Working Paper: NBER ID: w24958
Authors: Richard Murphy; Felix Weinhardt
Abstract: This paper establishes a new fact about educational production: ordinal academic rank during primary school has long-run impacts that are independent from underlying ability. Using data on the universe of English school students, we exploit naturally occurring differences in achievement distributions across primary school classes to estimate the impact of class rank conditional on relative achievement. We find large effects on test scores, confidence and subject choice during secondary school, where students have a new set of peers and teachers who are unaware of the students’ prior ranking. The effects are especially large for boys, contributing to an observed gender gap in end-of-high school STEM subject choices. Using a basic model of student effort allocation across subjects, we derive and test a hypothesis to distinguish between learning and non-cognitive skills mechanisms and find support for the latter.
Keywords: Ordinal Rank; Educational Production; Gender Gaps; STEM Subjects
JEL Codes: I21; J24
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Ordinal rank in primary school (A21) | Age-14 test scores (I24) |
| Ordinal rank in primary school (A21) | Age-16 test scores (C12) |
| Ordinal rank in primary school (A21) | Subject choice for A-levels (M00) |
| Ordinal rank in primary school (A21) | Self-reported confidence in subjects during secondary school (I21) |
| Ordinal rank in primary school (A21) | Noncognitive skills (G53) |
| Ordinal rank in primary school (A21) | Big fish little pond effect (C92) |