Framing effects on risk-taking behavior: evidence from a field experiment in multiple-choice tests

Abstract We exploit testing data to gain better understanding on framing effects decision-making and performance under risk. In a randomized field experiment, we modified the of scoring rules for penalized multiple-choice tests. tests, right answers are typically framed as gains while wrong losses (Mixed-framing). Loss-framing proposed, both non-responses presented in loss domain. According our theoretical model, expect change decrease students’ non-response increase performance. Under Loss-framing, reduces by 18%-20%. However, it fails scores. Indeed, results support possibility impaired Loss-framing.