were overage for grade at ages 6, 9, and 14. In 1992, for example, almost 40% of all 14-year-old males were overage for grade compared to 20% of all females. Over one-half of black 14-year-old males and fully 48.5% of Hispanic males were enrolled below ninth grade.

The proportion of youths promoted from one year to the next is largely determined by schools systems' promotion policies and by teachers' and principals' attitudes regarding the benefits of retention. Two developments have contributed to increases in rates of non-promotion in the past two decades. First, during the 1980s many school systems adopted strict promotion policies, often tied to scores on curriculum-referenced or basic-skills tests. Second, kindergarten and preschool enrollments have increased steadily, bringing with them rising academic demands in kindergarten and first grade and increasing emphasis on "academic" as well as developmental readiness in making promotional decisions in the early years.(2)

Over the past three decades, grade retention policies have fluctuated from one extreme to another. During the 1970s, the prevailing philosophy was that social promotion was most beneficial to youths.(3) Policies of social promotion came under sharp criticism during the standards-raising movement of the 1980s, catalyzed by the publication in 1983 of A Nation At Risk. This report pointed to declines in student achievement test scores as evidence that such lenient policies as social promotion had caused a dilution of standards and a decline in the quality of American education. In response, many school systems drafted strict promotion policies, often tied to scores on curriculum-referenced or basic-skills tests.

The most dramatic indicator of the shift in attitudes and practices regarding promotion is the increase in the proportion of youths who are overage for grade. From 1970 to 1980, the proportion of 12- to 14-year-olds enrolled below their modal grade level hovered around 20%. From 1980 to 1993, however, this proportion increased significantly, peaking in the early 1990s at nearly 32% (see Table 2).

The 1990s witnessed yet another pendulum swing in educator's attitudes toward retention. Many school systems began to review their retention policies and search for alternatives. In 1990, for example, New York City's school chancellor revised the system's strict promotion policy, citing evidence that dropout rates among retained youths were higher than dropout rates among promoted youths with comparable reading levels, even though retained youths had received special services.(4) In Chicago, research showed that overage students were more likely to drop out, even when controlling for prior school achievement, and this condition led to a revocation of basic-skills testing as a criterion for retention.(5) The Chicago School Reform Law, passed in 1989, set a central goal of education reform reducing retention rates by 10% over four years.(6)

GRADE RETENTION AND SCHOOL DROPOUT

As evinced by the experiences of Chicago and New York City, opponents of grade retention often cited the strong association between retention and dropping out as evidence that nonpromotion is harmful. Indeed, the strength of the association is impressive. A widely quoted finding from the Youth in Transition Study is that one grade retention increases the risk of dropping out by 40% to 50%, and being two grades behind increases the risk by 90%.(7) In the more recent High School and Beyond Survey, sophomores who reported that they had repeated at least one previous grade dropped out at more than twice the rate of youths who reported that they had never repeated a grade.(8) Using data from individual school systems, several studies have shown that students who are retained or who are overage for grade drop out at significantly higher rates, even when controlling for prior achievement or grades and attendance.(9) In addition, students who are retained drop out at higher rates, regardless of whether retention occurs early or later in their school careers.

The literature on grade retention suggests three important aspects of the retention experience that combine to place students at risk of school failure and early school leaving. First, grade retention as a remediational strategy does not appear to fix the problem it purports to address and may, at least when it occurs in higher grades, exacerbate poor school performance. Second, grade retention is perhaps the strongest message that a teacher and a school can send to a student that she or he is not making the grade and is not as capable as other children, a failure that is permanent and cannot be remediated by extra effort. The permanency of retention and the message it sends students may have long-term effects on self-esteem and school attachment that may override even short-term academic benefits. And third, grade retention, regardless of when it occurs, may increase the chances of school leaving because it makes a student overage for grade during adolescence, and, for those who are already having difficulty in school, it may increase the likelihood that they will feel frustrated and become disengaged.

