Analysis of short-answer question styles versus gender in pre-clinical veterinary education.

One large study in medical education has reported that the choice of question format (or question content) could introduce a gender bias, with men outperforming women on questions with a true-false component or that required knowledge of anatomy or physiology. The purpose of our study was to ascertain whether this finding is also true in veterinary medical education. Two veterinary student cohorts were analyzed across four different modules over a three-year period (804 questions in total). The results of the study show that the women's and men's performance did not differ in any of the question types analyzed across any module or year. When students' (both women and men) overall average performance on different question types was compared with their performance on standard multiple-choice questions (MCQs), performance levels increased when students were asked to answer MCQs that contained an image-based prop (IMCQ) such as a photograph, X-ray image, or diagram. In contrast, students' performance was consistently lower when answering assertion-reason questions (ARQs), and this performance could not be explained by the demographic makeup of the two cohorts analyzed. When comparing standard MCQs with MCQs that contained a true-false question stem, no specific trend in the data could be determined. In conclusion, this study suggests that the short-answer question style does not bias against one gender in veterinary medical education, but that overall students do perform differently according to question type and, in particular, less well when ARQs are used in examinations.