Framing Effects on Risk-Taking Behavior: Evidence from a Field Experiment

Abstract

We exploit testing data to gain a better understanding of the influence of framing effects on decision-making and performance when facing risk. In a randomized field experiment we modified the framing of grading instructions of multiple-choice tests. In the business-as-usual framing (BAU framing), correct answers were framed as gains while wrong answers were framed as losses. Under our proposed loss framing, both non-responses and wrong answers were presented in the loss domain. According to our theoretical model, we expect the change in framing to decrease student non-response and increase student test scores. Our experiment confirms that framing matters for risky decisions in the context of testing. It reveals that under the loss framing, student non-response is reduced by 17%-19%, with risk-aversion being the main candidate for explaining this reduction. However, it fails to increase students’ scores. Indeed, our results support the possibility of impaired performance under the loss framing.