as a PDF

Review of Educational Research
Fall 1990, Vol 60, No. 3, pp. 501-504
The Effects of Ability Grouping in Secondary
Schools: A Response to Slavin's
Best-Evidence Synthesis
Maureen T. Hallinan
University of Notre Dame
The motivation for this response to Slavin's meta-analysis of ability grouping in
secondary schools is the concern that social scientists, and especially educational
practitioners, might accept without further qualification the conclusions Slavin
reaches and proceed to make decisions regarding practice and policy based on these
conclusions. This outcome, I feel, would be inappropriate and perhaps misguided,
for reasons to be outlined here.
The time is ripe for a review of research on the effects of ability grouping.
Recently, new conceptualizations of the organization of schools have been formulated, renewed interest in the instructional process has occurred, empirical research
on ability grouping has burgeoned, and more and more of this research is being
disseminated to the educational community. It is time to take stock in order to
facilitate the difficult process of applying social science research on ability grouping
to educational practice in schools.
Slavin's review on ability grouping in secondary schools (middle, junior high,
and high schools) parallels his recent meta-analysis of research on ability grouping
in elementary schools (Slavin, 1987). In both reviews, the final conclusion Slavin
reaches is predictable. Based on what he judges to be the best of the available
research on ability grouping, he concludes that ability grouping has no direct effect
on student achievement in English, Reading, or Mathematics. Ability grouping
provides no advantage or disadvantage, compared to heterogeneous grouping, for
student learning.
The fact that the studies Slavin examines show no direct effect of ability grouping
on student achievement is not surprising. The studies compare mean achievement
scores of classes that are ability grouped to those that are not. Since means are
averages, they reveal nothing about the distribution of scores in the two kinds of
classes. Ability grouping may increase the spread of test scores while leaving the
mean unchanged. This would occur if the practice had a differential impact on
students with different abilities. Since teachers generally gear instruction to the
ability level of the students being taught, students in a high ability group are likely
to receive more and faster instruction and those in low ability groups less and
slower instruction than pupils in an ungrouped class where instruction is geared to
the average of the class. If greater gains of high achievers balance lesser gains of
slow students in a grouped class, there should be no overall impact on the mean
achievement of the class, compared to a heterogeneous class, even though the
variance of the test scores in the two classes may differ markedly. Studies comparing
only mean would show no direct effect of grouping on achievement.
501
Downloaded from http://rer.aera.net at PENNSYLVANIA STATE UNIV on March 5, 2016
Maureen T. Hallinan
What is more important than observing no main effect of ability grouping on
achievement is the implication of this finding. Simply stated, the conclusion suggests
that ability grouping has no consequence for educational practice. It identifies
neither grouping practice as superior to the other in fostering learning. Interpreting
the results this way could lead educators to rely on other rationales in deciding
whether or not to employ ability grouping in their schools. The risk here is that of
ignoring a possible differential impact of grouping on different students which
could lead to significant inequities in educational practice.
Slavin's second conclusion is more astonishing. He claims that the studies he
reviewed show that ability grouping provides no advantages for students assigned
to high ability groups and no disadvantages for students assigned to low ability
groups. He reaches this conclusion despite the wealth of social science research that
purports to show otherwise. Given the controversial nature of Slavin's conclusion,
the supporting evidence he presents deserves close scrutiny.
Slavin reviews 21 studies, none of which, according to the meta-analytic technique he uses, show a statistically significant differential effect of ability grouping
on students by ability level. Three comments are in order regarding this finding.
First, and perhaps most important, the studies Slavin cites ignore a fundamental
mechanism that would predict and explain differences in ability group outcomes—
namely, the instructional process. The research he cites simply compares outcome
measures of achievement across high and low ability groups. Yet, the most fruitful,
recent line of research on ability grouping argues that the main mechanism
producing differential effects of ability grouping is the instructional process (e.g.,
see Barr and Dreeben, 1983; Gamoran, 1986). None of the studies referred to by
Slavin takes into account either the content and pace of instruction or the pedagogical practice. The research systematically ignores instructional and curricular
differences across classes. This is a fatal flaw in studies aimed to evaluate grouping
effects. As a result, the studies Slavin reviews are unable to provide a real test of
whether the learning process differs significantly across ability grouped and heterogeneously grouped classes, because the fundamental cause of such differences has
been ignored.
Second, Slavin's studies rely almost solely on standardized test scores to measure
achievement. This outcome measure has well-known limitations. Standardized
tests are not adequate measures of what students are taught in school. They can be
viewed more accurately as tests of general ability or intelligence rather than of
mastery of the curriculum. Failure to observe differences in standardized test scores
across students is a poor measure of grouping effects. Moreover, the growing body
of research showing cultural and gender bias in standardized tests suggests that they
are a particularly inappropriate measure of learning when comparing students who
differ in social origins. Since students' background characteristics often correlate
with assignment to a high or a low ability group, reliance on standardized test
scores confounds the results. The number of studies in Slavin's sample that use
other measures of achievement is too small to permit separate analysis. In general,
Slavin's conclusions are based on a limited andflawedmeasure of student learning.
