• by Pam McCollum, Ph.D. • IDRA Newsletter • November – December 1998 • Pam McCollum

A review of education columns in newspapers across the country shows that politicians and education policy-makers are taking a stance against “social promotion” in the public schools. In his 1998 State of the Union Address, President Clinton linked ending the practice of social promotion to improving schools. The centerpiece of his education policy was a proposed voluntary national test based on national standards in fourth grade reading and eighth grade math. The president’s plan would retain students in-grade who do not meet those standards and it would give them the tools they need to learn.

The intuitive appeal of holding students back who have not mastered grade-level knowledge and skills is so strong and its history is so long that its efficacy is rarely questioned, even though research overwhelmingly shows that retention has negative personal and academic effects.

During a 30-year period, the educational pendulum has alternated between advocating social promotion and supporting in-grade retention. Social promotion refers to the practice of passing students who have failed to master part or all of the grade-level curriculum on to the next grade with their cohort of age-grade peers. In-grade retention, on the other hand, requires students to repeat the same grade a second time in order to master problem material.

The pendulum has changed directions by decade. For example, in the 1970s, social promotion was favored, but with the call for raising educational standards in the 1980s and its attendant minimal competence testing, the favor returned to retention. By 1990, however, two of the largest school districts in the country, Chicago and New York City, were advocating promoting students with their age-appropriate cohort.

As we approach the year 2000, the pendulum clearly indicates in-grade retention as the favored response to ameliorating poor academic achievement. Across the country, cities such as Philadelphia, Long Beach, Milwaukee, Detroit, Boston, Oakland (Calif.), Springfield (Mass.) and Corpus Christi (Texas) have already abolished or given notice that social promotion will be abolished by the year 2000.

The political nature of joining the “end social promotion” bandwagon cannot be ignored. The governors of California, Delaware, Michigan, Wisconsin and Texas have all spoken out against social promotion, while others are sure to follow suit. “Getting tough” on social promotion is coupled with a call for higher educational standards.

In 1997, the Chicago Public Schools took the lead in abolishing social promotion. Students who do not master curriculum at certain checkpoints or “promotional gates” are required to attend summer school to master the content or repeat their grade the following year.

A rarely mentioned consequence of ending social promotion is a sharp rise in summer school enrollment for students who fail to meet promotion standards. Summer school now functions as the educational system’s “release valve” for dealing with large numbers of students who have not met the new standards. The following examples give an indication of the magnitude of the problem.

  • In Chicago, 130,200 students attended summer school this year at a cost of $65 million (Spielman, 1998). Of those, more than 11,500 students did not successfully complete summer school and will repeat their previous grade. Even more disturbing is the fact that 14 percent of the 11,294 students who were retained last year will be retained a third time. Could it be the Chicago Public Schools has adopted the concept of a “three-peat” from the Chicago Bulls?
  • The Denver Public Schools required 2,500 students to attend summer school as a condition for promotion in the 1997-98 school year (Harrington-Lueker, 1998).
  • Washington, D.C., schools had difficulty financing summer school for the 20,000 students who did not meet testing standards (Harrington-Lueker, 1998).

A little-known fact regarding summer school is that some districts require students to pay for each course they must take. When one considers that the majority of students who are required to attend and pass summer school as a condition for promotion are low-income minority students, it becomes obvious that the requirement of summer school attendance turns out to be financed by those families who are least able to pay.

Another corollary structure to in-grade retention is the transitional school or academy for students who are not yet ready (in terms of mastering grade-level skills) to transition from one level of schooling to another. In the case of double and triple retainees, pre-high school and pre-junior high schools serve as “stopovers” for those who are physically too large to attend classes with students two or three years their junior. Transition rooms are no longer adequate to hold the burgeoning numbers of students who are being retained in-grade.

The pendulum has no memory. The schools chancellor of New York City recently called for raising educational standards by reinstituting the Promotional Gates Program in the year 2000 and thus curtailing social promotion. This program, in essence, constitutes a warmed-over version of the same program initiated in 1981 to check the educational progress of students in grades four and seven, based on reading scores. In the second year of the program, officials also tested math achievement. The program was a failure that cost in excess of $40 million to implement (House, 1989). Instead of restructuring or redesigning education to serve students, the schools simply gave stronger doses of what had not been effective previously. One researcher stated, “As a consequence, it was not uncommon to find 12-year-olds stuck in fourth grade and 17-year-olds repeating seventh grade” (Darling-Hammond, 1998).

