H O P K I N S    I N    T H E    N E W S

The Clinton administration has proposed a ban on social promotion. Should failing students be forced to repeat a grade? During his State of the Union Address, President Clinton listed ending social promotion as part of a larger package to improve education in America.

However, many studies argue that holding students back puts them at much greater risk for future failure than their equally achieving, non-retained peers.

Many education analysts say too many students are already being held back. Lorrie Shepard of the University of Colorado, co-editor of "Flunking Grades: Research and Policies on Retention," says the rates are as high as they were in the 19th century--approximately 25-30% of all students in the US have flunked at least one grade. She points to several studies that show that the stress and loss of self-esteem harms students' academic achievement in the long run. In one study, children said going blind or losing a parent were the only two life events that would be more stressful than being held back.

Karl Alexander of Johns Hopkins University ( Department of Sociology), co-author of "On the Success of Failure," takes a different view. He followed 775 students in the Baltimore system for 8 years and concluded that retention supplied the extra time needed by some students to perform at acceptable levels. Alexander argues self-regard is not irreparably harmed, and that retention stops academic free fall.

Other educators say the either-or approach misses the point. What is needed, they argue, is early identification of failing students and intervention through tutoring, after-school programs and summer school. With this in mind, the president's education proposal includes tripling the money for summer school and after-school help.

But critics say that even the proposed $600 million would not cover the great need for such programs. Their concern is that the president's ban on social promotion would cause harm in inner-city schools, where each year, several students are held back per class, and where money for intervention programs is much more scarce than in the suburbs, where retention is not as big a problem.

What do you think? In situations where intervention programs are not viable, should schools hold back failing students, or should they promote them to the next grade? Is this an issue for the Federal government, or should it be up to the local school boards?

Your questions are answered by Professor Shepard and Professor Alexander.

Bob Brunk of Los Angeles, CA, asks : I have a son and wonder if some day he will be held back a grade. What should a parent do if an educator proposes such a drastic measure? I was held back and I never truly recovered emotionally, even though my grades did.

Lorrie Shepard, University of Colorado, responds: If an educator proposes that your son be held back, you should inform yourself about the research findings on grade retention and then insist that your son be given the kinds of academic help that are known to be more effective than retention.

The most up-to-date summary of research on grade retention is provided in a 1999 report by the National Academy of Sciences called High States: Testing for Tracking, Promotion, and Graduation (J. P. Heubert & R. M. Hauser, Eds.; http://www.nap.edu). Contrary to popular belief, retention does not help students academically in the years following the repeated grade. Out of 63 matched-control studies reviewed by Professor Charles Holmes of the University of Georgia, 54 showed overall negative effects of retention on academic achievement; only 9 studies found positive outcomes. In addition to being an ineffective treatment for improving student achievement, retention has serious side effects. In the controlled studies reviewed by Holmes, retention had a negative effect on students' attitude toward school and on their social and emotional adjustment. Using in-depth interviews Professor Deborah Byrnes at the Utah State University documented that "flunking" is indeed a traumatic experience for many students. Moreover, being retained in grade dramatically increases the likelihood that students will dropout of school later on (again see the National Academy of Sciences report).

Researchers are not sure why retention does not improve achievement. Some speculate that it is because of retention's negative effect on student motivation that it also has a negative effect on learning. Others think that effects might be negative because repeating a grade still does not address the specific learning difficulties a child may be having. Tutoring, after-school programs, summer school, and intensive reading programs are all more effective interventions than retention for improving academic achievement. These "treatments" have all been documented to have positive effects in the research literature; they can all be more focused and targeted for specific learning needs than repeating an entire grade level; and they can all begin immediately, as soon as a student begins falling behind, instead of waiting for the dreaded end-of-year notice of failure.

