Ethnic Composition and Heterogeneity in the Classroom: Measurement and Relationship With Student Outcomes

• In Event: 49.034 - Student Heterogeneity in the Classroom: Precursors and Links to Outcomes

Abstract

As societies become more diverse in terms of ethnic background, the composition of the student body within educational systems diversifies as well. Various theoretical accounts commonly assume a negative relationship between the proportion of ethnic minority students and student achievement due to school resources, instructional quality, language spoken with peers, and learning culture (Driessen, 2002; Goldsmith, 2011; Raudenbush, Fotiu, & Cheong, 1998; Stipek, 2004). Research on the degree to which this ethnic makeup of schools or classrooms affects student achievement shows mixed results: The proportion of ethnic minority students in a school or classroom often has no or slightly negative effects on student achievement (Mickelson et al., 2013; Van Ewijk & Sleegers, 2010). Yet, some studies report that a higher proportion of heterogeneous minority students may lead to higher achievement (e.g., Benner & Crosnoe, 2011; Peetsma, Van der Veen, Koopman, & Van Schooten, 2006; Gurin et al. 2003; Tam & Bassett, 2004), as students in such learning environments encounter and have to work through contradictions and discrepancies in everyday life, which helps them to expand their intellectual capacities. These mixed results may be due to the operationalization of the ethnic composition: Most studies so far have only distinguished between ethnic minority and majority without addressing ethnic heterogeneity.

The present study explores various measures of ethnic composition and heterogeneity, such as Simpson’s D, Shannon’s H, and Kvalseth’s OD, used in different disciplines as a key to understanding the inconsistent findings (see Budescu & Budescu, 2012; McDonald & Dimmick, 2003). Our aim is to find out whether measures of ethnic diversity are related to student achievement and psychosocial outcomes over and above commonly investigated characteristics of classroom composition (i.e., average cognitive abilities, average socioeconomic status). Our multi-level analyses are based on data from a representative sample of 19,457 elementary school students in western states of Germany in 908 4th grade classrooms (Stanat, Pant, Böhme, & Richter, 2012). Student achievement outcomes in reading and mathematics were assessed with standardized tests. In addition to student achievement, we examine the effects of ethnic heterogeneity on students’ sense of belongingness to their classmates. This variable was assessed with a 4-item-scale in the student questionnaire.

Multilevel structural equation models revealed that the proportion of minority students in the classroom showed the strongest association with student outcomes (reading: beta = -.29, p < .01; mathematics: beta = -.36, p < .01; belongingness: beta = -.28, p < .01). After including measures of ethnic diversity in addition to the proportion of minority students, we found different patterns of results: While ethnic diversity in the classroom was not significantly related with reading achievement (beta = -.06, p > .05) or with sense of belongingness to classmates (beta = .05, p > .05), students showed higher mathematics achievement in ethnically more diverse classrooms (beta = .23, p < .01). The presentation will discuss possible explanations for the differential findings as well as implications of using different measures to describe the ethnic makeup of classrooms.

Authors

• Camilla Rjosk, Humboldt University Berlin

• Dirk Richter, University of Wuppertal

• Oliver Lüdtke, Humboldt University

• Petra Stanat, Humboldt University - Berlin

• Jacquelynne Eccles, University of California - Irvine