Reflecting on New York’s earlier unsuccessful experience with the Promotional Gates Program, it is difficult to understand why it will soon be re-implemented. One has to ask the question, “Who is not learning in this situation – the students or the administrators?” Why, in a span of less than 20 years, is the same unsuccessful policy of retaining students at promotional gates being re-implemented, particularly in light of what research has shown about the negative consequences of retention in the interim?

Retention Receives Failing Grades

Social science research (and educational research, in particular) is often criticized for not producing definitive results. Perhaps a greater irony lies in the fact that when definitive results are produced, they are ignored. Certainly that is the case with the research on in-grade retention.

The research is unequivocal – the effects of retention are harmful. As early as the 1930s, studies reported the negative effects of retention on academic achievement (Ayer, 1933; Kline, 1933). Retention does not benefit students academically or socially (Foster, 1993; Harvey, 1994; Holmes, 1989; Holmes and Mathews, 1984; Shepard and Smith, 1989; Walters and Borgers, 1995). Highlights from the literature include the following.

  • In-grade retention does not ensure significant gains in achievement for children who are academically below grade level (Bocks, 1977; Cárdenas, 1995; Holmes, 1983; Holmes, 1989; Holmes and Mathews, 1984).
  • A meta-analysis of 63 studies found that on average, retained children fare worse than their promoted counterparts on both personal adjustment and academic outcomes (Holmes, 1989; Foster, 1993).
  • The threat of non-promotion is not a motivating force for students to work harder (Bossing and Brien, 1979; Darling-Hammond, 1998).
  • Retention is strongly associated with dropping out of school in later years (Grissom and Shepard, 1989; Roderick, 1995); a second retention makes dropping out of school a virtual certainty (Setencich, 1994).
  • Small advantages that accrued to first graders who were retained washed out by the end of third grade (Butler, 1990a; Butler, 1990b; Karweit and Wasik, 1992; Shepard, 1989; Snyder, 1992).
  • Students who are retained suffer lower self-esteem and view retention as a punishment and a stigma, not a positive event designed to help them (Byrnes, 1989; Cárdenas, 1995).
  • Retained students on the average, compared to matched controls, were reported to do more poorly on follow-up measures of social adjustment, attitudes toward school, behavioral outcomes and attendance (Holmes, 1989; Meisels and Liaw, 1993; Rumberger, 1995).
  • Analysis of those who are retained in-grade shows that African American students and Hispanic students are retained at twice the rate of White students (George, 1993). Additionally, 40 percent of students who repeat grades come from the lowest socio-economic quartile, as compared to only 8.5 percent from the highest quartile (Meisels and Liaw, 1993).
  • Between 1990 to 1997, 66 studies were conducted on retention, with only one supporting it (Lenarduzzi and McLaughlin, 1990).

How many students are retained annually? The national prevalence of in-grade retention proves difficult to ascertain because national retention data are not collected by the U.S. Department of Education. However, the National Longitudinal Study of Adolescent Health found that 21.3 percent of adolescents were retained in at least one grade (Resnick, et al., 1997). State-level retention data are not collected uniformly, and when collected, they are not comparable due to their idiosyncratic nature.

In Texas during the 1995-96 academic year, 144,683 students (4.3 percent) were retained at a cost of $4,756 per pupil, causing an additional expenditure of $688 million dollars (TEA, 1997). The profile of retained students in Texas does not follow the national trend. While two-thirds of all retentions nationally occur between kindergarten and third grade (Meisels and Liaw, 1993), in Texas the highest rates of retention occur in the ninth grade among disadvantaged, urban minority males. For the years 1992 to 1995, one in six Texas ninth grade students, primarily Hispanic and African American, was retained (see box below). This is particularly disturbing considering that being over-age in grade more than doubles a student’s chances of being retained in-grade and of dropping out (Shepard and Smith, 1989; Roderick, 1994).

The next highest percentage of retentions in Texas occurred in first grade, where total retention rates for 1992-93 to 1995-96 were 6 percent, 5.8 percent and 5.9 percent, respectively. At this level, African American and Hispanic students also had higher rates of representation than did White students (TEA, 1997).

Texas law requires that students show academic proficiency to be promoted, but local districts set their own promotion policy. The Texas Assessment of Academic Skills (TAAS) test was not designed to be used to determine whether students should be promoted, but it is used that way in some districts. A case in point is Waco, Texas, where parents brought suit against the Waco Independent School District to block the board’s 1997-98 promotion policy, which requires students in grades one through eight to pass all of their core classes and maintain a 90 percent attendance rate. Students in third through eighth grade must score a combined average of 70 on the math and reading portions of the TAAS. The new policy resulted in 11.9 percent of elementary and middle school students being retained in-grade. The injunction was turned down recently by a state district judge, but a trial for a permanent injunction against the policy is expected next year (San Antonio Express-News, 1998).