More than any other single problem, students who are failing in school need help with reading. If your son is having trouble learning to read in first grade, or if he is in a higher grade and still cannot read fluently enough to comprehend grade level material, then ask for help from a reading specialist. Work with the specialist to find out what particular problems your child is having. Some children can sound out words perfectly but have never been encouraged to think about the meaning of what they are reading; other children have not mastered decoding, either because they were not taught enough about letter-sound correspondences or because they could not "hear" what was being taught. At the same time that you and your son's teacher formulate a plan to work on his more serious learning problems, I also encourage you to identify his strengths and what he likes best about school. See if you can think of ways that his interest in football, computer games, or whatever, can be used as the context for practicing school skills that he typically avoids.

Karl Alexander, Johns Hopkins, responds: Bob, I've done research on grade retention and feel as though I understand a bit about the practice from that perspective, but research tells us about what happens to children in the typical situation or "on average." That's not good enough for making placements in the individual case, as each child's situation is unique. It falls to parents, teachers and administrators together those closest to the child-- to decide what is the best course, and that depends on a good, firm understanding of the nature of the child's difficulties weighed against the available options. Children who are candidates for retention by and large are far behind academically and many have difficulty complying with classroom routines. These are problems that need fixing, and for that reason I'm not at all a fan of social promotion, which doesn't fix anything. But traditional grade retention, in my view, should be a last resort, and then only when there is reason to think the extra time is what's needed. Let me explain why I feel this way.

My research in Baltimore does not find that retention is positively harmful, as some critics of the practice contend. Indeed, my colleagues and I conclude that repeaters' school performance improves some as a result, and we don't find evidence of emotional scars (again, in the "typical case," which might not be anything like your experience. Almost half the children in our study were held back in the elementary grades, and I suspect the emotional fallout is different maybe more severe when retention is less common). At its best grade, retention costs children a year and is expensive (schools have to educate children for an extra year), and even then it generally doesn't bring children up to the level of performance we would want for them (rather, what my research shows is that it helps them catch up to other poor performing children who were promoted). We should be able to do better than this by our children.

Not all children struggle at school for the same reasons, or in the same way. I should think that you, as a concerned parent, would want to find out from the people at school who have been working closest with your child: exactly what kinds of problems your child is experiencing; what they think is behind those problems; and the corrective alternatives that might be recommended. Assessment and evaluation is key-- if you don't have a good understanding of the problem, I don't see how you can have much confidence in the recommended solution. The procedures for deciding the best course in problem situations vary greatly from school system to school system-- here in Baltimore the teacher or a counselor typically makes a recommendation to a Promotion Committee, which reviews the case, confers with the parents, and then makes a recommendation to the principal. Parents must agree if the recommendation is to retain. As a process, this seems about right, but I'm not convinced that the quality of the information that feeds into the process is as good as it should be, and in Baltimore the range of alternatives short of retention unfortunately is quite limited. In terms of information, I'm impressed with the assessment protocol that Jim Grant and Irv Richardson have developed in their "Retention/Promotion Checklist." It's an intelligent approach, that incorporates research-based "best practices." For more information, they can be contacted at the Society for Developmental Education, 603-924-9621.

Franklin Chavies of Himyar, Kentucky, asks : I attended school in the 30's and was considered very smart and was promoted ahead of my other classmates several times in the early years which made the rest of my schooling very hard. Are there studies on skipping grades? Do they have any bearing on the issue of promoting students who are not having such an easy time?

Karl Alexander, Johns Hopkins, responds: So far as I know there's not much research on how skipping grades affects children, and that's unfortunate. There are studies that show acceleration has positive effects on achievement, but I'm not familiar with studies that consider how children who are much younger than their classmates fare socially and emotionally.

A colleague of mine here at Hopkins, Julian Stanley in the Psychology Department, has been at the forefront of developing enrichment experiences for extremely bright middle-schoolers, and his programs attend very carefully to these children's social needs. Stanley's interventions, though, are supplementary to children's regular schooling they take place after school, on weekends, and during the summer. This problem of being "out of synch" may be a particular concern during adolescence, when "fitting in" is so important. And perhaps it weighs on repeaters and accelerated children alike-- repeaters who are older, and accelerated students who are younger. In fact, research by Melissa Roderick, reported in her book "The Path to Dropping Out," suggests that being overage for grade may help explain some portion for the association between grade retention and dropout a relationship that is well established in the literature and cause for concern. Roderick's research finds that children who were held back in the primary grades are at elevated risk of dropout, but so too are children who started school late (so-called "academic redshirts), who also would be older than their classmates when they get to high school and can legally leave school. This suggests to me that the social side of schooling plays an important role-- children who are not fitting in comfortably AND have other options may well avail themselves of those options.

