Criteria or quotas for success? Grade inflation and the role of norm-referencing

The ongoing increase in the number of upper degrees is generating significant debate over what constitutes a genuine and welcome improvement, and what represents more suspicious practice that lowers the standards that students are assessed against.

In his recent piece for Wonkhe, Iain Mansfield argues that too much focus on this question is standing in the way of meaningful reform. But, is imposing limits on the number of good degrees awarded to students, as his approach ultimately suggests, really the answer?

What does a norm-referenced classification tell us?

The theory goes that by limiting the number of firsts and 2.1s, institutions are effectively prevented from “over-awarding” and inflation goes away. Fewer good degrees are up for grabs, meaning only the most talented graduates receive them, and the value of a good degree as an indicator of excellence is re-established. Employers, and universities are better equipped to identify the highest performing graduates while also restoring public confidence in standards.

The catch is that these higher performing graduates can only ever be judged the most talented relative to their peers on the course. Unlike in schools, where comparisons can be made against a national curriculum and national assessments, the UK higher education system is founded on diverse courses, teaching and assessments offered by autonomous and diverse universities. Even with a “sufficiently large subset”, a norm-referenced classification awarded by an institution does not really tell you where a graduate sits relative to the wider graduate population.

A norm-referenced classification would also tell us nothing of the absolute level of knowledge or skills acquired over the course of a graduates’ studies, beyond the threshold requirement for a degree; simply where they ranked against their peers. Two graduates receiving the same overall mark could find their classifications differing; in a low-performing year, the mark may result in a first while in a high-performing year, where competition is greater, only a 2.1. For example, norm-referenced degree classification is not useful for courses such as medicine and nursing where the focus is on meeting set standards.

Norm-referencing is not only unfair on the graduate who has worked hard and scored well, yet not received a commensurate classification. It is also unfair on teaching staff and institutions who have been successful in supporting students to achieve genuine improvements in performance. It also makes incentivising and evaluating any innovations in teaching that much harder. Only raw scores will indicate where there have been changes.

Crucially, it, therefore, undermines a central objective of norm-referencing – namely, to restore the value of a good degree. Without context, the classification loses its meaning. If we are to believe there has been grade inflation in schools, it risks perpetuating this trend, students receiving grades based on the performance of their cohort and not according to the knowledge or skills they can demonstrate.

Motivating or demotivating?

The introduction of competition undoubtedly has the potential to motivate some students. There are those who will strive to achieve the highest marks possible, paying attention to the working habits of their contemporaries to try and better these. A first-class degree might be viewed by the outside world as indicative of those who best apply themselves and are most deserving. Even so, the absolute knowledge test should still remain.

Moreover, for others who judge themselves incapable of matching the level of the highest performers, irrespective of how hard they study, norm-referencing can have a demotivational effect.

If only 15% of students can receive a first, but 40% are guaranteed a 2.1, a student who feels they are “average” upon entry may have little appetite for trying to compete in a contest where the winners have already been decided. Even if they improve, what chance is there of bettering the top performers, who might also be improving?

For students lacking self-assurance this could be particularly damaging, and in extreme cases, confidence could fall to the point of questioning their continuation. We must be mindful of the implications this will have for widening participation, if students entering higher education with lower tariff scores or non-traditional backgrounds for their course are worried about how their ability matches up to their peers.

Where do we go from here?

We all know that getting caught up in the detail and trudging through evidence can be a distraction from making changes happen. If the sector is to get a genuine handle on a complex issue such as grade inflation and improvement it should not look for a quick-fix in the form of artificial limits on success. Doing so risks replacing one form of devaluation with another, while also penalising students.

Nor is a hybrid approach, where criteria are set to maintain a distribution of grades, a quick solution. We would still need to resolve how the quota can be set through criteria across different courses with different curricula and, crucially, how it can be recalibrated in response to legitimate improvements in student attainment. There is scope for greater awareness of the drivers of grade distribution to help identify and address inflation and even improve calibration, but this would be a step short of the managed distribution that Mansfield’s proposals take us toward.

To help address this complex challenge, we are working with Guild HE, the Quality Assurance Agency for higher education (QAA) and the UK Standing Committee for Quality Assessment (UKSCQA) to understand these trends and identify where targeted action or longer-term changes in assessment and calibration practice are required. First, however, we are developing a sector reference point to aid transparency and consistency of approaches to setting degree classification criteria by autonomous and self-critical academic institutions.

Ultimately this work aims to give students and their prospective employers and universities themselves confidence in the way attainment is assessed and classified across the sector. The work is ongoing and if you would like to participate in the conversation, please get in touch.

post list Latest articles

4 responses to “Criteria or quotas for success? Grade inflation and the role of norm-referencing

  1. Exactly what I’ve been thinking. I spoke to some students at a university where they rank every student in some kind of league table of success. The students said they felt as though it brought in competition instead of collaboration and prevented them from taking academic risks.

    They understood it trained them for the “real world” of competitive labour markets, but I hadn’t realised how much criteria-based assessment might be keeping the neoliberal world of league tables, shameless competition and individualism at bay.

    Norm referencing can’t be good for students’ intellectual pursuits, nor their mental health.

  2. An anecdote. A friend qualified as a physiotherapist in the 1970s, when there were national exams for that diploma qualification. She was offered a position after an extremely short interview because, according to her interviewer, wherever she had trained her qualification was of the same worth, and showed that she had the right set of skills. \

    Maybe that should be the model for HE in general? We would have large enough numbers to ensure that norm referencing could work, and we’d also be sure of equity between institutions.

  3. >>For example, norm-referenced degree classification is not useful for courses such as medicine and nursing where the focus is on meeting set standards.<<

    That'll be why medicine doesn't award classified degrees, then.

    One of my biggest concerns with norm-referencing – which is also one of my biggest concerns about grade inflation – is that it makes it impossible to compare fairly between students graduating in different years. A student who had graduated with a 2i one year competing for jobs with a graduate from the same course who gained a 1st the following year with lower marks (e.g. as a result of a change in the classification algorithm) has a right to be angry, and to consider legal action against the university.

  4. >>There are those who will strive to achieve the highest marks possible, paying attention to the working habits of their contemporaries to try and better these.<<

    This is a major issue for courses that try to develop collaborative working habits, which employers say they value. Setting quotas on the number of degrees of a particular class motivates students not just to improve their own performance but also to try to undermine others' (the Dick Dastardly effect). In my undergraduate days (pre-internet), pharmacy students were in the habit of razoring recommended articles out of journals so that other students couldn't access them. This kind of Hunger Games ethos persists among students on competitive courses who are convinced – despite repeated assurances to the contrary – that their outcomes will be based on how they perform relative to other students rather than how they perform against specific criteria.

Leave a Reply