Literature on retention focuses on both short term and long term outcomes for retained students. Short term outcomes typically include academic achievement, self-esteem issues, and occurrence of behavior problems. Long term outcomes include academic growth and persistence to graduation. Regardless of the focus and intent of the study, research on retention is characterized by conflicting conclusions.

Overall, “the preponderance of evidence argues that students who repeat a grade are no better off, and are sometimes worse off, than if they had been promoted with their classmates” (David, 2008). Smith (2004) argued that “the evidence is remarkably one-sided: no short-term benefits, substantial long-term risks, substantial costs to taxpayers, and, because students of poverty and color are disproportionately represented among retainees, a failure of social justice” (p. 7). Burkam, LoGerfo, Ready, and Lee (2007) had similar conclusions in that most empirical research showed that retained student show short-term gains in achievement followed by a long-term fade in progress as well as poor emotional adjust and negative attitudes toward school.

Longitudinal research is a particularly important methodology for studying retention because it allows the researcher to understand how the effects of retention change or persist over time (Tanner and Galis, 1997). Longitudinal literature on retention focuses on a cohort of retained students and their long-term outcomes such as academic achievement, self-esteem, and dropping out of school. A cohort of retained students is typically matched by same grade or same age with promoted students. This analysis of longitudinal literature is organized to discuss those studies that show negative outcomes for students and studies showing positive student outcomes. This paper also highlights the interesting longitudinal work on retention that has come out of Chicago and Texas.

Peterson, DeGracie, and Ayabe (1987) conducted a three year study to determine retention’s effects on a group of 65 first grade students, 26 second grade students, and 15 third grade students. The authors examined more than 100 students in both retained and promoted groups and students were matched on sex, race, age, and scores on the California Achievement Test (CAT). The authors reported that separate analyses on all three grade levels showed that retained students performed higher in reading and math when compared with promoted students. “Although the advantage of first grade retention no longer existed in reading by third grade and mathematics by second grade, the benefits to second and third grade persisted over the next three years” (Lorence, 2006, p. 746). Lorence (2006) pointed out that one flaw in the research study was that initial raw CAT scores were not normed using the same reference group at all three grade levels.

In a study that seems to show a weak but positive impact of retention, Rust and Wallace (1993) conducted a longitudinal, same-age study to analyze the relationship between retention and further achievement scores by retained students in later grades. The authors divided 120 students, located in a large, southern metropolitan school district into two groups. One group was low achieving students who were retained and one group were low achieving students who were promoted. Students were matched on race, gender, free lunch status, and grade previous to the retention year. During the study, the retained students were in fifth grade and the promoted students were in sixth grade. They then collected data on the 120 students for the next four years. Data collected included not only demographic data but academic comparisons on both math and reading scores on the CAT and various forms of the Stanford Achievement Tests. Rust and Wallace concluded that their findings “indicated that retention lead to higher achievement during the repeated year but leveled out for the third and fourth years of the study” (p. 165). The authors also stated that the low achieving students who were promoted had success the following year. The promoted students’ academic achievement significantly increased the second year of the study. Overall, the authors concluded that the findings of the study were weak evidence that retention was positive during the first year. “In fact, during the year of retention both promoted and retained students did well even though the promoted students were working on the next grade assignments” (p. 166).

Similar to places such as Texas and Chicago, Florida’s third grade students must meet specific proficiency standards on the Florida Comprehensive Test (FCAT) in order to be promoted. Positive findings within this Florida data were reported by Greene and Winters (2007) and Greene and Winters (2004) who were interested in the impact of retention on student performance one and two years after the retention year. Their data set included Florida students in grades 3 through 10 for the school years 2001-2002 and 2004-2005. The authors compared low performing students who were promoted before the adoption of a state retention policy to low performing students who were retained after the adoption of the policy. Comparisons were calculated using a regression discontinuity by across-year, a comparison also done with Chicago Public School data by Jacob and Lefgren (2002). In their comparisons, Greene and Winters reported that after one year retained students outperformed promoted students by 0.05 standard deviations. After two years, the reading increase was 0.40 standard deviations. When discussing their regression discontinuity findings, Greene and Winters stated “that after two years, students who were retained outperformed promoted students by about 0.46 percent” (p. 336).

