The effects of exam frames on student effort and performance.

We conduct an online experiment in which participants take a multiple-choice vocabulary exam with scores described using either a "loss frame" or a "gain frame" to examine how exam framing affects item skipping, overall performance, and effort. In the loss frame, participants begin with an "endowment" of points and lose a point for incorrect answers, receive no points for skipped items, and gain a point for correct answers. In the gain frame, participants do not lose points for incorrect answers, gain one point for skipped items, and gain two points for correct answers. Contrary to traditional choice theory, where students may choose to skip more questions if the framing induces loss averse preferences, we find that loss-framed exams lead to less skipping. We also find that the loss frame led to improved overall performance, which appears to be driven by increased effort on the exam among participants in the loss frame group. We find little evidence of a gender differential response to frames. [ABSTRACT FROM AUTHOR]