This was supported by the significant interaction effect between actors and partners, which explained unique variance of the frequency of peer interaction. Moreover, the supplementary analysis indicates that the homophily of language and literacy skills appeared later in the school year. This result is compatible with a peer socialization effect. Perhaps as children’s language and literacy skills develop, their language and literacy skills becomemore like theirpeers’ so thatthey are able to engage in better communication and interactive literacy activities with them. However, the peer socialization account is only one of the possible interpretations for the current finding. A longitudinal study is needed to further examine the underlying causal mechanism in the future. Previous studies focusing on children’s language and literacy skills suggest that such emergent language and literacy skills are closely related to individuals’ social behaviors and relationships with peers (e.g., Doctoroff et al., 2006; Mendez et al., 2002). The current study extends previous findings in three ways. First, our study incorporated both individual-level and dyad-level factors in the models, which allowed us to compare the relative contributions of individual and dyadic attributes to peer interactions. Our findings suggest that peer interaction was more associated with actor’s and partner’s language and literacy skills rather than individuals’ skill levels alone. Second, the supplementary analysis of the pre- and post-test language and literacy skills suggests that the developmental trajectory of children’s language and literacy T.-J. Lin et al. / Early Childhood Research Quarterly 37 (2016) 106–117 115 skills can be shaped by socializing these skills with peers. Third, our findings are consistent with the peer effects literature that has predominantly focused on urban preschool children (e.g. Delay et al., 2016).