How Good Am I? Effects and Mechanisms Behind Salient Ranks

Abstract

How can individuals respond to their ordinal ranking when they are not aware of it? We present evidence on the effects and mechanisms of achievement rank effects in middle schools when ranks are salient to students and their parents. For identification, we rely on the random assignment of students (and teachers) to classrooms in China. That is, students with the same baseline test scores end up having different achievement ranks in their assigned classroom. We find positive and large effects of being assigned a higher rank on subsequent performance, especially for males and overconfident students. We show that students with higher ranks spend more hours on autonomic studying. What drives these effects is still an open question, especially when ranks are salient to both students and their parents. Using rich survey data, we show that these academic gains are not only mediated through (1) students' higher self-perception and higher subject learning confidence, but also through (2) better parental understanding of their child's ranks, stricter parental requirements for their child's study, and higher parental expectations regarding their child's educational attainment and career prospects. We show that these two channels make similar contributions to explaining salient rank effects, and when combined they explain 46.80% of the increase in test scores. We find no impact on teachers' investment or attention to students as a result of rank effects.

Keywords: achievement rank, saliance, quasi-random classroom assignment, mechanisms, survey data, middle schools, mediation analysis

JEL Classification: I21, J24