Much literature has been written concerning the state of Texas and its high stakes testing procedures. Literature concerning Texas data is mixed. Toenjes, Dworkin, Lorence, and Hill (2000) reported that to assess problem solving skills and critical thinking skills, the state of Texas implemented the Texas Assessment of Academic Skills (TAAS). Starting in 1993-1994 the reading and math test was given to students in grades three through eight and a writing test was given to students in grades four through eight. The authors reported that in 1994, 55.6% of students tested passed the TAAS and in 1999, 78.3% passed the test. This accountability system seemed to increase overall averages in reading scores by 6 points and math scores by 11 points. Examining the Houston Independent School district in detail revealed increased achievement similar to the overall state (Toenjes, Dworkin, Lorence, and Hill, 2000). However, Thomas and Stockton (2003) cited a study by Haney (2000) who credited retention as the cause for nearly 700,000 students dropping out school in an eight year period, 1992 to 1999.

Neblett (2007) reported similar positive increases in achievement for Texas students. The authors used a sample of retained third and fifth grade students to determine if retention increased reading scores one and two years after the retention year. The sample consisted of 33 third grade students and 49 fifth grade students. Neblett’s findings showed that both third grade and fifth grade students showed an improvement that was statistically significant in their reading and math scores the year after retention. This included Hispanic and African American students. Finally, findings also showed that students who were retained in the third grade scored above the proficiency standard for math in the fifth grade. A major weakness in the argument of the report was the small sample size.

Lorence, Dworkin, Toenjes, and Hill (2002) argued that their study was superior to previous research because it was large, contained all academically challenged students who were retained or promoted, and included data for six years. The authors used data provided by the Texas Education Agency (TEA) which provided information on 159,218 third grade students test scores on TAAS. The authors analyzed TAAS data starting in the 1994 school year and running through the 1999 school year. Two groups were compared in this data set, those with low reading scores who had been promoted and those who had low reading scores but had been retained. The authors described this approach as quasi-experimental. Data was analyzed in many ways with their conclusion being that requiring low-performing students to repeat a grade does not harm their academic performance. The findings support the position that holding low-performing students back an additional year in school provides them more time to obtain the skills necessary for later success in school (p. 38).

Specifically, the authors reported that the group of third grade students who had low reading scores on the TAAS in 1994 and were retained, increased their score about 18 points when they retook the TAAS reading assessment in 1995. This same group of students had a gain of nearly 17 points in the fourth grade. The authors also stated that this group of students passed the TAAS reading assessment in their fifth grade year.

Using the same Texas data as Lorence, Dworkin, Toenjes, and Hill (2002), Lorence and Dworkin (2006) followed third grade students who had failed the 1994 Texas reading test through the tenth grade and compared their reading scores with low-performing students who had been promoted in 1994. The authors compared “the average reading scores of those failing students required to repeat third grade (the experimental group) with that of the socially promoted students who also did not have a passing third grade reading score (the control group)” (p. 1005). The sample for the study consisted of 38,445 Texas public school students. Of this sample, 1,244 were retained at least one time. Lorence and Dworkin used a nonequivalent control group design to study the sample. The authors concluded that students who were retained in the third grade increased their reading performance when compared to similar at-risk students who were not retained but promoted. Summarizing their findings, the authors asserted that “comparing adjusted same-grade reading scores through six grades after retention revealed that socially promoted pupils lagged behind the reading ability of the retainees” (p. 1026-1027).

Lorence and Dworkin (2006) explained why their findings were different than most retention studies. The authors stated that their analysis only looked at retention in the third grade. The authors hypothesized the impact of retention may be grade specific, with retention more likely to beneficial for older students. They argued that retention at the third grade level, as opposed to the primary years, possibly helps to avoid any learning disabilities students might have because many learning disabled students have already been identified. In other words, these third grades students simply needed more time to meet the standard.

