Given the inherently social context of classrooms, students' beliefs are dependent on the beliefs of their classmates, particularly their classroom friends and popular peers (Ryan & Shin, 2018). I investigated the ways in which students observe and interpret teaching practices in a context of their peers with three studies with the following aims: 1) determine the extent to which classroom friend groups predict changes in student academic engagement and perceptions of teaching practices; 2) determine whether student perceptions of teaching practices are similar to their friends' perceptions. If so, is this similarity related to friends selecting, maintaining, or influencing friends; and 3) determine the extent to which peer descriptive and status norms of perceptions of teaching practices predict changes in individual student perceptions of these practices. The studies use data from the Classroom and Peer Ecologies (CAPE) Project (PI Allison Ryan), a study of 58 elementary to middle school classrooms in the fall and the spring of the school year. We surveyed students about their perceptions of three autonomy-supportive teaching practices (disciplinary harshness/upholding fairness, promoting student voice and choice, and fostering relevance) and three measures of student engagement (emotional engagement, behavioral engagement, and disruptive behavior) in their math and science class. For 48 classrooms, students indicated which classmates they considered friends and whom they considered "really cool" in the fall and the spring. Study I highlighted the relatively minimal role that peer groups play in shaping classmates' perceptions of autonomy-supportive teaching practices but demonstrated that different facets of student perceptions of teaching practices relate to peer groups in distinct ways. There was significant variance between peer groups in disruptive behavior. Peer group perceptions of teaching practices did not affect student engagement. Therefore, there is no evidence that teachers need to worry about leveraging the influence of students' friend groups to shape their perception of the climate. Study II utilized one type of longitudinal social network approach, stochastic actor-based modeling, to understand how changes in friendships from fall to spring coincided with changes in student perceptions of the teaching practices. Different teaching practices in relation to student friendship connections operate in distinct ways. For student perceptions of autonomy support, patterns of peer homophily differed based on observed teacher emotional support. Study III reinforced the idea that classroom-wide characteristics, particularly peer descriptive norms, matter more than the influence of peers with social status in shaping beliefs. The findings across the three studies provide evidence that students choose friends with similar perceptions as themselves and do not substantially influence their classmates' perceptions apart from class-wide perceptions. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]