This paper provides causal evidence on the effects of attending better schools on students' academic performance, self-perceptions, aspirations, and subsequent schooling choices. Students in Mexico City are assigned to public middle schools based on the results of a placement exam. The morning shift in double-shift schools is often oversubscribed, leading to the stratification of school shifts by test score results. Using the cut-off scores for the morning shifts as exogenous variation in school offers, I find that students who have the opportunity to attend more selective schools experience small improvements in standardized language scores but perform worse on non-standardized school-based assessments. These students have lower grade point averages and are less likely to pass all the courses required to complete middle school. I provide evidence to suggest that students evaluate themselves based on their relative performance: marginally admitted students report feeling academically inferior to their peers, have lower self-reported perseverance and time management scores, and are more likely to shift their aspirations and subsequent schooling choices from academic to vocational programs. These findings highlight the importance of frames of reference in educational settings and the potential trade-offs between a better school and a worse position in the ability distribution.