Fairness in grades and admissions

In the summer of 2020, there was a fairly big scandal in Britain concerning secondary school children’s grades. Earlier that year, the government had decided to cancel all examinations, owing to the pandemic. (At the time, forcing those least at risk from Covid to sit and home and pretend to learn things on Zoom was considered good public policy, you see.) Yet without any examinations, there weren’t any exam results. And without any exam results, there wasn’t a fair way of deciding who should get to go to which university. Quite a pickle.

The government’s proposed solution was for teachers to submit predicted grades, which would then be “moderated” by the examinations regulator. Since predicted grades tend to be on the optimistic side (i.e., teachers expect their students will do somewhat better than they actually do), the moderation was done to prevent grade inflation. It took the form of an algorithm that adjusted grades up or down based on factors like the school’s average performance in previous years. Not ideal, of course, but else are you going to do? (I guess a timed online IQ test might have worked better.)

When the moderated grades were finally announced, there was a public outcry. Since the moderation had been designed to prevent grade inflation, it ended up reducing a lot of grades (surprise, surprise.) When it came to A-Levels (the exams taken at age 18), almost 40% of moderated grades were lower than what teachers had predicted, and only 2.5% were higher. This left a lot of students, parents and teachers feeling indignant. While the moderated grades might have been about right at the level of schools, there were a lot of individuals for whom they weren’t right at all.