IMPACT OF RETENTION ON SCHOOL PERFORMANCE AND ACHIEVEMENT

A large body of research has examined the impact of grade retention on measured achievement.(10) In most of this research, the effect of retention is estimated by comparing the achievement test scores of retained students to a matched group of promoted youths. Using this method, estimates differ depending on whether students' scores are compared when they are the same age or after they have completed the same grade. Studies that use same-age comparisons generally find that promoted students perform better than retained students in the year after retention and that the academic performance of retained pupils continues to lag behind that of promoted youths in later years.(11) For example, C. Thomas Holmes's most recent meta-analysis indicates that retained students score on average .45 standard deviations lower on tests of achievement than promoted students in the year after retention occurs.(12) Studies comparing retained and promoted students after they complete the same grade and same material (same-grade comparisons) do not show large negative effects on achievement and often find small positive ones. These positive effects, however, appear to be short term, lasting no more than two to three years.(13) Research findings also differ depending on the grade in which students are retained. In general, early grade retentions are associated with moderate positive benefits for school performance, although such effects are short term.(14) Studies that compare the performance of retained to promoted students in later grades tend to find large negative effects of grade retention.(15)

In sum, results of studies investigating the effects of retention on academic performance generally indicate that retention as a means of remediation does not work. At best, it leaves students who were already lagging behind their peers even further behind. At worst, retention has negative effects on measured achievement, particularly in later grades. If grade retention increases the likelihood that a student will drop out because it leads to lower school performance, then reductions in the number of students retained in grade should lead to better performance and lower dropout rates. Researchers at the Consortium for Chicago School Research are just beginning to look at this possibility in a study of achievement trends and school reform in the Chicago Public Schools.(16) They estimated the effect of declining grade retention rates on achievement gains among cohorts of students. From 1988 to 1992, retention rates fell by one-half, mostly in response to school system efforts. By comparing the achievement growth of students retained under pre-reform retention rates to that of students in later cohorts retained under the old policy but now promoted, these researchers estimated that, across grade levels, a decline in the use of retention was associated with greater learning gains. Whether these benefits will translate into lower dropout rates has yet to be determined.

GRADE RETENTION AND ATTITUDES TOWARD SCHOOL

While much is known about the effect of grade retention on measured achievement, studies about how grade retention influences students' attitudes toward themselves and school are scant and results are mixed. Qualitative studies most often find that students react negatively to retention. Younger children often perceive retention as a form of punishment and a stigma.(17) Qualitative studies of adolescents often conclude that retention exacerbates disengagement from school and leads to increases in frustration at doing poorly.(18)

Quantitative studies examining the effects of retention on self-esteem and attitudes toward school have found substantial negative effects when retained students are compared to matched promoted youth. In their meta-analysis, C. Thomas Holmes and Kenneth Matthews conclude that across studies retention has negative but relatively moderate effects on measured self-esteem and school attachment.(19) In a recent study conducted in a school system that served predominantly African-American students, Denise Gottfredson and her colleagues compared school attachment and attitudes of retained sixth- and seventh-grade students to a matched sample of promoted students: retention was not associated with negative effects on self-esteem, peer associations, attitudes toward school, or school attachment and behavior.(20) Indeed, the authors found that retained students showed greater attachment to school and reported less negative school behaviors than promoted students. The authors concluded that among high-risk students retention may have led to a reduction in negative behavior, because it placed these youths among younger students, where they may have enjoyed higher status, perhaps delaying later problems.

