Measuring the ratio students actually lived through

A student finishing in 2026 did not experience one year's staffing ratio, they experienced three. The top chart shows which years each group of students studied under, and which of those years the published data actually covers. The bottom chart shows what is left once the ratio is measured that way and the pandemic rebound is taken out.

1. Which years each cohort studied under

Three-year degrees shown – Scottish universities add a fourth year at the start

Studied, ratio published Studied, no ratio published
The 2025 and 2026 leavers rest on exactly the same two published years. Their experienced ratios correlate at 0.995 and differ by less than a tenth of a point on average, so comparing those two surveys cannot detect a staffing effect even if one exists. Finding nothing there proved nothing.

2. What survives once the rebound is taken out

Correlation between the change in experienced ratio and the change in each score, after allowing for where the university started in 2023. Bars are 95 per cent confidence intervals – anything crossing the zero line is not distinguishable from no effect

Before adjustment After adjustment Survives the adjustment
Change in experienced ratio against change in each score
ThemeBefore adjustmentAfter adjustment95 per cent intervalp

106 universities with at least 100 NSS responses in both 2023 and 2026. The experienced ratio averages the Guardian University Guide staff to student ratio across the published years of each cohort's own degree. Two academic years students lived through, 2022/23 and 2025/26, have no published ratio, so every window rests on two years rather than three. The adjustment removes the university's 2023 starting score, which is what the pandemic rebound runs through.