Physical appearance and peer effects in academic performance

ABSTRACT

A large literature examines the role of peer effects in shaping student academic outcomes. This article adds to that literature by introducing a new kind of peer effect – the effect of classmate physical appearance. We document that college students are assigned higher grades when in a classroom with peers who are rated as very attractive. This effect is strongest for female students and appears to be concentrated among the courses of younger and male instructors.

KEYWORDS:

• Peer effects
• appearance
• grades
• classmate appearance

JEL CLASSIFICATION:

• I21
• J71

Acknowledgements

We thank Geoffrey Lehr, Steven Matthias and all the raters who participated in the process. This project has been approved by the MSU Denver IRB.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Student-invariant traits (such as confidence) and instructor-invariant traits (such as responsiveness to student appearance) will be absorbed by the fixed effects. However, they remain possible mechanisms for the peer appearance effect identified in our results because peer appearance varies by classroom, which means that a student or instructor response to that appearance also varies by classroom and will not be captured by the fixed effects.

² Although the ages and backgrounds of our raters may vary substantially from the instructors’ backgrounds, studies have shown assessments of beauty to be similar across individuals (Hamermesh 2011). Our sample also exhibits a high degree of inter-rater reliability (Hernández-Julián and Peters 2017).

³ Students with grade observations in the final sample are not substantially different along observable characteristics from students not in the final sample.

⁴ The results of Table 1 are robust to an ordered-probit specification without fixed effects.

⁵ We also find that peer group appearance does not have differential impacts according to student race, grade point average, class size or whether the course is upper or lower division. These results are consistent with the hypothesis that the appearance peer effect is due to instructor discrimination, in that it is unaffected by student and course traits.