Individual Submission Summary

Direct link:

Compositional features of elementary school classrooms and students’ social-emotional and academic skills

Thu, March 21, 12:30 to 1:45pm, Baltimore Convention Center, Floor: Level 1, Exhibit Hall B

Integrative Statement

Systems theories emphasize the role of contexts in shaping human development. For elementary school-aged children, classroom composition is a key contextual feature, determining the characteristics of peers with whom children interact daily. Social and behavioral compositional features have been linked to individual children’s social-emotional development (Thomas et al., 2011; Yudron et al., 2014); less is known about such compositional influences on academic skills. Demographic composition, particularly racial/ethnic diversity, has often been associated with children’s academic skills (Orfield et al., 2014), with mixed results regarding associations with children’s social-emotional development (Graham et al., 2014; Juvonen et al., 2017; Seaton & Yip, 2009). Supportive classroom interactional processes may promote positive development in classrooms with high levels of behavioral problems or racial/ethnic heterogeneity (Denessen et al., 2011; Verkuyten, 2008). This study examines associations between social/behavioral and demographic classroom composition and students’ aggression, internalizing problems, social competence, and academic skills. It also will examine moderation of these associations by classroom interactional quality and teacher burnout.

Data come from a cluster-randomized controlled trial of a social-emotional learning and literacy program in 60 NYC public elementary schools. Participants were 1,916 3rd- and 4th-grade students in 131 classrooms in 27 schools (cohort 1) (see Table 1). In winter and late spring 2016, teachers reported on students’ social competence (α=.93; CPPRG, 1999), aggressive behavior (α=.95; Reynolds & Kamphaus, 1998), and literacy skills (α=.96; ECLS-K). Students reported on their own aggressive behavior (α=.83; Orpinas & Frankowski, 2001), anxiety, and depressive symptoms (α=.82; Reynolds & Kamphaus, 1998). Classroom racial/ethnic diversity was calculated based on proportions of enrollment of Hispanic, Black, White, Asian, and Other students using Simpson’s index (Simpson, 1949). Classroom poverty level was calculated as the proportion of students eligible for free/reduced-price lunch.

Multilevel modeling (HLM v7.03) was used to predict students’ self-reported aggression, anxiety, and depression, and teacher-reported social competence and literacy skills in spring from average classroom-level student-reported anxiety and depression and teacher-reported aggression, and Simpson’s index in winter, controlling for winter scores on each outcome. Covariates included child gender, classroom poverty level, and school random assignment status.

Higher classroom-level aggression predicted decreased literacy skills (B = -0.20, p = .05) and decreased social competence at the trend level (B = -0.16, p < .10), even after controlling for winter scores on outcomes and other compositional measures. Higher classroom-level depression predicted increased student-level anxiety (B = 0.18, p < .01), but higher classroom-level anxiety predicted decreased student-level anxiety (B = -0.21, p < .001). Higher classroom-level anxiety during winter standardized test prep and lower student-level anxiety after spring testing might explain the inverse relationship. Greater classroom racial/ethnic diversity predicted decreased child-reported depression (B = -0.11, p < .05) (see Table 2).

Results suggest that both social/behavioral and demographic classroom compositional features influence students’ social-emotional and academic development, with evidence of cross-domain effects. Data on all 60 schools (2 cohorts) and tests of moderation by classroom quality and teacher burnout will be included in the presentation. Findings have critical implications for preventive efforts and school policies and practices.


©2020 All Academic, Inc.   |   Privacy Policy