Q J Econ. 2015;130(3):1329-1367. doi: 10.1093/qje/qjv021.
The quarterly journal of economics
Leonardo Bursztyn, Robert Jensen
PMID: 26740727 PMCID: PMC4698889 DOI: 10.1093/qje/qjv021
When effort is observable to peers, students may try to avoid social penalties by conforming to prevailing norms. To test this hypothesis, we first consider a natural experiment that introduced a performance leaderboard into computer-based high school courses. The result was a 24 percent performance decline. The decline appears to be driven by a desire to avoid the leaderboard; top performing students prior to the change, those most at risk of appearing on the leaderboard, had a 40 percent performance decline, while poor performing students improved slightly. We next consider a field experiment that offered students complimentary access to an online SAT preparatory course. Sign-up forms differed randomly across students only in whether they said the decision would be kept private from classmates. In nonhonors classes, sign-up was 11 percentage points lower when decisions were public rather than private. Honors class sign-up was unaffected. For students taking honors and nonhonors classes, the response depended on which peers they were with at the time of the offer, and thus to whom their decision would be revealed. When offered the course in a nonhonors class (where peer sign-up rates are low), they were 15 percentage points less likely to sign up if the decision was public. But when offered the course in an honors class (where peer sign-up rates are high), they were 8 percentage points more likely to sign up if the decision was public. Thus, students are highly responsive to their peers are the prevailing norm when they make decisions.
