Peer effects in the elementary school classroom

While it is assumed frequently that children who spend time with aggressive or prosocial peers will learn eventually and adopt these forms of social behavior, there is limited evidence to support a causal impact of peers' behaviors and attributes on individual outcomes. To date, important methodological limitations have prevented researchers from distinguishing the causal impact of peer attributes on developmental outcomes from the selection processes that motivate similar individuals to cluster together. In this study, I employ the instrumental-variables approach described by Bramoulle, et al. (2009), which capitalizes on the availability of detailed social-network data among classroom peers, to estimate the causal impact of "hanging around" aggressive and prosocial peers on the development of anti– and pro–social behavior of elementary-school children. In addition, I explore the extent to which individual differences in aggressive and prosocial behavior influence their future academic skills. I test these effects using data collected as part of the New York City Study of Social and Literacy Development, from approximately 1200 3 rd to 5th grade students, nested in 18 elementary schools. I find support for the existence of average peer effects for prosocial and aggressive behavior, with similar effect sizes, in this population, acting simultaneously on the behavior of elementary school-aged children. Additionally, I find that for a segment of the population -- the girls -- higher levels of both aggressive and prosocial behavior are associated with higher ratings of their academic skills one year later. I conjecture that this unexpected finding, evidenced exclusively for girls, speaks directly to the differential functions that different forms of social behavior may have, within the classroom context during the elementary school years.