Grade inflation: a misplaced myth?

Do improved results at degree level really indicate a rise in standards? Nicholas Pyke examines a new report that aims to find out

Hard-working women rather than "grade inflation" is the reason for the astonishing increase in the number of top-class degrees handed out over the past decade. That is the conclusion of two economists who have subjected British universities to a new statistical analysis.

Professors Geraint Johnes and Bob McNabb, from Lancaster and Cardiff universities respectively, conclude that there has been no significant decline in degree standards over the past decade. In other words, criticism directed at the senior common room, supposedly willing to sacrifice rigour for cheap popularity, has been misplaced.

The widespread scepticism about grade inflation has been generated by a transformation in the scale of students' achievements. Between 1988 and 1993, the proportion graduating with a "good" degree (a first or upper second) increased from around a third to a half. Over the same period, the proportion of school leavers going on to higher education doubled from 15 to 30 per cent.

Johnes and McNabb, who presented their findings to the Royal Economic Society's annual conference at Warwick University yesterday, conclude that the improvement in degree results at one end of the university system is broadly in line with an improvement in the quality of students, as measured by A-level results, at the other.

The large increase in the number of women in higher education has probably played an important role in driving up degree results, they say, pointing out that universities with more men tend to produce lower grades (although this may be to do with the subjects men favour).

"The main driver for the change in degree results has been a gradual change in the personal characteristics of the typical student," explains Professor Johnes. "These include such factors as gender mix: more women go to university now than did a generation ago, and they tend to get better degrees than men."

So far, their theory has been broadly accepted. "The general thesis makes sense," says Roger Brown, the director of Southampton Institute and a former head of the Higher Education Quality Council. "The increasing proportion of good grades does reflect the quality of the candidates' A-level scores."

Not that Johnes and McNabb let the universities off scot-free. Using the same statistical techniques, they also conclude that there probably was a period of grade inflation between 1984 and 1988, when the proportion of firsts and upper seconds increased by 14 per cent. This was without any obvious improvement in the quality of candidates: "Our findings suggest that there has been no recent decline in academic standards in British higher education. However, there was a decline during the mid-1980s and this was quite sharp."

Alternatively, it may be that students were working especially hard at that time. But their study found no evidence of this.

Professor Alan Smithers, of Liverpool University, believes that the unexplained increase in the number of top degrees in the mid-1980s could be the result of old universities competing with their newer rivals. Sensing for the first time that they were being judged alongside former polytechnics, the established institutions adjusted their results upwards to demonstrate the superiority of their students.

His chief reservation about the conclusions reached by Johnes and NcNabb is that, in his view, the "difficulty" of degrees varies greatly from one university to another and from one discipline to another, making overall conclusions hard.

But, as its authors acknowledge, the main problem with the study is the fact that it relies on the value of A-levels remaining constant over time.

If the improvements in A-level results are a sham, then so is the parallel increase in the number of good degrees. And the question of A-level standards is vexed, to say the least.

'Academic Standards in UK Universities: More for Less or More for More?' by Geraint Johnes and Bob McNabb; e-mail G.Johnes@lancaster.ac.uk or McNabb@cf.ac.uk