Gottfresdon's findings as well as the general direction of results in grade retention research suggest that a critical ingredient in understanding retention effects lies in identifying and disentangling the interaction of grade retention with age and a student's developmental stage. In a study of teachers' attitudes about grade retention conducted by Ellen Tompchin and James Impara, teachers often believed that early grade retentions (kindergarten through third grade) give immature students a chance to catch up and have few negative impacts on self-esteem.(21) Teachers were much less sanguine about the effects of retention in later grades (fourth through seventh grade). While only 17% of the teachers surveyed agreed that early grade retention harms children's self-concept, over 60% agreed that later grade retention would have a negative effect, and 44% agreed that children should not be retained later than third grade. Research findings, however, suggest that teachers overestimate the potential benefits of early grade retention. At the same time, there is general agreement that as a student gets older, relative status among his or her peers becomes increasingly important in shaping self-concept and, ultimately, school engagement and performance.(22) Indeed, fully 74% of the teachers surveyed by Tompchin and Impara agreed that overage students cause more behavior problems in later grades.

In most of the literature the terms "overage for grade" and "grade retention" are used interchangeably. It is unclear, however, to what extent teachers' beliefs about the negative effects of grade retention on older students are influenced by their observations that students overage for grade cause and have more problems. Distinguishing the effect of being overage for grade from the effect of grade retention has important policy implications. If negative effects occur primarily when students are retained later in their school careers, then school systems should pursue alternatives to retention in the middle grades. In this case, there would be less basis for alarm about high retention rates in early grades, since early retention would appear to have relatively neutral effects. If, on the other hand, being overage for grade, regardless of when a student is retained, negatively impacts school performance and attachment during adolescence, then early grade retention would increase the risk of school dropout, even if it had short-term positive benefits earlier in a student's school career.

When you make a student overage for grade during adolescence, he or she may feel self-conscious, particularly as his or her age cohort moves to high school. The impact of being overage for grade was described by Mario, a student in my current study of the transition to high school in Chicago. Despite reporting that his eighth-grade grades were better than he had expected, Mario's frustration at being overage for grade largely colored his attitudes toward his elementary school. When asked if he would miss the elementary school he had attended for five years and in which he was retained in the fifth grade, Mario responded:

Nah, too many years. I was the longest student there, me and some other guys. Had like five years already. Teachers came and past and other principals were there and we're still here. It was embarrassing. All my cousins and brothers had passed to high school already and I was still there. And like I was supposed to be a sophomore this year and so they go "man you're still here." Everybody tell me like "what year of high school are you in" and I say . . . and they say, "Eeeh, you been there a long long time."

The results of my own research on the effect of grade retention on school dropout lends some evidence to the hypothesis that there is an effect of being overage for grade that influences students' school performance during adolescence. Students can be overage for grade for a number of reasons -- because they were retained, because they entered the school system overage for grade, or because they immigrated to the school system and were placed in a grade below their modal grade level. In an exploratory analysis of school transcript data among one cohort of students from an urban Massachusetts public school system, I found that students who were overage for grade experienced substantial disengagement from school during the late middle-school years.(23) Students who began sixth grade overage for grade were substantially more likely to drop out of school during middle school (sixth to eighth grade). Even those overage students who went on to high school were beginning to shows signs of withdrawal. By the end of middle school, those who ended the sixth grade overage for grade and who had not dropped out were absent an average of seven days more than those enrolled at their modal grade level, even when accounting for differences in grade and attendance just two grades earlier. These students continued to drop out at higher rates. In fact, the effect of being overage for grade may explain a large proportion of the higher dropout rates among retained students.

CRITICAL POLICY ISSUES

Reviews of grade retention research that compare teachers' beliefs about the benefits of retention to the research evidence are numerous.(24) Many of these reviews conclude similarly. Teachers promote retention because they believe it is an effective remediational strategy. To the extent that retention, at least in the early grades, shows some short-term benefits for students, teachers' conclusions are understandable. Teachers in early grades are not able to follow their students over time to see the effects of their decisions on students' later development. While research on the relationship between retention and school dropout is not conclusive, nor have the processes and causal links that fully explain high dropout rates been identified, there is enough evidence to conclude that grade retention provides few long-term academic or developmental benefits and that it places students at risk of dropping out. The conclusion of most reviews, then, is to take a strong stand against retention and to advocate for the education of teachers.