Thormalen (1998) conducted a study to determine how primary grade retention impacted Texas students’ academic achievement and self perception in adolescent years. The sample in the study consisted of 41 eighth grade students who were retained in first or second grade. The academic performance of these students on the TAAS and Total Reading and Total Math fifth grade Texas Learning Index (TLI) during their fifth and eighth grade years was analyzed. Surveys were also sent to the students and parents concerning their school experience. The author found that 76.7% of the retained group failed to meet the reading standard in the fifth grade. In the eighth grade, Thormalen reported that 57.5% of the eighth grade students failed to meet the reading standard. In regards to student perceptions, a greater number of students indicated that retention had helped them academically even if the data did not show this. Parents’ perceptions of retention were similar to the students, in that retention was a beneficial intervention. Thormalen conclusion was “that retention of this population of students in the primary grades had no observable impact on their academic performance either four years later or seven years later as measured by the mandated Texas Assessment of Basic Skills” (p. 122).

Similar to Texas, several studies have used Chicago Public School retention data as a basis for discussion on retention. Starting with the 1995-1996 eighth grade class, Chicago Public School students were held accountable by high-stakes testing. The students’ performance on Iowa Test of Basic Skills was the single factor that determined if they were promoted to high school or whether they were retained. In its first year of implementation, 1,800 eighth grade students were retained (Nagaoka, & Roderick, 2004). House (1998) predicted the Chicago high stakes testing policy, similar to New York City’s retention program implemented in the 1980’s, would fail to solve the problems of social promotion. Results from analysis of the Chicago data so far have been mixed.

Researchers with Consortium of School Research have analyzed the data from Chicago in a series of studies (Greene & Winters, 2004). Nagaoka and Roderick (2004) attempted to determine how successful the retained Chicago students were in raising their tests scores to the promotional standard during their retained year. Using a same age, matched methodology, Nagaoka and Roderick compared the achievement of both third grade and sixth grade students who scored below the promotion standard with similar students who scored above the promotion standard. The authors measured the improvements of both groups without accounting for retention. They then statistically adjusted whether a student was retained or promoted (Greene, & Winters, 2004). The authors reported that less than 60% of retained third and sixth graders in 1998 and 1999 were able to increase their scores to minimum standards. The authors also stated that approximately 20% of retained third and sixth graders were placed in special education classes within two years after the retention year. Overall, the authors concluded that there was little evidence that students who were retained in third grade did better than similar students who were socially promoted. The authors also found that sixth graders who were retained had lower achievement growth than low-achieving sixth graders who were promoted.

Allensworth (2004) reported on seven cohorts of Chicago students who were followed to determine whether a mandatory retention policy influenced their tendency to drop out of school and academic achievement. Results showed both positive and negative outcomes for the Chicago students. Cohorts were selected and compared before the implementation of the policy and after the implementation of the policy. The study consisted of 171,471 students. Allensworth concluded that the students who were retained at the eighth grade level had an 8% greater chance of dropping out of school by the time they turned 17 than those who were promoted. By the time these retained students reached 19, they had a 13% greater chance of dropping out when compared to other students in their cohort who had been promoted. In regards to student achievement data, Allensworth reported that student achievement “improved substantially with the post policy cohorts” (p. 11). By the 1998 cohort, student achievement had risen one-half of a standard deviation higher than pre-policy cohorts. It seemed to indicate that the policy had motivated students to perform academically for fear of retention. Allensworth warned that the degree to which the policy caused student achievement to rise was debatable and other factors in the Chicago Public Schools should be considered.

Jacob and Lefgren (2002) conducted a regression discontinuity design research format, similar to Greene and Winters (2007), to analyze data collected from the Chicago Public Schools data. This design controlled for the students’ previous test performance as well as social and demographic characteristics. The authors stated their population consisted of students who were in third and sixth grades from the 1993-1994 school year to the 1998-1999 school year. This included a sample of 346,909 student records. When the authors applied their regression discontinuity design to the data, they concluded that grade retention had a positive academic impact on math and reading at the third grade. Jacob and Lefgren concluded that summer school and grade retention had increased student achievement by 20% in students in the third grade. After the second year, the effect was not as large but was still statistically significant. Findings for sixth grade students were not significant for any year analyzed. Jacob and Lefgren concluded their findings by stating the evidence “suggests that summer school and grade retention have a modest but positive net impact on student achievement scores for third grade students” (p. 27).