Finally, Slavin's selection of studies is skewed heavily in favor of experimental
research. There are only a few surveys in his sample, and there is a complete
absence of case studies. Slavin excludes case studies by requiring that at least three
ability grouped and three control classes be in an analysis for inclusion in his
502
Downloaded from http://rer.aera.net at PENNSYLVANIA STATE UNIV on March 5, 2016
Response to Slavin 's Best-Evidence Synthesis
sample. In so doing, he disregards important field work, such as that by Oakes
(1985), Rosenbaum (1976), and others. Their work shows distinct differences in
instructional techniques, teacher interactions, reward systems, student motivation,
effort and self-esteem, student behavior, disciplinary measures, administrative load,
role modeling, and peer influences by level of ability group. It is difficult to believe
that these dramatic findings are not related to differential learning patterns across
ability groups. It may be that the design of some of the experimental studies Slavin
examines hides the richness of the learning process—a complexity that is better
detected by more in-depth studies.
Of the fourteen surveys that Slavin examines, three were conducted before 1940.
Only four studies were conducted since 1970, when researchers began using more
sophisticated multivariate models in survey research. Many of the earlier studies
were correlational; they simply examined the bivariate relationship between ability
grouping and achievement without statistically controlling for other factors. One
must be very cautious in interpreting the results of these studies lest spurious
relationships be given scientific importance or the failure to observe bivariate
relationships be interpreted as the absence of a causal process.
It is not clear why Slavin did not include more survey research in his metaanalysis. It may be that his criteria for inclusion biases his sample toward experimental research by requiring a control and treatment group. Or, it may be that a
number of recent survey analyses are excluded because they examine only differential effects of ability group level on achievement without also comparing grouped
and ungrouped classes. Moreover, much of the recent survey research conducted
by sociologists focuses on the effect of course selection and completion rather than
on ability grouping or tracking on achievement gains. The omission of more survey
research is unfortunate since much of the support for the proposition that ability
group favors high ability students and disadvantaged low ability pupils has been
found in this work as well as in smaller, in-depth field studies. Again, one is tempted
to conclude that the skewness of the sample toward experimental research permits
only a partial picture of the impact of ability grouping on student learning.
Slavin places considerable emphasis on the experimental studies in his sample.
Examining this research shows that, of the six randomized experiments, four had
a duration of a year or more while the other two were only a semester in length.
Of the nine matched experimental studies in Slavin's meta-analysis, seven studies
had a duration of a year, and the other two had only a semester. Three of the four
year-long randomized experiments examined Reading and Language scores, two
looked at Social Studies, and one Mathematics. Of the seven matched experiments,
four investigated Mathematics differences, two English, two Language, one Reading, two Social Studies, and one Biology. Several grade levels, ranging from fifth
through twelfth, are represented in the sample. Most of the experimental studies
are old, with twelve of the fifteen being conducted before 1970 when the practice
of ability grouping may not have been implemented in the same way that it is
today. One must question the validity of each of these strategies: pooling results
across subjects, pooling across grade levels, and relying heavily on old research.
Slavin draws rather firm conclusions from this limited sample.
In sum, it seems that Slavin's conclusions, at least with respect to the impact of
ability grouping on students assigned to different level groups, rest on shaky ground.
Careful systematic research is currently underway by several social scientists ex503
Downloaded from http://rer.aera.net at PENNSYLVANIA STATE UNIV on March 5, 2016
Maureen T. Hallinan
amining the interaction between the organizational practice of ability grouping and
the instructional process as it affects student opportunities to learn and academic
achievement. This research promises to be far more fruitful in addressing the
complex issues related to the effectiveness and equity of ability grouping than
previous studies that were limited in conceptualization, scope, and methodology.
References
Barr, R., & Dreeben, R. (1983). How schools work. Chicago: University of Chicago Press.
Gamoran, A. (1986). Instructional and institutional effects of ability grouping. Sociology of
Education, 59, 185-198.
Oakes, J. (1985). Keeping track: How schools structure inequality. New Haven, CT: Yale
University Press.
Rosenbaum, J. E. (1976). Making inequality: The hidden curriculum of high school tracking.
New York: Wiley.
Slavin, R. E. (1987). Review of Educational Research, 57, 293-336.
Author
MAUREEN T. HALLINAN is Professor of Sociology, 400 Decio Hall, University of Notre
Dame, Notre Dame, IN 46556. She specializes in sociology of education.
504
Downloaded from http://rer.aera.net at PENNSYLVANIA STATE UNIV on March 5, 2016