Exploring differential attainment by assessment type in mathematics, chemistry and life sciences

Denholm-Price, James, Page, Nigel, Williams, Neil and Dourado, Luis (2019) Exploring differential attainment by assessment type in mathematics, chemistry and life sciences. In: Horizons in STEM Higher Education Conference : Making Connections and Sharing Pedagogy; 03 - 04 July 2019, Kingston-upon-Thames, U.K.. (Unpublished)


A BME attainment gap persists across the HE sector: the proportion of students who attain a 1st or 2:1 honours degree is lower for students from a BME background than for non-BME students, even when controlled for prior attainment or entry profile. The reasons for the gap are unclear (Richardson, 2015) and many theories have adopted a student-deficit model. However, Broeke and Nicholls (2007) found that even when controlling for other factors (including prior attainment, subject of study, age, gender, disability, a proxy of socio-economic background, type of HE institution attended, type of level 3 qualifications, mode of study and term-time accommodation) being from a minority ethnic community remained a statistically significant variable in explaining final attainment. More recent research directed at understanding the gap has focused on institutional practice; including assessment practices. This research also identifies the importance of positive relationships with staff, which provide students with support and encouragement (HEFCE, 2015). Assessment practices must consider a significantly more diverse student body nowadays and measures taken to improve assessment should promote inclusivity. Poor legitimacy in assessment harms student confidence when assessments are not inclusive, authentic or linked to developing employability. Modes or types of assessment have received relatively limited literature scrutiny with respect to the BME attainment gap debate, although there is well-established evidence of differences related to student gender and contradictory evidence around assessment type, possibly confounded by subject. For example, Leathwood et al. (2011) analysed records of students graduating from Kingston in 2009 and found that examination (as opposed to coursework) disadvantaged BME students. However, when looking at assessment in Mathematics modules, Fairclough (2015) offers examples where time-constrained assessments reduces attainment disparity with respect to ethnicity compared to less traditional portfolio assessment. This project aims to contribute to discussion by expanding the evidence base. It focuses on the outcomes of a broad range of assessments with a sample from 25 undergraduate maths, life science and chemistry modules combined with demographic information from nearly 1000 students to examine student performance on different forms of assessment (encompassing formal exams, coursework, “in-class tests” and oral presentations). The sample data show 14 statistically-significant differences in the attainment of certain groups of students in some assessment modalities, but not all. This talk will present some of the results of this analysis and their possible interpretation that is relevant to strategies for addressing the BME attainment gap. The talk also aims to give attendees an appreciation of the fact that students perform differently on different types of assessment across different subjects, irrespective of the relative perceived “difficulty” of the subject.

Actions (Repository Editors)

Item Control Page Item Control Page