The dramatic rise in retention rates over the past two decades has not been generated by teacher attitudes alone, however. At the beginning of this bulletin, I cited two developments that have contributed to the rise in retention rates over the past two decades: changes in school systems' policies and changes in the nature of early grades education. Policy makers need to be aware of the extent to which district-wide policies influence individual school's practices. While many school systems have moved away from making retention decisions based on scores on standardized tests, the increasing emphasis on using test scores to hold schools and teachers accountable for performance creates disincentives for teachers to allow heterogeneity in the classroom and students to develop at their own pace, even as early as first grade.(25) There is also little evidence that the trend toward increasing academic demands in kindergarten will abate, despite calls from early childhood educators for more age-appropriate practices that allow for diversity in students' development.(26) For minority and lower income students, who are less likely to be enrolled in preschool and have parents who emphasize academic skills development before kindergarten, increasing academic demands in the early grades will most likely translate into continued high retention rates in the absence of policy attention.(27)

ENDNOTES

1. See, for example, U.S. Department of Commerce, Bureau of the Census, School Enrollment: Social and Economic Characteristics of Students: October 1984, Current Population Reports, Series P-20, no. 404, Washington, DC: GPO, 1985.

2. Lorrie A. Shepard and Mary Lee Smith, "Escalating Academic Demand in Kindergarten: Counterproductive Policies," The Elementary School Journal 89, no. 2 (1988): 135-145; Jerry West, Elvie Germino Hausken, and Mary Collins, "Readiness for Kindergarten: Parent and Teacher Beliefs," Statistics in Brief (Washington, DC: U.S. Department of Education, National Center for Education Statistics, 1993).

3. Janet S. Rose et al., "A Fresh Look at the Retention-Promotion Controversy," Journal of School Psychology 21 (1983): 201-211.

4. New York City Board of Education, Office of Educational Assessment, A Follow Up Study of the 1982-1983 Promotional Gates Students (New York, 1988).

5. John Q. Easton and Sandra Storey, June 1989 Grade Retention in Chicago's Public Elementary Schools, Chicago Panel on Public School Policy and Finance (Chicago, 1990).

6. Ibid.; Stuart Luppescu et al., "School Reform, Retention Policy, and Student Achievement Gains" (paper presented at the annual meeting of the American Educational Research Association, San Francisco, April 1995).

7. Jerald G. Bachman, Swayzer Green, and Ilona D. Wirtanen, Dropping Out -- Problem or Symptom? vol. 3, Youth in Transition (Ann Arbor, MI: Institute for Social Research, University of Michigan, 1971).

8. Stephen M. Barro and Andrew J. Kolstad, Who Drops Out of High School?: Findings from High School and Beyond, no. CS 87-397 c (Washington DC: U.S. Department of Education, National Center for Education Statistics, 1987).

9. James B. Grissom and Lorrie A. Shepard, "Repeating and Dropping Out of School," in Flunking Grades: Research and Policies on Retention, ed. Lorrie Shepard and Mary Lee Smith (London: The Falmer Press, 1989), 34-63; Melissa Roderick, "Grade Retention and School Dropout: Investigating the Association," American Educational Research Journal 31 (1994): 729-759.

