What If We Ended Social Promotion?
|In its plan to end social promotion, the Clinton administration appears to have mixed a number of fine and credible proposals for educational reform.|
Last year, I chaired a study of appropriate uses of testing for the National Research Council. The NRC panel was a diverse group of 15 scholars from all over the country. We wrote our report, "High Stakes: Testing for Tracking, Promotion, and Graduation," in response to a congressional mandate. The study was prompted by the Clinton administration's proposal, in 1997, for voluntary national tests of 4th grade reading and 8th grade math. The panel took no position about the value of voluntary national testing for its stated purposes--to tell American students, parents, and teachers how well they are doing relative to high national standards--but we recommended strongly against such tests' use for any high-stakes purpose. The report, just published, has a lot of useful information about proper test use, and I commend it to readers. One of the strongest recommendations is that "accountability for educational outcomes should be a shared responsibility of states, school districts, public officials, educators, parents, and students. High standards cannot be established and maintained merely by imposing them on students."
Early in its work, the NRC panel decided to consider whether good tests could serve bad purposes. Thus, we evaluated the consequences of high-stakes educational decisions that may be based, at least in part, on test scores. In particular, we found--as American schools presently operate--that decisions to place students in typical lower-level tracks and decisions to hold students back to repeat the same grade are not educationally sound. Those decisions hurt students, and good tests will not improve them. This is not to say that all forms of tracking are bad for students, or that all grade retention is necessarily bad for students. Our findings were based on the actual and typical, not the ideal. But research evidence based on actual experience should inform new policies.
Here, I may leave some of my colleagues on the NRC panel. Although the facts and ideas are based on the NRC report, I speak for myself about the Clinton administration's proposal to "ban social promotion."
W e should know that a new policy works before we try it out on a large scale. In its plan to end social promotion, the administration appears to have mixed a number of fine and credible proposals for educational reform with an enforcement provision--flunking kids by the carload lot--about which the great mass of evidence is strongly negative. And this policy will hurt poor and minority children most of all.
Everyone is in favor of creating high standards and holding students to them. No one is in favor of social promotion, if that means promoting students who have not mastered the work of one grade and who are not ready for the next. But the question is, "What is the alternative?" Is holding students back in grade--flunking them--good for students? The research evidence shows that it is not.
It makes much more sense to identify learning problems early and to remedy them with solutions that really work--long before the only choices are flunking and social promotion. We know a lot of things that work: smaller class sizes, better-trained teachers, a challenging curriculum, high expectations, after-school and summer school help. There is good evidence that these things work--and good evidence that flunking kids does not work. Why should we tie these good ways of teaching to decisions that hurt children?
Students who have been held back typically do not catch up; in fact, low-performing students learn more if they are promoted--even without remedial help--than if they are held back. One reason for this is that the elementary and secondary school curriculum does not change radically from one grade to the next; there is a lot of review and overlap. Another is that it is simply boring to repeat exactly the same material.
|Well-meaning leaders are pushing a huge national experiment with a policy that has not worked in the past—and without any evidence that it will work in the future.|
Students who have been held back are much more likely to drop out before completing high school. That effect often occurs many years after a student is held back in grade and thus is invisible--without careful longitudinal study--to those who make the retention decision. The teachers and administrators who make decisions to hold children back do not have to live with the long-term consequences of their decisions.
It also costs a lot to hold students back. Children lose a year of their lives when they are held back. Overage students are out of place in classes with younger children--especially in the teenage years--and their presence in the classroom is a problem for teachers and administrators. Flunking kids--repeating grades--is expensive for school systems, even if they do not invest in remediation.
What is the evidence about the effects of retention? And do we really have social promotion? How much retention is there in our schools?
In 1989, Thomas Holmes reported a careful scientific summary of 63 controlled studies of grade retention in elementary and junior high school through the mid-1980s. When promoted and retained students were compared one to three years later, the retained students' average levels of academic achievement were at least 0.4 standard deviations below those of promoted students. In these comparisons, promoted and retained students were the same age, but the promoted students had completed one more grade than the retained students had. Promoted and retained students were also compared after completing one or more grades, that is, when the retained students were a year older than the promoted students but had completed equal numbers of additional grades. Here, the findings were less consistent, but still negative.
Of the 63 studies reviewed by Mr. Holmes, 54 yielded overall negative effects of retention, and only nine yielded overall positive effects. Some studies had better statistical controls than others, but the best studies--those with subjects matched on test scores, sex, and/or socioeconomic status--showed larger negative effects of retention than studies with weaker designs. Mr. Holmes concluded, "On average, retained children are worse off than their promoted counterparts on both personal-adjustment and academ-ic outcomes."
