Sense of Belonging and Engagement in Racially and Ethnically Diverse Classrooms: The Role of Classroom Peers

• In Event: Poster Session 09
  In Poster Session: PS 09 Topic - Social Relationships

Abstract

Although evidence suggests that students benefit from racially diverse educational settings (Mickelson, 2015), the extent of cross-ethnic engagement within these settings, and its implications for students’ academic development, is not well understood (Orfield & Frankenberg, 2014). Drawing upon a social capital perspective (Coleman, 1988), the present study examines the extent to which early adolescents from historically underserved racial/ethnic groups may face inequitable opportunities to learn from and with peers in the classroom, that may contribute to academic disparities. We hypothesize that greater inclusion in the classroom’s “academic collaboration” network will be associated with greater feelings of belonging and support and greater engagement in middle school classrooms. We further hypothesize unique roles of same- and cross-group relationships, particularly for youth of color (Williams & Hamm, 2017).
We examine data from 915 students (50.1% female) across 53 racially and ethnically diverse classrooms from two middle schools. Students identifying as White (24.30%) or Latinx (26.10%) made up the largest proportion of our sample, followed by Black (18.90%), Multiracial (18.11%), Asian (6.86%), Arab American (4.50%), Native American (0.56%), or another race/ethnicity (0.67%). Students from historically underserved groups were defined as those who identify as Black, Latinx, Native American, or as multiple race/ethnicities that included any of these categories. Surveys asked students whom they like to work with on class assignments, and these peer nominations were used to compute students’ overall indegree centrality (number of nominations received; Freeman, 1979) as well as their separate cross- and same-racial/ethnic group indegree centrality. Students also reported on their sense of belonging and peer support in the classroom (Goodenow, 1993), and their emotional and behavioral engagement (Midgley et al., 2000). In a series of linear regression models, we predicted each of these academic outcomes from either students’ overall centrality, cross-group centrality, or same-group centrality, controlling for gender, grade, and classroom racial/ethnic composition. We also tested for moderation of these associations by a binary indicator of membership in a historically underserved racial/ethnic group, and examined simple slopes within each group.
Overall, findings were consistent with expectations that students from historically underserved racial/ethnic groups were significantly less “well connected” (lower overall centrality) in the classroom academic network (t(868) = 1.77, p = .039). Moreover, findings suggest these differences matter for students’ academic development. Greater overall centrality in the academic network was significantly associated with greater sense of classroom belonging (β = .25, p < .001) and peer support (β = .22, p < .001) for all students, and greater emotional (B = .08, p = .001) and behavioral engagement (B = .04, p = .034) for historically-underserved students. While centrality among same-group peers was important for sense of belonging among youth from historically underserved groups (B = .09, p = .040), centrality among cross-group peers predicted these students’ behavioral engagement (B = .05, p = .016), and was associated with greater sense of belonging (β = .19, p = .001), peer support (β = .19, p < .001), and emotional engagement (β = .11, p < .038) for all students.

Authors

• Lauren Molloy Elreda, University of Virginia - Charlottesville
  Presenting Author

• Haley E. Johnson, University of Virginia - Charlottesville
  Non-Presenting Author

• Joanna Lee Williams, Rutgers University - New Brunswick
  Non-Presenting Author

• Patrick Tolan, University of Virginia - Charlottesville
  Non-Presenting Author