The association between higher education and political attitudes is well-recognized, and research suggests that socialization amongst peers is one of the most probable mechanisms explaining a possible attitudinal change. However, identifying a socialization effect is difficult due to self-selection mechanisms and limitations in data availability. This article empirically investigates the underlying peer socialization by employing a longitudinal network study of undergraduate students in Swedish and Finnish top-ranked business schools (N = 2651). The study tests the hypotheses that (1) attitudes within a cohort converge over time, and (2) socially embedded students experience more attitudinal change during their studies than students who do not engage with their peers. The paper leverages the Covid-19 pandemic as an exogenous shock that severely affected the networking dynamics of specific cohorts since it moved their education off campus. It is thereby among the first studies to directly test the relationship between socializing behaviors within higher education and political attitudes using both subjective and objective measurements. Contrary to previous conjectures, the findings challenge a general socialization mechanism to foster political attitudes. Instead, they point toward the importance of self-selection mechanisms in studies of higher education and for the specific student composition.