Peer Network Dynamics in Linguistically Diverse Middle School Classrooms: Associations with Students’ Academic Development

• In Event: 1-190 - The Role of Peers in Learning across Distinct Life Stages

Integrative Statement

Network approaches to understanding classrooms point to a specific set of classroom network-level dynamics theorized to have implications for student outcomes. For instance, classroom peer networks that are more densely interconnected and with social connections that are relatively equally distributed throughout the network (versus hierarchically distributed) are thought to provide “equitable” access to and exchange of resources among students, and to promote a sense of belonging and “mattering” in classroom communities (Cappella et al., 2013; Neal, 2014). In the present study, we examine each of these network structural features as well as what we view as a critical third dimension of “equitable” classroom networks: the extent to which the network is integrated across language boundaries (versus segregated by English Learner (“EL”) status). We hypothesize that students in classroom peer networks with greater density, and a more egalitarian and linguistically integrated network structure, will show greater across-year growth in academic outcomes.

We examined data from 501 students (30% designated as ELs; Mage = 12.41 years; 51% female) across 29 linguistically diverse math and English classrooms from 2 middle schools. In fall and spring, teachers rated each student’s class participation on a scale of 1 (very weak) to 5 (very strong). In addition, we collected students’ course grades each quarter, and standardized test scores from the end of the study year and previous year. Social network surveys administered in the spring asked about students’ academic interactions (who they go to for academic help, and who they frequently work with), and social relationships (close friendships, who they “hang out” with outside of class). These data were used to quantify the density (Wasserman & Faust, 1994), unequal distribution (classroom-level SD of students’ indegree centrality; Hanneman & Riddle, 2005), and segregation by EL status (Freeman, 1972) of academic and social peer relationships in the classroom. Associations between these network features and students’ academic outcomes (controlling for baseline academic scores and demographic variables) were assessed in a series of multilevel models. We also tested interactions of the network variables with EL status.

As expected, findings suggest unique roles of each of the network structural features in contributing to student outcomes. EL students in more densely connected academic networks showed relatively greater across-year growth in classroom participation (=0.16, p<.001); and students (both EL and non-EL) in more egalitarian academic networks showed greater growth in course grades (=-0.19, p=.009). In addition, students in more socially integrated classrooms showed greater growth in standardized test scores (=-.44, p=.039).
Together, findings highlight the importance of the peer network structures within which youth are embedded; the number of ties and ways in which those ties are distributed within a classroom may be particularly important within linguistically or culturally diverse classrooms. As such, this study suggests the importance of improving our understanding of the conditions which best facilitate the development of these equitable classroom network dynamics (e.g., Carhill-Poza, 2017; Dryden-Peterson, 2010), and call for consideration of ways in which teachers can play a role in fostering these dynamics (e.g., Farmer et al., 2011).

Authors

• Lauren Molloy Elreda, University of Virginia
  Presenting Author

• Amanda Kibler, Oregon State University
  Non-Presenting Author

• Yin Gu, University of Virginia
  Non-Presenting Author