Students in larger module groups ‘tend to get lower grades’

Effect also more pronounced for science subjects, according to study of class sizes at UK university

May 16, 2021
Lecturer, lecture hall
Source: iStock

Students who take course modules in the largest groups are likely to get lower grades, particularly in science, technology, engineering and maths subjects, a new study suggests.

Researchers looked at data on grades and module enrolments from about 25,000 first-year undergraduates across seven years at a large UK research-intensive university to see if there was any link between class size and performance.

After controlling for student ability by statistically isolating how a certain student fared as class size varied, they found that the bigger the group on a module, the lower their grades.

The effect was even more pronounced for STEM modules, while a separate analysis also suggested that students from the most disadvantaged backgrounds tended to suffer more from larger classes.

In the paper, published in the Economics of Education Review, the authors say that their findings have “important implications” for the current focus of many governments on STEM in higher education.

Although it “might seem obvious”, the findings suggest “the drive to expand enrolment in STEM subjects should be accompanied by investment in teaching resources to avoid a deterioration in students’ achievement”.

“Our analysis highlights that the negative effects in STEM subjects would be more pronounced than in non-STEM subjects, suggesting that there might be a need to funnel scarce resources in the STEM direction.” 

They add that the data on disadvantaged students also suggests “developments that would lead to higher student to staff ratios in these fields would disproportionately impact a group that is already disadvantaged in its access to tertiary education”.

Although the paper does not fully investigate why larger classes have a particular impact on STEM students, it refers to previous research highlighting the instructional nature of such subjects – rather than the student-centred model of learning employed in non-STEM disciplines – and how this could be a reason for the difference.

Data for the study were taken from first-years studying from 2007-08 to 2013-14, but co-author Michael Vlassopoulos, professor of economics at the University of Southampton, added that an interesting area for further research would be how variations in class sizes might have affected student performance during the pandemic.

“We know relatively less about how staff-student ratios impact learning in online settings, although some research on this is now emerging due to the pandemic. Certainly, [this is] an area where more research is needed going forward,” he said.

simon.baker@timeshighereducation.com

Register to continue

Why register?

  • Registration is free and only takes a moment
  • Once registered, you can read 3 articles a month
  • Sign up for our newsletter
Register
Please Login or Register to read this article.

Related articles

Reader's comments (1)

One factor not discussed in the article is why such modules exist. Large modules in STEM are more liely to be foundational material offered to a wide cohort of students rather than designed specifically for thos on specialist programmes, e.g. introduction to analysis or introduction to programming. In some case these may be recommended or required modules for other programmes. Thus the students will have a wider range of backgrounds than say those on 'complex analysis' or 'scientific computing' where the cohort have a deeper background in the topic, and have possibly met higher entry requirements. The benefit of running larger modules with a broader intake is probably seen as ensuring that students can get exposure to the material usually delivered by specialist in the home dept for the material. Obviously the challenges need to be offset by support elsewhere, e.g. more tutorial/lab classes But to quote Euclid: " There is no royal road to geometry" If some material is offered to larger more diverse cohorts. It is hardly surprising that outcomes will be mixed esp so in STEM where lower prior exposure may create a greater mountain to climb than i the dual case of a large mixed-cohort studying non-STEM material, where you'd hope/expect an undergraduate to have sufficient background from [say A-levels] to engage with the material

Sponsored