Reasons for Retention

The reasons for retention vary with districts’ promotion policies, but in most cases, they are related to a state’s accountability measure. In a study of characteristics of retained children, D.A. Byrnes determined through a survey of teachers and principals that children retained in-grade were described by educators as being immature and having low self-esteem and low motivation (1989). In the particular district where the study was conducted, limited English proficiency was also characteristic of children who were retained.

Another underlying reason for retention is that it is perceived as efficient and meritocratic. The assumption is that those who work sufficiently are promoted; those who do not work hard enough, do not earn promotion to the next grade and are retained in-grade until they learn the curriculum. It also serves to keep the structure of schools intact, maintaining the status quo. It is a strategy that projects the aura of cracking down on students who are “academic slackers” by instituting stricter standards and policies for promotion. Such policies play well to the public but, unfortunately, are harmful to students.

E. House summarizes some of the detrimental effects of in-grade retention:

Students are retained in rather arbitrary and inconsistent ways, and those flunked are more likely to be poor, males and minorities, although holding students back is practiced to some degree in rich and poor schools alike. The effects of flunking are immediately traumatic to the children, and the retained children do worse academically in the future, with many of them dropping out of school altogether. Incredibly, being retained has as much to do with children dropping out as does their academic achievement. It would be difficult to find another educational practice on which the evidence is so unequivocally negative (1989).

Alternatives to Retention

Traditional offerings for students who are retained in-grade include two-year kindergartens, transitional rooms or “half grades” and tutoring programs. Two-year kindergartens and transitional rooms operate on the premise that students just need more time to mature and develop the appropriate skills. An ancillary line of thinking is that curricula should be presented in a linear fashion. In general, the purpose of these traditional responses to improving retained students’ achievement is to give them a larger dose of what failed to work the first time. There is also a tendency to place students in remedial tracks that often become permanent. The approach is deficit in nature and situates the reason for failure within the child, ignoring the possibility that the educational program, the instructional approach or the teacher played a part in the child’s failure.

Linda Darling-Hammond calls for a shift in perspective from a deficit model (which situates the problem of poor achievement solely within the child) to one that acknowledges that classroom and school practices may be contributing factors to a student’s lack of achievement (1998). Acknowledging the ineffectiveness of both retention and social promotion, she offers administrators four complementary alternatives to retention.

Strategy 1: Enhance the professional development of teachers to ensure they have the knowledge and skills to teach a wider range of students to meet standards. This strategy involves staff development in the sense of sustained learning over time, where teachers learn effective strategies to help students learn standards. Such an approach to staff development would include teacher academies, professional development laboratories or university offerings that support individual and collective teacher learning (Darling-Hammond, 1998). Staff development needs to be sustained and responsive to teachers’ needs. Darling-Hammond makes an often-ignored point that the two most common motivators for achievement – standards and assessments – do not operate without competent teachers.

Strategy 2: Redesign school structures to support more intensive learning. Age-grade cohorts were adopted in the mid-19th century to efficiently move groups of students through a sequential curriculum correlated to grade. Given the different rates of development, particularly for children in the early grades, multi-age classes (where teachers stay with the same class for more than one year) provide many advantages: (a) exposure to older, more competent peers who can help provide appropriate models of behavior and academic assistance; (b) opportunities for more intensive instruction; and (c) teachers who come to know their students better over time.

Another example of structural change that might support learning is cross-grade grouping. Students do not have to be locked into their appropriate age-grade cohort for all of their instruction. For example, a fourth grade student who has trouble with reading could attend reading instruction with the third graders; likewise, more advanced readers could attend a reading class with the fifth graders. Some students may only perform poorly in one subject but excel in others. In these cases, repeating an entire grade deprives them of learning new academic material.

Strategy 3: Provide students the support and services they need in order to succeed when they are needed. This strategy is presented together with the fourth due to their interrelation.

Strategy 4: Use classroom assessments that better inform teaching. In order to know that students need help and to be able to structure lessons appropriately, teachers should use performance-based or informal assessment techniques to understand how students approach learning. Informal assessment techniques such as keeping anecdotal records and using checklists and rubrics as part of the instruction, provide information on the process of learning rather than just the product (i.e., standardized test scores). Once teachers identify the problems, schools should provide opportunities for students to receive instructional support or resources as they need it. Saturday school has proven to be effective in providing these services. Darling-Hammond cites Reading Recovery and Success for All as two successful literacy programs that stress individual instruction in the early grades (1998).

