Grade inflation – an international perspective
David Kernohan is Deputy Editor of Wonkhe
Tags
A timely new research summary from QAA attempts to put our ongoing moral panic over grade inflation into global perspective.
If you want an overview of the way the issue is framed in England, there was an OfS briefing released last week that hits most of the key notes. A UKSCQA “Statement of Intent” outlines the response of universities.
Is this an international issue?
The accusations of “soft marking”, “inflated norms”, and – in tension with this – “marking on a curve” has led to benchmarking upwards around the world. Outside England, the trend is generally ascribed to corporate rather than academic judgements having an undue impact on assessment – an issue of “corruption” in the words of the Council for Higher Education Accreditation (CHEA also notes issues like “unfair practices in appointment of officials (eg through nepotism or favour)” and “political or commercial interference in regulatory decisions” as other areas of concern globally – which is a mood).
However, published evidence internationally pours cold water on the “more effective students” hypothesis. There’s evidence that students are choosing less challenging assessment methods, choosing courses with reputations for higher grades (this would be courses in the US/global sense, we would say “modules” in the UK), and are complaining when they don’t get the grades they feel they deserve. Students are spending less time on their studies, standardised test performance (in Turkey and the US) has remained stable or decline, and growth in student numbers generally means less spending on teaching and support per students.
QAA also identify a Hungarian study that finds that grades have an impact on student evaluations of teaching – recommending that surveys take place before marks are known.
What’s being done about it?
The UK statement of intent is globally unique. Elsewhere approaches involve accreditation reviews, the introduction of externality (such as the UK external examiner system) to assessment and calibration, and improvements to audit and monitoring. There’s little evidence of direct quality assurance interventions in the problem – either by individual agencies or trans-national organisations.
When individual institutions have taken unilateral measures to limit the use of higher grades, there have been notable climb-downs when results have disappointed students. A fascinating paper documents how this played out when Princeton University and Wellesley College experimented with institutional quotas. Despite the existence of grade inflation concerns in the US, other national systems (including the Netherlands and Malaysia) are interested in moving to US-style approaches like ranks, percentiles, and median grades.
Recommendations
This is a research review rather than a policy intervention, but we do get a series of possibilities for future action. Many of these suggest further research – improving evidence and analysis, further tracing of the impact of policy interventions, and the development of national guidance and support.
What’s notable looking at international work is how rarely academic judgement – rather than institutional policy – is blamed for an “unexplained” issue in student attainment. The shift towards beefing up external examiner processes globally (and I understand there may be a QAA report on practice in the UK on the way) points to a need to reinforce and support academic judgement rather than algorithms.
There does need to be room to identify and reward excellent student performance with a grade that reflects the quality of work rather than the position of a student on a curve of their peers. Quite how we get there remains to be seen.
One could argue that the recent talk of ‘grade inflation’ is a result of more support and more equitable assessment formats during Covid. For years (certainly the UK) has rested on traditional assessment methods, which advantage some over others (I.e. those who are better at academic writing) yet in the UK we rarely train students how to do this (nor is it necessarily a reasonable expectation, and is not specifically outlined as a learning outcome in QAA FHEQs). Covid also allowed more time for exams, with students getting 24hrs and it being possible to confer with the peers (more akin to a real work situation, but of course brings with it challenges). And of course, we cannot underestimate the impact of market forces in HE in ‘grade inflation’ – league tables, threat on jobs in the event of poor performance etc. If you pit a sector in competition with one another, do not be surprised when they compete on outcomes. In terms of solutions mentioned above, guidance is key, but so is training. The focus on “externality” to me is flawed in its current form. While useful to have other opinions on assessments and standardised marking, it’s important to note that, with no initial teacher training in HE and very little in the way of formal training in assessment methods, much of this is based on opinion by non-experts (in assessment writing) and not on evidence. Until the sector realises that ‘teaching’ should not play second fiddle to research (bearing in mind up to 70% of funding can often come from tuition fees is some institutions), standards are always likely to be more variable. A good researcher does not necessarily make a good teacher, nor a good materials/assessment writer – trying to address this by adding more informal quality assurance only makes sense if those involved are trained experts in assessment writing. Guidance is important, but unlikely to be taken on board without assessed training. Initial teacher training as standard for teaching at HE is key, followed by mandatory training throughout. And policies that prevent HE’s from removing programmes/staff with lower attainment, but instead investing in further training in those circumstances.