Lorrie Shepard, University of Colorado, responds: As your question implies, there are interesting parallels between repeating grades and skipping. Both practices derive from a factory model of schooling developed early in this century as cities were faced with educating massive numbers of students migrating from rural areas and from abroad. According to this model, which works like a factory assembly line, both "laggards" and "precocious pupils" were problems that had to be eliminated because they didn't fit the standard curriculum aimed at the average pupil.

Skipping is rarely used in public schools today and has been such an uncommon practice for decades that I know of no studies investigating its effects. The idea of skipping students fell into disfavor for both academic and social reasons. First, high achieving students are usually more than one grade level ahead in their strongest subject (usually reading). Should they be skipped more than one grade? At the same time, students who skip have gaps in their knowledge from the curriculum they missed. Second, students who skip often have difficulty adjusting socially.

The parallels, which would likewise argue against retention, are striking. Students who are retained are typically behind by different amounts in their weakest and strongest subjects, yet repeating a grade is a uniform corrective that treats all subject areas the same. A fifth grader could be two years behind in reading, should he go back to third grade? What about math where he is only 6 months behind? Retention is also known to carry a social stigma, which may have a negative effect on student motivation and achievement.

An important difference between skipping and repeating is that schools have turned to curriculum enrichment and gifted and talented programs as a means of addressing the learning needs of high achieving students while keeping them in their normal grade placement. Because of beliefs in this country that equate automatic promotion with lax standards, however, schools have not been as effective in providing special programming for below-average learners.

June of Westlake, LA, asks : How does the present retention rate compare to rates of 40 years ago? We've seen falling academic performance in international studies do rates of retention fit into that pattern in any way? What is the attitude toward retention in other countries?

Lorrie Shepard, University of Colorado, responds: Retention rates have ebbed and flowed over the last 40 years, but there has never been a time when there have not been substantial numbers of students retained. A national longitudinal study conducted in the 1980s by the National Center for Education Statistics found that 20% percent of all eighth graders had been retained at least once. The recent National Academy of Sciences report, High States: Testing for Tracking, Promotion, and Graduation (J. P. Heubert & R. M. Hauser, Eds.; http://www.nap.edu), used U.S. Census data to examine trends in retention. "Overall, a large share of each birth cohort now experiences grade retention during elementary school. Among children ages 6 to 8 from 1982 to 1992, age-grade retardation has reached 25 to 30 percent by ages 9 to 11" (p. 152). As Professor Alexander found with his study in Baltimore, retention rates are much higher in urban school systems, with 50% or more of all students having been retained at least once.

Statistics on retention rates are too inaccurate to do a close analysis of how changes in retention rates might affect international comparisons. Given that retention has not been shown to improve achievement either in controlled studies or in massive non-promotion programs tried in the 1980s (e.g., New York City's Gates Program and Chicago's stricter eighth-grade promotion standards), there is no reason to believe that increasing retention rates would increase the performance of U.S. students in international comparisons. Of course, cynically, we do have to consider that increasing retention could improve the ranking of U.S. students if more low achieving students will have dropped out of school by the time the tests are given.

Data available from UNESCO show tremendous variability in the retention practices of other nations. In developing countries such as Brazil, the annual retention rate is 20 percent. Some children may spend three or four years in the first grade; 84 percent have dropped out by eighth grade. Among European nations, France has relatively high retention rates; only one-third of working-class children complete elementary school on time compared to an 84 percent rate for the children of professionals and executives. Some countries, such as Japan, Korea, Denmark, Norway, and Sweden have "automatic" promotion policies for elementary education meaning that no students are retained; for Japan the automatic promotion policy applies to secondary education as well.