The Talent Development model originated from research conducted by the Center for Research on the Education of Students Placed at Risk (CRESPAR) at John Hopkins University. It provides a fresh approach for reconceptualizing instruction for students who have not been well-served by traditional education. Everything about Talent Development schools’ organization, curriculum teaching and student support structure stems from the belief that schools must develop talent and that they can do this best in schools where every student has access to an engaging standards-based curriculum in heterogeneous classrooms. In addition, schools must be a place where every student is in a classroom with caring teachers and peers who are “rooting for them to do well, who are encouraging them to give their best in the classroom, and who are doing everything in their power to help them improve their skills and increase their understanding” (MacIver and Plank, 1996).

A. Wheelock cautions that setting up Talent Development schools is a complex process that takes several years (1998). It is not based on funding a series of “add-ons” but depends on a coherent culture of high standards grounded in research-based strategies to improve student achievement.

As we approach the new millennium, educators speak of becoming part of the “information superhighway” and advancing classroom learning through technology. It is ironic that futuristic conceptualizations of education admit discussions of retention, a practice that was first shown to have negative effects on students as early as the 1930s. Since then, the literature has grown in depth and breadth. Yet, predictably, every 10 years we decide to give it another chance.

Retention gets the public’s attention by advocating a kind of educational “tough love,” but unfortunately it does not deliver what it promises. The growing numbers of students who are retained two and three times and the implementation of special schools for the legion of students who cannot pass to the next level of schooling, attest to the fact that we need to explore other alternatives. Students need to have access to different ways of acquiring the knowledge and skills they need for the 21st century. It is time to admit that retention has never worked well; moreover, it does not offer a viable alternative for the future.

Historical Review of Grade Level Retention




Total Students*



Number Retained



Retention Rate

























* The value for “Total Students” is derived using information from two consecutive PEIMS submissions. It is not the same as fall enrollment counts that districts report directly to PEIMS.

Source: Texas Education Agency, 1998.


Ayer, F.C. Progress of Pupils in the State of Texas 1923-33 (Austin, Texas: Texas State Teachers Association, 1933).

Bocks, W.M. “Non-Promotion: A Year to Grow?” Educational Leadership (1977) 34 (5), 379-83.

Bossing, L. and P. Brien. A Review of the Elementary School Promotion Retention Dilemma, Murray, Kentucky, Murray State University (ERIC Document Reproduction Service NO. ED 212-362, 1979).

Butler, J. “Effects of Retention on Achievement and Self-Concept of Kindergarten and First Grade Students,” Paper presented at the annual meeting of the Mid-South Educational Research Association (November 1990).

Butler, J. “Effects of Retention Among Children in the United States,” Pediatrics (November 1990) 93(3), 481-487.

Byrnes, DA“Attitudes of Students and Educators Toward Repeating a Grade.” In L.A. Shepard and M.L. Smith (Eds.), Flunking Grades: Research and Policies on Retention (Philadelphia: The Falmer Press, 1989) 108-131.

Cárdenas, J.A. Multicultural Education: A Generation of Advocacy (Needham Heights, Mass.: Simon and Schuster Custom Publishing, 1995).

Darling-Hammond, L. “Alternatives to Grade Retention,” Internet posting, The School Administrator (August 1998).

Foster, J. “Review of Research: Retaining Children In Grade,” Childhood Education (1993) 70:38-43.

George, C. Beyond Retention: A Study of Retention Rates, Practices, and Successful Alternatives in California (Sacramento, Calif.: California State Department of Education, 1993).

Grissom, K.B. And L.A. Shepard. “Repeating and Dropping Out of School.” In L.A. Shepard and M. L. Smith (Eds.), Flunking Grades: Research and Policies on Retention (Philadelphia: The Falmer Press, 1989), 34-63.

Harvey, B. “To Retain or Not? There is No Question,” Paper presented at the annual meeting of the American Association of School Administrators (San Francisco, Calif.: AASA, 1994).

Harrington-Lueker, D. “Retention vs. Social Promotion,” The School Administrator (1998) 7(55), 6-12.

Holmes, C.T. “The Fourth R: Retention,” Journal of Research and Development in Education (1983) 17, 1-6.

