Individual differences and dyadic processes in conversations with peers in middle childhood

The goal of the current study was to investigate the contribution of both trait-like individual differences and dyadic processes to the content of children's conversations. Fifty-two groups typically consisting of four same-sex unfamiliar nine-year-old children (N = 202) interacted in all possible dyads, resulting in six dyads per group. Each dyad completed a 5-min frustration task and a 5-min planning task. Observers coded children's verbalizations into 10 categories and further summed these categories into prosocial (suggest, agree, solicit input, ask, encourage, state personal) and antisocial (command, disagree, discourage, aggress) verbalizations, resulting in 24 variables (12 per task). Across both tasks, Social Relations Model analyses provided evidence of the role of both individual differences [significant effects for actor variance (15 of 24 variables), actor-actor correlations, and intrapersonal correlations] and dyadic processes [significant effects for partner variance (4 of 24 variables), relationship variance (18 of 24 variables), dyadic reciprocity correlations (10 of 24 variables), and interpersonal correlations] in children's conversations with peers.