New York City instituted in the 1980s the Promotional Gates program, which combined high rates of retention with efforts at remediation. The remedial efforts soon faded, and eventually the program was dropped. A blue-ribbon panel evaluated the program and found--no surprise--that it reduced achievement and increased dropping out. Today, New York is back on the same track. What reason does Chancellor Rudolph F. Crew have to believe that what failed in the '80s will work now and in the new millennium?
A major new longitudinal study of Chicago children was undertaken by Arthur Reynolds and his colleagues in connection with a successful experiment in early and sustained educational intervention (before the recent so-called reforms in Chicago). They found that "grade retention was significantly associated with lower reading and math achievement at age 14 above and beyond a comprehensive set of explanatory variables" ("Grade Retention and School Performance: An Extended Investigation," Institute for Research on Poverty, 1998). Another analysis of the Chicago data found that retention during kindergarten through grade 8 increased dropping out by 12 percentage points, after controls for social background, program participation, school moves, and special education placement. Several earlier studies of Chicago in the 1980s also found that grade retention increases the number of dropouts.
Douglas K. Anderson in 1994 carried out an extensive, large-scale national study of the effect of grade retention on high school dropout rates. He analyzed data from the National Longitudinal Study of Youth for more than 5,500 students whose school attendance was followed annually from 1978-79 to 1985-86. After extensive statistical controls for sex, race/ethnicity, social background, test scores, adolescent deviance, early transitions to adult status, and several school-related measures, students who were currently repeating a grade were 70 percent more likely to drop out of high school than students who were not currently repeating a grade.
|Low-performing students learn more if they are promoted—even without remedial help—than if they are held back.|
R.W. Rumberger and K.A. Larson analyzed high school dropout statistics and completion of the General Educational Development diploma in longitudinal data for almost 12,000 students in the National Educational Longitudinal Study of 1988 (American Journal of Education, 1998). After statistical controls for social and family background, school characteristics, student engagement, and academic achievement in the 8th grade (test scores and grades), they found that being held back before the 8th grade increased the relative odds of dropping out by the 12th grade by a factor of 2.56. Furthermore, they wrote, "students who were held back before the 8th grade were more than four times as likely as students who were not held back to not complete high school or receive a GED by 1994." Reliable negative evidence of that strength in a clinical trial would lead to its early termination.
Well-meaning leaders throughout the country are pushing a huge national experiment with a policy that has not worked in the past--and without any evidence that it will work in the future. For example, we have good experimental evidence that smaller class sizes in the elementary grades have important, lasting effects on learning; the evidence for banning social promotion is just wishful thinking. There is no credible, large-scale evidence of its success as an educational policy.
T here is one more critical point on which the National Research Council report provides strong evidence: We do not practice social promotion in the United States now, and we have not practiced it for many years. Our statistics are not very good; neither the federal government nor most states collect the right data, but we do know a few things.
Age at entry to 1st grade has increased since 1970. At that time, almost all 6-year-olds were in the 1st grade (about 4 percent of 6-year-old boys and 8 percent of 6-year-old girls were enrolled below the 1st grade). In 1996, 18 percent of 6-year-olds were enrolled below the 1st grade. Part of that change is due to holding children back in kindergarten.
Many students are held back during elementary and secondary school. Nationally, among children who entered school in the late 1980s, 21 percent were enrolled below the usual grade at ages 6 to 8; 28 percent were below the usual grade at ages 9 to 11; 31 percent at ages 12 to 14; and this rose to 36 percent at ages 15 to 17. Not counting kindergarten and the later grades of high school, this means that at least 15 percent of children--and probably 20 percent--have been held back at some time in their childhood.
Worse yet, minorities and poor children are the most likely to be held back. Black, Hispanic, and white children enter 1st grade at just about the same ages, but between entry and adolescence, about 10 percent of white girls fall behind in grade, while 25 percent to 30 percent of minority children fall behind. By ages 15 to 17, 45 percent to 50 percent of black and Hispanic youths are below the expected grade levels for their ages.
Holding students back--flunking them--has a much greater impact on minority and poor youths than on majority, middle-class children. It decreases educational opportunity, and it makes opportunities less equal among groups. For 35 years, American education has aimed to reduce social inequality. While much remains to be done, we have made major gains--narrowing differences in test scores in the 1970s and 1980s and reducing the dropout difference between majority and minority children. If we start holding back ever larger numbers of children, we are likely to reverse the progress of the past four decades.
Robert M. Hauser is the Vilas research professor of sociology in the Center for Demography at the University of Wisconsin-Madison. He is the editor, with Jay P. Heubert, of the National Research Council's report "High Stakes: Testing for Tracking, Promotion, and Graduation" (National Academy Press, 1999).