Do we know what skills our students think are being tested in exams?
Jennifer Lingard, Laura Minasian-Batmanian, Ian Cathers,
In a substantial assessment policy change, the University of Sydney is moving to grade allocation based on published standards, rather than cohort distribution. Examiners often allocate higher grades to students who exhibit higher level skills, as based on a taxonomy such as Bloom’s (Krathwohl, 2002) or SOLO (Boulton-Lewis, 1998). However, many students do not understand exactly what skills are required in terms of gaining these higher grades. Questions thus arise as to whether or not students have the same perception of the grading system as those who set the assessment tasks and also if students with more accurately aligned perceptions perform at the higher levels. As students’ learning behaviour is influenced by their perception of what skills are required to obtain higher marks (Trigwell & Prosser, 1991; Prosser, Trigwell, Hazel & Waterhouse, 2000), it is crucial to know what our students think is being tested.
This research aimed to find out if students performed better when their perception of the skills being tested in a particular exam question was well aligned with staff judgments. First year Physiotherapy students (160) undertook an optional online trial exam in their first semester biochemistry unit. This exam consisted of thirty multiple choice questions that were prepared and graded for skill requirement by three content and educational experts. Students indicated their perception of the skill level being tested by each question. Multiple attempts were permitted. Detailed feedback on content and skill testing was also provided online.
About 80% of students completed the assessment, and many have attempted it again. Preliminary data analysis indicates that students view the skills being tested in a much narrower band than do staff i.e. questions testing lower level skills are seen as being more difficult, and vice versa. It is intended to correlate student perceptions with several factors, including their prior level of topic knowledge and final exam performance.
Keywords: grade descriptions, student perceptions, assessment, student learning