The thesis focuses on the influence of friends on multiple measures of students' educational aspirations in Hungarian primary schools before students are distributed to stratified tracks. The dissertation discusses and examines various mechanisms through which friends can influence students' aspirations, such as the adaptation to norms, values, and attitudes shared by friends and the potential effect of having access to the educational resources of friends' parents. The dissertation also considers the contribution of friendship selection to friends' similar aspirations. 1. The first empirical chapter of the dissertation examined the effects of social influence and selection on academic ambitions and achievement in two school subjects: Hungarian literature and mathematics. Using longitudinal social network models, the analysis disentangled social influence on academic ambitions from friendship selection based on academic ambitions. The study found that changes in adolescents' academic ambitions were related to the ambitions of their friends. Nevertheless, the mechanisms underlying the tendency towards similarity were slightly different for the two subjects. In Hungarian literature, the attractiveness of friends with high ambitions was more pronounced than that of friends with low ambitions, while the opposite process was observed in mathematics. Academic achievement seemed to be a relevant attribute for friendship selection, but only in Hungarian literature. 2. The second empirical study investigated the influence of friends on the secondary school track preferences of Hungarian adolescents. The results indicated that peers could influence the adjustment of adolescents' educational preferences through their parental resources, as having friends with at least one parent with a tertiary education could motivate students to adjust their preferences towards the grammar school track. The average preferences of friends did not have a direct positive effect on preferences for the grammar school track, while having friends who preferred the vocational secondary school track had a negative effect on students expressing the same preferences. 3. The third empirical study examined whether students' preferences for the grammar school track in their secondary school applications were associated with the preferences of their friends and classmates by applying Generalized Structural Equation Models. The results indicated that students' own preferences for the grammar school track were negatively associated with the proportion of friends and classmates who preferred the grammar school track in their applications. Friends' previous preferences were positively associated with students' preferences in their applications, but only when friends' preferences in their applications were also taken into account. On the whole, the results did not provide sufficient support for the positive association between friends' educational preferences. 4. Overall, the results suggest that friends' aspirations may have a positive effect on aspirations regarding school grades, i.e., aspirations that have immediate consequences and are not bound by external limits. The differences in social influence and social selection between the school subjects studied underscore the importance of studying academic motivation as a domain-specific concept. With regard to aspirations expressed as preferences for secondary school tracks, the negative effects may be explained by perceived constraints that prevent or encourage individuals from having certain aspirations or dreams for themselves. The contribution of friends' parental background to the development of students' educational preferences highlights that schools are social institutions where social networks play an important role alongside formal education.