Similar to their 2002 study, Jacob and Lefgren (2007) used the Chicago Public Schools data and attempted to determine if retention in the sixth or eighth grades increased the chances of students dropping out in high school and its academic outcomes for the students. The authors reported their final data sample consisted of 8,573 sixth grade students and 5,402 eighth grade students during the 1997, 1998, and 1999 school years. Students were followed to determine if they had graduated from the Chicago Public School by the fall of 2005. Jacob and Lefgren found that grade retention in the sixth grade had little effect on the probability of dropping out of school at a later date. Based on their findings, the authors concluded that it appeared “eighth grade retention is more harmful than sixth grade retention, possibly due to the greater social dislocation caused by preventing students from moving to high school with their peers” (p. 22).

Reynolds (1992) conducted a same-age, same-grade longitudinal study following retained first, second and third grades students in the Chicago Longitudinal Study (CLS). The final sample consisted of 1,255 student records. A majority of the retained students were African American and from a low socioeconomic background. Reynolds stated that data from four areas was analyzed: reading scores, math scores, teacher ratings of school performance, and self perceptions reported by the students themselves. The author stated that data was analyzed using a hierarchical regression analysis using two steps. After controlling for socioeconomic factors and school readiness, the author found that retained students scored significantly lower in reading and math than their promoted peers. Specifically, retained students scored eight months lower in reading achievement and six months lower in math achievement. Reynolds concluded by stating “The effect sizes on academic achievement in the present study…are over twice the size of the one-third of a standard deviation effect found in many previous studies” (p. 115).

In extension of the CLS data by McCoy and Reynolds (1998) found similar results using the CLS data to determine retention’s effects on school achievement, perceived school competence, and behavior occurrences up to 14 years old. The authors reported that their sample consisted of 1,164 low income children, which was 93% of the population sample for the Reynolds (1992) study. Retained students were tracked from first through seventh grades and data was collecting on reading achievement, math achievement, and behavior referrals. Data was analyzed using both same grade and same age comparisons (Malone, West, Flanagan, & Park, 2006). Same age results indicated that retained students scored 9.5 standard score points lower in reading and 8.9 standard score points lower in math. Same grade comparison showed that retained students scored significantly lower in reading but not in math. Finally, retention did not significantly impact behavior referrals.

The Reynolds (1992) and McCoy and Reynolds (1998) studies are considered strong retention research by Lorence (2006) because both studies statistically controlled for many variables of achievement previous to the retention year. “Most important is the availability of similar low-achieving pupils…which allowed for the creation of a control group more comparable to the retained pupils” (Lorence, 2006, p. 757).

The majority of educational research has come to the conclusion that retention is not a sound intervention and does more harm than good (Lorence, & Dworkin, 2006). For many, retention seems to help the struggling learner the year after the retention, but gains quickly drop off after several years. Research as early as the 1970's has reported that retention is harmful to students (Jackson, 1975). These negative student outcomes appear as lower academic achievement, poor self-esteem, and dropping out of school (Jimerson, & Kaufman, 2003; Anderson, Whipple, & Jimerson, 2002).

Others contend that much of this retention research is flawed leaving a legitimate argument that retention may be beneficial for the struggling student (Lorence, 2006). Retention’s positive effects on students’ cognitive development have also been reported in the several research articles (Powell, 1997; Southard, & May, 1996; Lorence, Dworkin, Toenjes, & Hill, 2002). While methodological issues plague much of this research there is also the question of how retention works differently for different types of students.