Karl Alexander, Johns Hopkins, responds: The second part of your question, June, is easier to answer than the first grade retention is not commonplace in other industrial countries, although it's not altogether clear how they manage to avoid it.

One suggestion I've seen is that "looping" is more common in other countries the practice of having the same teachers stay with children over several years. Time trends here in the U.S. are harder to assess, as no data on retention are compiled at the national level. The best you can do from national Census data is estimate the percentage of children who are older than the typical age for their grade level, but that's very imprecise for gauging retention birthdate cutoffs vary from place to place and across time, as do school starting dates, and then many children start school late (I've seen one estimate that 7% of first-graders nationally are academic "redshirts," but I have no idea how accurate that figure is). And even these data, with all their inadequacies, only are available back to 1960. That means the picture has to be patched together from local sources, and Dr. Shepard has done as good a job as anyone of organizing that information. However, it might be of interest to you that a recent National Academy of Sciences report ("High Stakes," by J. P. Heubert and R. M. Hauser) estimates that about a third of students nationally are overage for grade by age 17 - 18.

David Larabee, in an article published in 1984 in the Harvard Education Review (a very well respected journal), reviews the history and ideology of promotion/retention practices here in the U.S. It's good reading if you're interested in an historical perspective on the issue.

Casey Wyland of Lambertville, NJ, asks: Why have you two come to different conclusions? Is it a question of method or interpretation?

Karl Alexander, Johns Hopkins, responds: To be honest, Casey, I don't know why our perspectives on the effects of grade retention are so different, but I don't think we're all that far apart on what we think is good practice. I believe my own research is sound and we apparently read somewhat differently the implications of other research. However, these differences involve whether retention is helpful or harmful, compared to promotion, when children are far behind academically. That's vastly different than asking whether EITHER retention OR promotion helps poor performing children master the school curriculum at acceptable levels.

Dr. Shepard and I recently wrote point/counterpoint essays for the Scripps-Harris News Service that, in my reading of them, come to quite similar conclusions. These essays were published recently in several newspapers, but I was upset when I saw that mine had been edited down (I would guess that this was done to Dr. Shepard's also)-- upset because important points had been deleted. In any event, that brief essay says pretty much what I want to say about the promotion/retention controversy, so here it is: "President Clinton's call to end social promotion in his State of the Union address has made retention policy a front burner-issue. This is a healthy development in my view, but only if it encourages serious discussion of how best to meet the needs of children who are not achieving at acceptable levels. That requires getting beyond the stale "to retain or not" debate-- it's too confining.

With colleagues Doris Entwisle and Susan Dauber, in 1994 I published a book on the effects of grade retention ("On the Success of Failure," Cambridge University Press) that monitors the school progress of some 800 Baltimore children from first grade thorough middle school. More than 40% of the children we studied were held back during that time, many twice. Such numbers signal a problem of immense proportions, for the young people involved, who lose precious time, and for the schools, which must bear the cost of educating repeaters for an extra year or two. But what of the impact of retention, is it helpful, as intended, or harmful, as critics of the practice claim? We looked, carefully, at school performance (achievement test scores and marks) and attitudes (toward self and school) after children were held back in relation to the same criteria from before they were held back and in relation to promoted children's school performance and attitudes over the same time frame (some of these comparisons spanned eight years). Contrary to the critics, retention was not positively harmful; more than that, it helped. Children held back in the early elementary years (our focus) made up ground academically and suffered no emotional scars. These youngsters-- who were far behind academically before being retained did not catch up altogether, but they were better off than before. On that basis, we deemed retention a qualified success.

That conclusion seems to have cast us in the role of "friends of retention," but that misreads our position. Retention "works" only in a narrow sense-- when the later situation of repeaters is compared against other poor performing students who were promoted (or, equivalently, against all promoted children, using statistical means to adjust for achievement differences that predate retention). This approach attempts to compare like against like. It is the frame of reference used in all evaluations of grade retention, and by this standard our evidence is sound. But if we believe that all children can, and should, realize success at school, then simply seeing repeaters keep up with other low achievers is small comfort.

