The structure of classroom social networks predicts educational outcomes

Humans are deeply social, and in early life, classroom friendships shape academic and socio-emotional development. Yet, research has only begun to explore how social network structures are related to classroom-level outcomes. Using a large-scale data set of over 46,000 students in nearly 2,000 classrooms, we extracted 313 social network metrics from four different network types (i.e., friendship, rejection, help, and break-time contacts) and applied machine learning to predict various classroom-level educational outcomes, including academic achievement, motivation, and social integration. Classroom networks predicted both academic achievement and well-being, beyond school tracks, socioeconomic status, and demographics. The predictive power for different educational outcomes was highly dependent upon the type of network and categories of network metrics. These insights position social networks as a powerful tool for understanding educational environments and informing network-based interventions.