Peers relations have been regarded as a powerful developmental context that can influence multiple aspects of children's functioning and well being (Rubin, Bukowski, & Parker, 2006). Given that to a great extent these peer interactions take place within classrooms, this particular social context is an essential domain for the study of children’s socialization.  The present dissertation sought to further our understanding of the multiple effects that the classroom may have on children’s social development, particularly for antisocial behaviour. Through a series of three interrelated studies, the classroom was examined as a fundamental context of experience and measurement, as a predictor of individual development, and as dynamic system that changes over time. These issues were addressed in a sample of 1594 Colombian children, who were assessed at four time points during a school academic year. The first study (Chapter 1) describes a statistical technique that capitalizes on the valuable information that peer-nomination-based assessments provide about individual members of the classroom, while ensuring the reliability and validity of the measures. Particularly, a regression-based technique that corrects scores obtained from sociometric and peer assessments to prevent the risk of bias due to differences in classroom size (i.e., number of nominators) is proposed and examined. The advantages of the regression-based technique over the widely-used standardization technique, in terms of practicality, efficiency, and interpretability are presented. The second study (Chapter 2) examined the effect of classroom cohesion and aggression-related norms on changes in individuals’ aggressive behaviours across the school year. Results from this study revealed that classroom norm salience predicted higher and increasing levels of students’ aggression. In addition, it was found that classroom cohesion prevented direct forms of aggression, while it enhanced conformity to aggression norms in girls. Finally, the third study (Chapter 3) explored how classroom norm salience changes over the course of the school year and to what extent teachers’ cognitions and behaviours explain such changes. Evidence indicates that teachers’ support prevented classroom norm salience from increasing over time. Results from these studies are discussed in light of their practical implications for interventions directed at preventing aggression and promoting social competence in the school context.