10. Roy P. Doyle, "The Resistance of Conventional Wisdom to Research Evidence: The Case of Retention in Grade," Phi Delta Kappan 71 (November 1989): 215-220; Gregg B. Jackson, "The Research Evidence of the Effects of Grade Retention," Review of Educational Research 45 (1975): 613-635; C. Thomas Holmes, "The Fourth R: Retention," Journal of Research and Development in Education 17 (1983):1-6; C. Thomas Holmes, "Grade Level Retention Effects: A Meta-Analysis of Research Studies," in Flunking Grades: Research and Policies on Retention, ed. Lorrie A. Shepard and Mary Lee Smith (London, The Falmer Press, 1989), 16-33; C. Thomas Holmes and Kenneth M. Matthews, "The Effects of Nonpromotion on Elementary and Junior High School Pupils: A Meta-Analysis," Review of Educational Research 54 (1984): 225-236; Nancy L. Karweit, Repeating a Grade: Time to Grow or Denial of Opportunity, no. 16 (Baltimore, MD: Center for Research on Effective Schooling for Disadvantaged Students, Johns Hopkins University, 1991); Lucille Niklason, "Non-Promotion: A Pseudoscientific Solution," Psychology in the Schools 21 (1984): 485-499; Monica Overman, "Student Promotion and Retention," Phi Delta Kappan 67 (April 1986): 609-613; Rose et al., op. cit.; Lorrie A. Shepard and Mary Lee Smith, "Synthesis of Research on School Readiness and Kindergarten Retention," Educational Leadership 44 (1986): 78-86; Mary Lee Smith and Lorrie A. Shepard, "What Doesn't Work: Explaining Policies of Retention in the Early Grades," Phi Delta Kappan 69 (October 1987): 129-134.

11. Holmes, "The Fourth R"; Holmes, "Grade Level Retention Effects"; Holmes and Matthews, "The Effects of Nonpromotion on Elementary and Junior High School Pupils."

12. Holmes, "Grade Level Retention Effects."

13. Holmes, "Grade Level Retention Effects"; Panayota Mantzicopoulos and Delmont Morrison, "Kindergarten Retention: Academic and Behavioral Outcomes through the End of Second Grade," American Educational Research Journal 29 (1992): 182-198.

14. Holmes, "Grade Level Retention Effects"; Mantzicopoulos and Morrison, "Kindergarten Retention."

15. Holmes, "Grade Level Retention Effects."

16. Luppescu et al., "School Reform, Retention Policy, and Student Achievement Gains."

17. Deborah A. Byrnes, "Attitudes of Students, Parents, and Educators Toward Repeating a Grade," in Flunking Grades: Research and Policies on Retention, ed. Lorrie A. Shepard and Mary Lee Smith (London: The Falmer Press, 1989).

18. Judy Roberts Campbell, "Failing Students: A Qualitative Study on the Grade Retention Experience for Seventh Graders," in Phi Delta Kappa 1993 Outstanding Doctoral Dissertation Awards (Bloomington, IN: Phi Delta Kappa, 1993); Michelle Fine, Framing Dropouts: Notes on the Politics of an Urban Public School (Albany, NY: State University of New York Press, 1991).

19. Holmes and Matthews, "The Effects of Nonpromotion on Elementary and Junior High School Pupils."

20. Denise C. Gottfredson, Carolyn M. Fink, and Nanette Graham, "Grade Retention and Problem Behavior," American Educational Research Journal 31 (1994): 761-784.

21. Ellen M. Tompchin and James C. Impara, "Unraveling Teachers' Beliefs About Grade Retention," American Educational Research Journal 29 (1992): 199-223.

22. Susan Harter, "Self and Identity Development," in At the Threshold: Developing Adolescent, ed. S. Shirley Feldman and Glen R. Elliott (Cambridge, MA: Harvard University Press, 1990), 352-387.

23. Roderick, "Grade Retention and School Dropout."

24. Doyle, "The Resistance of Conventional Wisdom to Research Evidence"; Holmes,"The Fourth R"; Holmes, "Grade Level Retention Effects"; Holmes and Matthews, "The Effects of Nonpromotion on Elementary and Junior High School Pupils"; Karweit, "Repeating a Grade"; Overman, "Student Promotion and Retention"; Shepard and Smith, "Synthesis of Research on School Readiness and Kindergarten Retention."

25. Shepard and Smith, "Escalating Academic Demand in Kindergarten."

26. West, Hausken, and Collins, "Readiness for Kindergarten: Parent and Teacher Beliefs."

27. Condition of Education: Overview (U.S. Department of Education, Washington, DC: GPO, 1995); West, Hausken, and Collins, "Readiness for Kindergarten."