Differentiation and Inequality in Cross-National Perspective
By Adam Gamoran
A persisting dilemma for educators is finding the best way to arrange students for instruction. In the U.S., students are often divided into separate classes or groups based on their academic performance or interests. Ability grouping (dividing students on a subject-by-subject basis) typically begins as early as first grade, as students are divided into reading groups within the classroom. Similar divisions within or between classes are common throughout elementary and middle school. More rigid forms of tracking (dividing students on a single criterion for all academic subjects), once common in U.S. high schools, are no longer prevalent, but researchers such as Samuel Lucas have shown that ability grouping in high school more or less amounts to the same thing as tracking, as students in high levels in one subject tend to be highly placed in other subjects.
It may seem logical to divide students for instruction in these ways. After all, students do vary in their performance levels, and if teachers are trying to target instruction to meet student needs, they can do so more efficiently when students are more homogeneous in their academic levels. Yet ability grouping is problematic for several reasons. First, we live in a stratified society, and due to conditions outside of school, when we divide students by academic performance, we also tend to divide them by race, ethnicity, and social class. If instructional efficiency is one goal, and bringing together students from diverse backgrounds is another goal, it can be difficult to serve both of these aims.
Second, scores of research studies cast doubt on the notion that ability grouping is a more efficient way of teaching than mixed-ability teaching. Students in high groups tend to learn more than similar students in mixed-ability classes, but students in low groups tend to learn less. Thus, ability grouping generally does not boost productivity in the school as a whole, but it does tend to magnify inequality.
What lessons can we learn from international research on ability grouping? Decades of research on the topic have been conducted in the United Kingdom, and more recent years have witnessed many new studies in other countries as well. Three findings seem clear. First, countries with more differentiated systems tend to exhibit more inequality in student achievement. For example, Eric Hanushek and Ludger Woessman used two international tests to show that in countries that use tracking in the early grades, inequality tends to increase more between the elementary and secondary grades than in countries that delay the onset of tracking. Also, Min-Hsiung Huang found that among countries that participated in the 2003 Trends in International Mathematics and Science Study in grades 4 and 8, inequality in mathematics achievement increased more among countries that used between-class ability grouping in mathematics than in countries that used grouping less extensively.
Second, international research shows that even though the forms of differentiation vary across countries (some divide students between schools, others within schools; some track in many subjects, others track in few; some track earlier and others later; some are more rigid in their tracking systems and others are more flexible), the results of differentiation are largely the same: as in the U.S., dividing students for instruction on the basis of academic performance results in gains for high achievers and losses for low achievers.
Third, a few cases provide examples of how the harmful effects of tracking on low achievers may be mitigated. When standards are clear and closely linked to curriculum and assessments; when low achievers have incentives to perform well on the assessments (e.g., when success provides postsecondary access or job opportunities); and when schools have incentives to promote the performance of low achievers - differentiation of instruction may result in less inequality rather than more inequality. These findings have emerged in countries as diverse as Taiwan and Israel. Also, cases of education reform in both Scotland and Australia - in which the use of tracking was reduced and the academic demands in low tracks was increased - resulted in higher achievement and lower inequality overall. These international examples suggests that differentiation is not inevitably associated with inequality, but rather its effects are context-dependent.
It seems clear from international as well as U.S. research that ability grouping is a problematic practice, but as long as it promotes the outcomes of high achievers it will be difficult to give up entirely. To the extent that ability grouping continues to be used, it must occur in the context of serious incentives for low achievers. In our current system in the U.S., that is rare. A system that truly focused on college and career readiness for all students would provide meaningful rewards for students in less academically advanced programs of study to elevate their levels of performance. Better articulation between secondary school and postsecondary educational and occupational opportunities, and accountability for schools that focused not just on graduation rates but on producing students who can succeed at the next level, would be a good start.
Note: References for these statements, along with a more extensive argument, may be found in my essay, "Tracking and inequality: New directions for research and practice."
Adam Gamoran is the MacArthur Professor of Sociology and Educational Policy Studies and the director of the Wisconsin Center for Education Research at the University of Wisconsin-Madison. His co-edited works include Methodological Advances in Cross-National Surveys of Educational Achievement (National Academies Press, 2002) and Stratification in Higher Education: A Comparative Study (Stanford University Press, 2007).