Gender, race, socioeconomic status, and parental involvement all play a role in the retention of students (Thomas, & Stockton, 2003). Retention occurs more often to minority males who are from a low socioeconomic background. These students are retained by educators for reasons such as low academic achievement, maturation of the student, and parental support (Witmer, Hoffman, & Nottis, 2004; Freeman, Gum, & Blackbourn, 1999). If retention is not to be used for low performing students, educators must find research based interventions such as multiage classrooms, looping, personal learning plans, and reduced class size that remediate struggling learners.

Allensworth, E. (2004). Ending social promotion: Dropout rates in Chicago after the implementation of the eighth-grade promotion gate. Chicago, IL: Consortium on Chicago School Research.

Anderson, G. E., Whipple, A. D, & Jimerson, S. R. (2002). Grade retention: Achievement and mental health outcomes. The California School Psychologist, 7, 51-62.

Bhattacharya, S. (2007). The effect of grade retention on child test scores. Retrieved from the World Wide Web February 4, 2008 from www.aeaweb.org/annual_mtg_papers/2008/2008_445.pdf

Burkam, D. T, LoGerfo, L., Ready, D., & Lee, V. E. (2007). The differentials effects of repeating kindergarten. Journal of Education for Student Placed At Risk, 12(2), 103-136.

Eamon, M. K., & Altshuler, S. J. (2004). Can we predict disruptive school behavior? Children & Schools, 26(1), 23-37.

Fine, J. G., & Davis, J. M. (2003). Grade retention and enrollment in post-secondary education. Journal of School Psychology, 41(6), 401-411.

Freeman, G., Gum, M., & Blackbourn, J. (1999). Proactive approaches to improving outcomes for at-risk students. (ERIC Document Reproduction Service No. 430948).

Greene, J. P., & Winters, M. A. (2004). An evaluation of Florida’s program to end social promotion. New York: Manhattan Institute for Policy Research.

Greene, J. P., & Winters, M. A. (2007). Revisiting grade retention: An evaluation of Florida’s test-based promotion policy. Education Finance and Policy, 2(4), 319-340.

Haney, W. (2000). The myth of the Texas miracle in education. Education Policy Analysis Archives, 8(41). Retrieved from the World Wide Web July 10, 2008 from http://epaa.asu.edu/epaa/v8n41/

Hong, G., & Yu, B. (2007). Early-grade retention and children’s reading and math learning in elementary years. Educational Evaluation and Policy Analysis, 29(4), 239—261.

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.

House, E. R. (1998, November). The predictable failure of Chicago’s student retention program. Paper presented at the Rethinking Retention to Help All Students Succeed conference, Chicago, IL.

Jacob, B. A., & Lefgren, L. (2007). The effect of grade retention on high school completion. Retrieved from the World Wide Web February 4, 2008 from http://sitemaker.umich.edu/bajacob/files/ret_nber_10_2_2007.pdf

Jacob, B. A., & Lefgren, L. (2002). Remedial education and student achievement: A regression-discontinuity analysis. Retrieved from the World Wide Web February 4, 2008 from http://sitemaker.umich.edu/bajacob/files/restat_2003.pdf

Jackson, G. B. (1975). The research evidence on the effects of grade retention. Review of Educational Research, 45(4), 613-635.

Jimerson, S. R. (1999). On the failure of failure: Examining the association between early grade retention and education and employment outcomes during lat adolescence. Journal of School Psychology, 37(3), 243-272.

Knoll, N. A. (2003). The impact of elementary retention on school related variables: A longitudinal study. (Doctoral dissertation, State University of New York at Albany, 2003). Dissertation Abstracts International, 64, 1970.

Lange, R. (2004, February). The truth about grade level retention and social promotion: How state and national policies are destroying the future of American youth. Paper presented at the Truth in Testing Conference, Orlando, FL.

Lorence, J. (2006). Retention and academic achievement research revisited from a United States perspective. International Educational Journal, 7(5), 731-777.

Lorence, J., & Dworkin, A. G. (2006). Elementary grade retention in Texas and reading achievement among racial groups: 1994—2002. Review of Policy Research, 23(5), 999—1033.