At this point I part company with the critics. Many seem so preoccupied with the practice that they lose sight of why these children are candidates for retention in the first place. Most are far behind academically, and social promotion-- simply moving them ahead unprepared for what awaits them won't fix anything. The time bought by repeating a grade can help, but the decision to promote or not should be made case by case, based on firm understanding of a child's difficulties and needs. For most poor performing children under most circumstances, this should be the option of last choice, not first choice. For them, we need a "third way," one that shores up their skills before problems mount. Partial promotion, summer programs, ungraded classes, looping, cooperative learning, and supplemental services (e.g, tutoring) are promising avenues, but social promotion has no place on the list.

I applaud President Clinton's call for an end to social promotion if it is a mandate to find this "third way." My fear is that many instead will take it simply as license to hold back more children."

Lorrie Shepard, University of Colorado, responds: The differences in our conclusions are due to both method and interpretation.

In his book, On the Success of Failure, Professor Alexander discounts the findings from Holmes's review of 63 matched-control studies on several grounds, especially he argues that match-control studies are never as good as randomization in accounting for initial differences between groups. While I agree with this basic scientific principle, it is possible to evaluate the degree of control and possible sources of error across a whole body of work, just as has been done with the non-randomized studies showing the relationship between cancer and lung cancer. When Holmes examined the degree of control, the better controlled studies showed more serious negative effects from retention.

Professor Alexander also points to the fact that negative findings don't outweigh positive findings by quite as much in recent studies as they did in past decades, though negative findings are still in the majority. Again, I agree in principle to his preference for relying on more recent studies. However, according to Holmes's review the small number of recent studies that show positive effects are highly unusual both in the characteristics of students retained and in the special help provided. All of the nine positive studies were conducted in suburban settings with few minority students. Although retained students in these studies were reading below grade level, they were scoring at the national average in mathematics and language arts. In addition these students were not simply recycled through the regular curriculum but were placed in special classes with low student-teacher ratios. Thus, I find it hard to generalize these findings to more typical settings where grade retention is being used, and I am inclined to attribute the positive outcomes from these studies to the special interventions rather than to retention. In the one study that I know of where retention-with-special-help was compared to promotion-with-special-help, low achieving students benefited more from promotion-with-special help.

I also disagree with Professor Alexander regarding the findings from his study in Baltimore. My colleagues Mary Lee Smith, Scott Marion, and I published a review of On the Success of Failure and a rejoinder to Alexander's reply in Psychology in the Schools (1996(3); 1998(4)). Some of our criticisms and reanalyses were quite technical, but a simplified summary is as follows. The largest sample of retainees in Alexander's study was 104 first-grade retainees. Alexander et al. acknowledged that the effect of retention on first graders was quite negative. "Not only did they trail badly, but they do so by more after being held back then they had before" (p. 120). For the sample of 62 fourth-seventh grade retained students, according to Alexander et al., "it is hard to say that retention either helps or harms children held back at higher grade levels: they seem to be doing satisfactorily all along, before, during, and after retention" (pp. 81-82). The best claim to benefits from retention is with the samples of 59 second-grade retainees and 37 third-grade retainees, where Alexander says that retention "stops academic free fall." However, his own statistical analyses show that these students benefited only in the repeat year itself and after that they were back at roughly the same percentile level (with their new grade mates) as they were when they first started the repeated grade. Given these results, we don't see how Alexander and his co-authors could report "a preponderance of positive effects" from retention.

Jerry McArthur of Fargo, ND, asks : Are there statistics on how many schools have programs that identify and help students who are having trouble and can we train our teachers to handle this task while they are juggling so may others?

