DP11716 | Gender Di erences in Academic Performance: The Role of Negative Marking in Multiple-Choice Exams

Publication Date

12/20/2016

JEL Code(s)

Keyword(s)

Programme Area(s)

Network(s)

Abstract

We investigate whether penalizing wrong answers on multiple-choice tests ("negative marking") makes females relatively worse o compared to males (the comparison being no penalties for wrong answers). With a cohort of more than 500 undergraduate students at a major Span- ish university, we conducted a eld experiment in the Microeconomics course. We created a fi nal exam, which was composed of two parts: one with penalties for wrong answers and one without. Students were randomly allocated to di fferent exam permutations, which di ffered in the questions that carried penalties for wrong answers. We fi nd that the penalties did not harm female students. Females performed better than males on both parts of the exam and did so to a greater extent on the part with penalties. Whereas risk aversion did not aff ect overall scores (despite aff ecting answering behavior), ability did. High-ability students performed relatively better with negative marking, and these were more likely to be women.