Lorence, J., Dworkin, A. G., Toenjes, L., & Hill, A. (2002). Grade retention and social promotion in Texas, 1994-99: Academic achievement among elementary school students. In Diane Ravitch (Ed.), Brookings Papers onEducation Policy (pp. 13-67). Washington, DC: Brookings Institution Press.

Malone, L. M, West, J., Flanagan, K. D., & Park, J. (2006, May). The early reading and mathematics achievement of children who repeated kindergarten or who began school a year late. Retrieved from the World Wide Web July 9. 2008, from http://nces.ed.gov/pubs2006/2006064.pdf

McCoy, A. R., & Reynolds, A. J. (1998). Grade retention and school performance: An extended investigation. Retrieved from the World Wide Web on July 9, 2008 from http://www.ssc.wisc.edu/irpweb/publications/dps/pdfs/dp116798.pdf

Meisels, S. J., & Liaw, F. R. (1993). Failure in grade: Do retained students catch up? Journal of Educational Research, 87(2), 69-77.

Nagaoka, J., & Roderick, M. (2004, March). Ending social promotion: Dropouts rates in Chicago after implementation of the eighth-grade promotion gate. Retrieved from the World Wide Web June 28, 2008 from http://ccsr.uchicago.edu/publications/p69.pdf

Neblett, P. S. (2007). An analysis of the benefits of the student success initiative in the 3rd and 5th grades in a district in Texas. (Doctoral Dissertation, University of North Texas, 2007). Retrieved from the World Wide Web July 10, 2008 from http://digital.library.unt.edu/permalink/meta-dc-3634:1

Peterson, S. E., DeGracie, J. S., & Ayabe, C. R. (1987). A longitudinal study of the effects of retention/promotion on academic achievement. American Educational Research Journal, 24(1), 107—118.

Reynolds, A. J. (1992). Grade retention and school adjustment: An explanatory analysis. Educational Evaluation and Policy Analysis, 14(2), 101-121.

Rust, J., & Wallace, K. (1993). Effects of retention for four years. Journal of Instructional Psychology, 20, 162-167.

Setencich, J. (1994). The impact of early grade retention on the academic achievement and self-esteem of seventh and eighth grade students. Washington DC: Office of Educational Research and Improvement. (ERIC Document Reproduction Service No. ED393026).

Silberglitt, B., Jimerson, S. R., Burns, M. K., & Appleton, J. J. (2006). Does the timing of grade retention make a difference? Examining the effects of early versus later retention. School Psychology Review, 35(1), 134-141.

Tanner, K. C, & Galis, S. A. (1997). Student retention: Why is there a gap between the majority of research findings and school practice? Psychology in the Schools, 34(2), 107-114.

Thomas, J., & Stockton, C. (2003). Socioeconomic status, race, gender, and retention: Impact on student achievement. Retrieved from the World Wide Web July 10, 2008 from http://www.usca.edu/essays/vol72003/stockton.pdf

Thormalen, K. C. (1998). The impact of primary grade retention on the academic performance and self perceptions of middle school students in an urban/suburban school district. (Doctoral dissertation, Texas A&M University, 1998). Dissertations Abstracts International, 59, 2809.

Toenjes, L. A., Dworkin, A. G., Lorence, J., & Hill, A. N. (2000). The lone star gamble: High stakes testing, accountability, and student achievement in Texas and Houston. Retrieved from the World Wide Web on July 10, 2008 from http://66.102.1.104/scholar?hl=en&lr=&q=cache:Nib7TYVVJT4J:www.brookings.edu/gs/brown/bc_report/2000/Houston.PDF

Wiley, R. F. (2006). Exploring the impact of grade retention on students’ aspirations’ and educational outcomes. (Doctoral Dissertation, Morgan State University, 2006). Dissertation Abstracts International, 67, 2006.

Witmer, S. M., Hoffman, L. M., & Nottis, K. E., (2004). Elementary teachers’ beliefs and knowledge about grade retention: How do we know what they know? Education, 125(2), 173-194.