Lorrie Shepard, University of Colorado, responds: There are national data available on the numbers of students being served in through Head Start, Title I reading programs, special education, and other federally funded programs. In addition, many states and districts have their own special programs. Yet, we know that these resources are still too few and spread too thin to provide high quality education to students who are at risk for academic failure. For example, Head Start reaches only a small fraction of the children who are qualified for services according to the program's criteria; and even with all of these programs in place we know that 40% of 4th graders are not performing at even the Basic level on the National Assessment of Educational Progress.

Teacher training will help, and indeed one of the goals of standards-based reform is to help teachers learn to teach more rigorous curricula in ways that help students "make sense" of difficult material rather than trying to memorize what they can't understand. Yet, as you suggest, teachers are already being asked to do much more than their counterparts of 10 or 20 years ago. That's why the issues of cost and the willingness of policy makers to invest in education are so critical. The Tennessee class size study is a case in point. This was a long-term, large-scale randomized experiment that showed significant achievement gains for students in smaller classes, and what's more these gains were sustained into later grades even when students were no longer in small classes.

One of the ironies we discovered in studying retention is that school administrators often lack the funds to provide special help to students who are struggling yet they can afford the high cost of retention, which costs about $5,700 per student. Because annual budgets are based on the total number of students regardless of grade level, and no one is held accountable for the cost of 14 years of education instead of 13, school administrators find it difficult to redirect the cost savings from not retaining students to interventions that would be of more direct benefit. In this respect it is important that President Clinton and other politicians say that they want to end both social promotion and grade retention, because it will take attention at the policy level to commit resources for summer school programs, intensive reading help, tutoring, and reduced class size. Unfortunately, when policy makers fall short on their promises, retention becomes the default policy and students suffer.

References:

Byrnes, D. A. (1989). Attitudes of students, parents, and educators toward repeating a grade.

In L. A. Shepard & M. L. Smith (Eds.), Flunking Grades: Research and Policies on Retention. London: Falmer Press. Heubert, J. P., & Hauser, R. M. (Eds.) (1999).

High Stakes: Testing for Tracking, Promotion, and Graduation. Washington, DC: National Academy Press. Holmes, C. T. (1989).

Grade level retention effects: A meta-analysis of research studies.

In L. A. Shepard & M. L. Smith (Eds.), Flunking Grades: Research and Policies on Retention. London: Falmer Press.

Karl Alexander, Johns Hopkins, responds: If there are such statistics, Jerry, I'm not aware of them. There are guides that try to help the parties involved make good decisions I mentioned above, for example, Grant and Richardson's "Retention/Promotion Checklist"-- but I have to say that the lack of good, reliable information on a matter of such importance is a serious concern. In general, these are local responsibilities here in the U.S. we don't have a national curriculum, national standards for teacher training and certification, and certainly not national standards for promotion/retention (as an aside, our emphasis on local standards is vastly different from most of the industrialized world, where typically the curriculum, standards for teacher training and certification, and tests for assessing student competence are developed by a centralized Ministry of Education and standardized throughout the country). Local control of schooling clearly is an important cultural value for us, but this makes it hard to gather good information that is comparable from place-to-place, and hard to implement reforms.

Numerous teacher surveys show that they feel grade retention has its uses and can help-- they don't want to see it eliminated as an option-- and of course it is teachers who are on the front line when we ask them to manage classes in which the reading level might span 3, 4 or even 5 grade levels.

Your question identifies a serious need-- to help teachers who are chronically overburdened-- but I'm not sure we're very far along in finding solutions. Still, there are promising developments. Bob Slavin (also here at Hopkins), for example, has developed a comprehensive classroom reform model that he calls "Success for All." It weaves together age-appropriate curriculum materials (along with intensive teacher training on thieir use), cooperative learning, parental involvement, frequent assessment and feedback on curriculum mastery, and supplemental services, including one-on-one tutoring.

The goal is to help children keep up all along, so that there is less need later for draconian measures like retention. Surely that's what all of us want, but until we get there, some children will continue to struggle and someone has to figure out how best to help them. Simply passing them along regardless of their mastery of the curriculum isn't the answer, but neither is traditional grade retention, at least not for most children under most circumstances.

Go to Headlines@HopkinsHome Page