Perceived Ability and School Choices

Abstract

We design and implement a field experiment in which ninth-graders from disadvantaged backgrounds in Mexico City are provided with individualized feedback about their performance on an achievement test. The intervention shifts both the mean and the variance of the subjective distributions of academic ability. Information provision also creates a steeper gradient of the relationship between realized skills and the demand for academically-oriented schools, which in turn affects the skill composition across high-school tracks. Follow-up data on schooling outcomes suggest that the intervention improved the match between students and education careers. The experimental variation in the individual distributions of subjective beliefs is embedded into a discrete choice model that is used to shed light on the channels through which informing students about their own skills shapes both observed and counterfactual sorting patterns across schools.