Holmes, CT“Grade-Level Retention Effects: A Meta-Analysis of Research Studies.” In L.A. Shepard and M.L. Smith (Eds.), Flunking Grades: Research and Policies on Retention (Philadelphia: The Falmer Press, 1989), 16-33.

Holmes, CT and K.M. Mathews. “The Effects of Nonpromotion on Elementary and Junior High School Pupils: A Meta-Analysis,” Review of Educational Research (1984) 54(2), 225-36.

House, E. “Policy Implications of Retention Research.” In LA Shepard and M. L. Smith (Eds.), Flunking Grades: Research and Policies on Retention (Philadelphia: The Falmer Press, 1989), 202-13.

Karweit N. and B. Wasik. A Review of the Effects of Extra-Year Kindergarten Programs and Transitional First Grades (Baltimore: Center for Research on Effective Schooling for Disadvantaged Students, Johns Hopkins University, 1992).

Kline, E. “Significant Changes in the Curve of Elimination Since 1900,” Journal of Educational Research (1933) 26, 608-616.

Lenarduzzi, G.P. and T.F. McLaughlin. “The Effects of Non-Promotion in Junior High School on Academic Achievement and Scholastic Effort,” Reading Improvement (Fall 1990) 27(3), 212-217.

MacIver, D.J. and S.B. Plank. The Talent Development Middle School. Creating a Motivational Climate Conducive to Talent Development in the Middle Schools: Implementation and Effects of Student Team Reading. Report No. 4 (Baltimore/Washington, DC: John Hopkins University and Howard University, Center for Research on the Education of Students Placed at Risk, September 1996).

Meisels, S.J. And F. Liaw. “Failure in Grade: Do Students Catch Up?” Journal of Educational Research (1993) 50(2), 69-77.

Resnick, M.D., P.S. Bearman, T.W. Blum, K.E. Bauman, K.M. Harris, J. Jones, J. Tabor, T. Beuhring, R. Sieving, M. Shew, L.H. Bearinger and J.R. Udry. “Protecting Adolescents from Harm: Findings from the National Longitudinal Study on Adolescent Health,” Journal of the American Medical Association (September 10, 1997) 278, 823-832.

Roderick, M. “Grade Retention and School Dropouts: Investigating the Association,” American Educational Research Journal (1994) 31, 729-759.

Roderick, M. “Grade Retention and School Dropouts: Policy Debate and Research Questions,” Phi Delta Kappan Research Bulletin (December 1995) 15, l-6.

Rumberger, R.W. “Dropping Out of Middle School: A Multilevel Analysis of Students and Schools,” American Educational Research Journal (1995) 32(3), 583-625.

Setencich, J. “The Impact of Early Grade Retention on the Academic Achievement and Self-Esteem of Seventh and Eighth Grade Students,” Paper presented at the Annual Convention of the National Association of School Psychologists (Seattle, Wash.: March 1-5, 1994).

Shepard, LA“A Review of Research on Kindergarten Retention.” In LA Shepard and ML Smith (Eds.), Flunking Grades: Research and Policies on Retention (Philadelphia: The Falmer Press, 1989), 64-78.

Shepard, LA and ML Smith. Flunking Grades: Research and Policies on Retention (Philadelphia: The Falmer Press, 1989).

Snyder, J. “The Effects of Retention in Elementary School on Subsequent Academic Performance,” Paper presented at the annual meeting of the Mid-South Educational Research Association (Knoxville, Tenn.: November 1992).

Spielman, F. “Daley Charts a Busy Summer for 200,000 Kids,” Chicago Sun Times (April 3, 1998).

Texas Education Agency. Report on Grade-Level Retention of Students: 1995-96 (Austin, Texas: September 1997).

Texas Education Agency. Report on Grade-Level Retention of Students: 1996-97 (Austin, Texas: September 1998).

San Antonio Express-News. “Waco Schools Keep Promotion Policy,” San Antonio Express-News (August 14, 1998).

Walters, D.M. and S.B. Borgers. “Student Retention: Is it Effective?,” School Counselor (March 1995) 42.4:300-310.

Wheelock, A. “Extra-Help and Support to ‘Meet Standards’ and Prevent Grade Retention,” Internet posting (1998).

Pam McCollum, Ph.D., is a senior education associate in the IDRA Division of Professional Development. Comments and questions may be sent via e-mail at feedback@idra.org.

[©1998, IDRA. This article originally appeared in the November – December 1998 IDRA Newsletter by the Intercultural Development Research Association. Permission to reproduce this article is granted provided the article is reprinted in its entirety and proper credit is given to IDRA and the author.]