The crackdown on grade inflation has hit students of colour harder than white students
Jim is an Associate Editor at Wonkhe
Tags
“BAME” hides all sorts of complexity – but even the Black awarding gap had fallen to 16.5 percentage points, again the lowest on record.
Different definitions abound, but here we’re looking at UK domiciled first degree qualifiers by classification of first degree and ethnicity marker, where we’re comparing percentage of white students getting a first or a 2:1 with the percentage of non-white students that achieve the same.
The problem with the champagne corks is that it all turned out to be a statistical trick. If in a given year everyone does better, then it would look like the gap was closing because more BAME students would hit the threshold for a 2:1.
You’d only spot if there was still a problem if you looked at firsts only. And guess what – the BAME firsts awarding gap was 9.9pp, and the Black firsts awarding gap was 18.2pp – the highest on record.
So when, back in January, HESA published this year’s data on degree classifications and Universities UK was celebrating “turning the corner on grade inflation”, the question was whether that would cause the reverse effect on the awarding gaps.
And guess what again. The BAME gap has increased 1.7pp to 10.5pp, and the Black gap has increased 1.8pp to 18.3pp. White students getting good honours fell 3.1pp – but BAME fell by 4.8pp, Black students fell 4.9pp, and Asian students getting good honours fell 5.3pp.
In other words, the crackdown on grade inflation has hit students of colour harder than white students. Whatever the sector and its regulators think is being done on awarding gaps, it’s not working. And in some cases, we’re going in reverse.
We can’t see students’ actual marks, obviously, but we can see those firsts. There the gap for BAME overall is the highest it’s ever been at 10.5pp, Black students is up at 17.6pp, and it’s also the highest it’s ever been for Asian students at 8.6pp.
The “cracking down on grade inflation” thing hitting BAME students hardest is interesting because the sector told itself and others that the flexibilities it found during over Covid hadn’t compromised the integrity of grades.
So the resultant re-tightening of academic regulations, and the accompanying return to pre-pandemic (and in some cases prehistoric) assessment practices may have cheered up the regulator and politicians – but appear to have made it harder for BAME students to demonstrate their abilities and competence.
Looks like it’s time to go back to the drawing board, sharpish.
It’s more complicated than this, but yes. Everyone doing better closed attainment gaps partly because some students from some demographic groups were disproportionately more like to get 2:1s and everyone did better. However, gaps in marks for online assessments were also smaller than physical exams, and this hasn’t changed (the gaps may not be as small, but they are still comparatively smaller). Similarly, the move to online assessment in 2019/20 (positive) and the shift to online *learning* in 2020/21 (more mixed) did different things
There’s plenty of work left to do looking at what causes attainment gaps (of all stripes) – what roles do different assessments, subject mix, prior attainment etc play; what about concepts like belonging etc – and intersectionality (some of our gaps bounced back, but others have remained smaller). I’m not convinced everyone had necessarily abandoned the drawing board in the first place.
As always in discussions around the awarding gap, Jim fails to mention the access gap: a much smaller percentage of white 19-year olds are in higher education than BAME 19-year olds (and also than black 19-year olds). A more valid statistic is the number of students achieving at least a 2:1 divided by the total number in that relevant cohort in the population (not just those who went to university).
One of the problems is that as a sector we haven’t had a proper debate about what the increase in ‘good’ degrees during the pandemic period meant. The key factor was the changes in assessment type, and we haven’t had a serious investigation and discussion of whether this allowed students’ learning to be assessed in a more valid, fairer way (so that the classifications weren’t ‘inflated’) or if these changes made assessments measurably ‘easier’ (and classifications therefore ‘inflated’). Instead as a sector we’ve responded to political pressure, without doing the research and evaluation that would allow us to truly understand the issue.
Bobby’s point about statistical ratios is important, no fudging of the facts please.
Having sat through my sons graduation in 2018 in Winchester it was revealing listening to other students comments about biases and preferential treatment SOME of the black students benefitted from. One who apparently rarely attended lectures, was dismal in group work sessions, but scored highly in the ‘intersectional victims game’ got a grade (1st plus a ‘special prize’) that did not fit, even with other black students assessment of her academic abilities. Skewing outcomes seems to be popular for reducing the ‘apparent attainment’ gap, though much of the problem we in our University see comes from previous education skewing results (via teacher assessments), often as far back as primary school, that to maintain ‘apparent attainment’ Universities have to do the same. The real damage of course is reputational, hence why some Universities Graduates are less employable than others.
As Andy notes, it’s much more likely that this is caused by a reversal of COVID-specific policies and practices, rather than ‘cracking down on grade inflation’ (I doubt many universities are basing their policies on the OfS’ pronouncements on that issue).
Beyond a move to online assessment, quite a lot of the sector put in specific assessment policies to mitigate the impact of COVID on marks and degree classifications (safety nets; changes to exceptional circumstances policies etc). These had the effect of aiding the most disadvantaged; thus we might expect that ending them will have reversed some of the gains in award gaps.
At my institution, and I’m sure at many others, we’re looking at what aspects of COVID policies we can keep (and have kept some), as part of more general work on award gaps.
The reversal of grade inflation is as much an enthusiastic commitment from the sector as it is a finger wag from OfS…
https://www.universitiesuk.ac.uk/what-we-do/creating-voice-our-members/media-releases/return-pre-pandemic-upper-degree
That was a commitment (on behalf of the sector) from UUK – made without consulting members(?) and presumably undertaken as an act of prostration and pre-emptive sacrifice before the mighty OfS – but the reversion to date was both expected (when we reviewed the impact, the drivers for the change included but were not limited to (a) a rapid shift to online assessment in 2019/20 which saw *some* assessments work less effectively than others, leading to some unexpected mark profiles and (b) no detriment style policies; as per David, moving away from the latter and improving the former saw an unexpected and appropriate reduction in outcomes) and not actually a reversion to pre-pandemic levels (it’s a drop below 2019/20, but still some way higher than pre-pandemic).
We didn’t do anything differently following the UUK announcement, and I’m not sure what impact it had on sector practice. We continued to do what you should do in the situation we were in – review what happened, look to improve anything that hadn’t worked, look to discontinue anything which was time-specific and no longer required, continue monitoring outcomes (specifically in relation to outcomes in the sector) and demographic variation. It will be interesting (and exciting) to see what happens in Summer 2023.
What Andy said. I’m not sure that most people in our institution, including those in charge of assessment policy, are even aware of the UUK announcement.
Having interviewed students about their practices in the pandemic, and their practices and experiences around assessment, maybe we need to face up to the reality that as a sector we still have Structural Racism, Unconcious Bias, and sadly, in some instances, we still have students experiencing outright racism.