Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
This paper investigates how group identification distorts people’s beliefs about the ability of their peers in social groups. We find that experimentally manipulated identification with a randomly composed group leads to overconfident beliefs about fellow group members’ performance on an intelligence test. This result cannot be explained by individual overconfidence, i.e., participants overconfident in their own skill believing that their group performed better because of them, as this was ruled out by experimental design. Moreover, we find that participants with stronger group identification put more weight on positive signals about their group when updating their beliefs. These in-group biases in beliefs can have important economic consequences when group membership is used to make inference about an individual’s characteristics as, for instance, in hiring decisions.