Ability grouping is one of the most controversial issues in educational research (Slavin, 1987). In theory, proponents of ability grouping emphasise that it provides instruction that is specific to the students’ needs. Instruction for high achievers takes place on a higher level and with a faster pace. Low achievers get more individual attention and time for repetition. Opponents of ability grouping, however, focus their arguments on the low achievers: In homogeneous classes they often experience a lower quality of instruction and have less experienced teachers (Oakes, 1985). By ability grouping, the low achievers lack a positive reference group and are deprived of the stimulation by high achievers. Moreover, ability grouping leads to social and ethnic disparities (Rosenbaum, 1980).

Most of the studies on compositional effects focus on the average achievement in learning groups. Only a few studies have yet analysed the effect of the variation of achievement within classes. If tracking has negative effects on attainment, we assume that a greater variation of achievement within learning groups – as it would be the case when teaching students of different abilities in the same classroom – could enhance individual achievement. Luyten & van der Hoeven-van Doornum (1995), however, find that classroom heterogeneity exerts a modestly negative effect on individual achievement for both high- and low-ability students. Dar & Resh (1986) report that the gain of low-ability students in heterogeneous classes outweighs the high-ability students’ decrease of achievement.

In Germany, students are allocated to different school tracks at a very early age. Though this allocation should only be based on achievement, it is strongly connected with the students’ social and cultural background. As a result, students with the same achievement are allocated to different secondary tracks. In consequence, there is a greater variation of achievement within as well as a greater overlapping of achievement between the different tracks than actually intended by this kind of ability grouping. Thus, even in a tracked school system as in Germany, classes are heterogeneous with regard to achievement.

For the German context, Lehmann (2006) finds a considerably high and positive effect of heterogeneity on individual reading achievement. However, he operationalises heterogeneity only by the standard deviation of cognitive ability on the class level, but not by the variation of prior knowledge. As he focuses on primary students, his analyses do not provide information about the effect of heterogeneity in the tracked secondary school system. Moreover, it is not obvious if heterogeneity has positive effects both on high- and low-ability students.

We take up these analyses and focus on three questions: Does heterogeneity as a compositional parameter of classes in secondary schools affect individual achievement at all? Do students learn more in homogeneous or heterogeneous classes? Can both high- and low-ability students equally benefit from heterogeneous classes? According to the previous research, we assume that a greater heterogeneity on the class level positively affects individual achievement. Because of the comparison with high-ability students, low-ability students are expected to reach a higher achievement in heterogeneous classes than in homogeneous classes.

Dar, Y., & Resh, N. (1986). Classroom intellectual composition and academic achievement. American Educational Research Journal, 23(3), 357-374. Lehmann, R. H. (2006). Zur Bedeutung der kognitiven Heterogenität von Schulklassen für den Lernstand am Ende der Klassenstufe 4. In A. Schründer-Lenzen (Ed.), Risikofaktoren kindlicher Entwicklung. Migration, Leistungsangst und Schulübergang (pp. 109-121). Wiesbaden: VS Verlag für Sozialwissenschaften. Luyten, H., & van der Hoeven-van Doornum, A. (1995). Classroom composition and individual achievement. Effects of classroom composition and teacher goals in Dutch elementary education. Tijdschrift voor Onderwijsresearch, 20(1), 42-62. Oakes, J. (1985). Keeping track. How schools structure inequality. New Haven: Yale University Press. Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models. Applications and data analysis methods. Thousand Oaks: Sage Publications. Rosenbaum, J. E. (1980). Social implications of educational grouping. Review of Research in Education, 8(1), 361-401. Slavin, R. E. (1987). Ability grouping and student achievement in elementary schools: A best-evidence synthesis. Review of Educational Research, 57(3), 293-336.