ED31B-0890
Does Question Structure Affect Exam Performance in the Geosciences?

Wednesday, 16 December 2015
Poster Hall (Moscone South)
Elizabeth A Day1, Mitch Keith D'Arcy1, Lorraine Craig1, Michael J Streule1, Emma Passmore1 and Jessica C E Irving2, (1)Imperial College London, Earth Science and Engineering, London, United Kingdom, (2)Princeton University, Princeton, NJ, United States
Abstract:
The jump to university level exams can be challenging for some students, often resulting in poor marks, which may be detrimental to their confidence and ultimately affect their overall degree class. Previous studies have found that question structure can have a strong impact on the performance of students in college level exams (see Gibson et al., 2015, for a discussion of its impact on physics undergraduates).

Here, we investigate the effect of question structure on the exam results of geology and geophysics undergraduate students. Specifically, we analyse the performance of students in questions that have a ‘scaffolded’ framework and compare them to their performance in open-ended questions and coursework. We also investigate if observed differences in exam performance are correlated with the educational background and gender of students, amongst other factors.

It is important for all students to be able to access their degree courses, no matter what their backgrounds may be. Broadening participation in the geosciences relies on removing systematic barriers to achievement. Therefore we recommend that exams are either structured with scaffolding in questions at lower levels, or students are explicitly prepared for this transition. We also recommend that longitudinal studies of exam performance are conducted within individual departments, and this work outlines one approach to analysing performance data.