Explaining "Unexplained" Grade Inflation in the UK's Universities

Lead Research Organisation: University of Bath
Department Name: Economics

Abstract

The UK's higher education (HE) sector has recently come under much criticism in public debate and from policymakers over grade inflation. According to the Office for Students (2019), 29% of students obtained a first-class honours degree in 2018, up from 16% in 2011. Comparing this rise with changes in students' prior attainment, the authors conclude that increases in top grades are "unexplained" for three-quarters of universities.

The observed grade inflation has coincided with two developments: (i) HE expansion, and (ii) increases in tuition fees. Both developments were Government initiatives to meet growing demand for HE whilst funding it by tuition fee income (HM Treasury, 2013; UCAS, 2013). As a result, the student population increased from 222,430 in 2010/11 to 259,185 in 2017/18, with the tuition fee cap raised to £9,000. The Government also encouraged competition among universities in attracting more students (HM Treasury, 2013, provision 1.202).
Consequently, the cause of grade inflation is widely believed to be the artificial lowering of grading standards by universities in competition for students and, thus, tuition income (Turner, 2019). The over 50% increase in 'good' degrees among students with the A-level entry qualifications of DDD or below (Coughlan, 2019) seems impossible to explain by anything but opportunistic grade inflation. This opinion is shared by the Government, which issued stark warnings to universities against "unfair practices" and proposed interventions aimed at restoring the value of grades (Department for Education, 2019).

In this project, we contrast this popular explanation for grade inflation with an alternative explanation that recognises the role of lecturers as educators (Zubrickas, 2015). We argue that the cause of grade inflation may lie in the expansion of HE, which brought in more students from the lower end of the ability spectrum (Belfield et al., 2018), while acknowledging that the approach to teaching is not constant but depends on the composition and needs of a class. Lecturers may have made their teaching and, accordingly, grading standards more accessible for more numerous weaker students with implications for grade inflation. These adjustments are not opportunistic and artificial but made for the good cause of facilitating learning.

For illustration, consider a lecturer teaching a strong class of students. To challenge them, the lecturer introduces some harder topics and reserves the highest grades for students who coped well with challenging material. Now suppose that the following year the same lecturer teaches a class of lower-ability students. In response to this change, the lecturer scraps harder topics in favour of more accessible topics so that most students are engaged in learning. The outcome of more accessible teaching is coarser grading and, consequently, grade inflation despite the adverse change in class composition. Such an approach pursues learning objectives to enhance the "average" student's knowledge rather than to generate more income.

We will empirically test these two competing explanations of grade inflation using data from the Higher Education Statistics Agency. The explanations make different predictions about grade inflation within a university. If grade inflation is driven by university administrations, then it should take place across all the departments of a university. However, if grade inflation is driven by lecturers' adjustments of teaching standards, then it should take place only in those departments that attracted a larger number of less able students. Our findings on the causes of grade inflation will be important for policy responses. If the popular explanation is true, then policymakers need to clamp down on the opportunistic behaviour of universities and implement policies to restore standards. If the alternative explanation finds support, then such a response would be unnecessary and potentially damaging to the sector.

Publications

10 25 50