That education policy in the UK is confused is nothing new, neither given the rapid turnover in education ministers is it surprising. But the latest turn, although rhetorical at the moment, is both strange and worrying.

In the last two weeks both the Prime Minister, Boris Johnson, and the Education Minister Michelle Donelan have criticised the quality and relevance of university courses. Johnson talked about “low-value courses” in his major set-piece speech on economic recovery post-Covid-19 while Donelan said that too many students “have been misled by the expansion of popular sounding courses” with what she described as poor standards and “no real demand from the labour market”.

Clearly most of this rhetoric is ideological. Johnson is talking about more funding for Further Education Colleges, which have been starved of funding through the period of austerity. However, it is being suggested that one motive may be that university cities tend to vote Labour, but in many of the towns in which the Conservatives won new seats in the election last November, there are not universities but are Further Education colleges.

To justify the talk of low value and poor quality courses the government produce various data as evidence. There are different surveys looking at issues related to satisfaction and student outcomes. The first is the student satisfaction survey conducted in every university. Although comprehensive it is doubtful that this survey has much greater validity than the happy sheets I used to hand out at the end of staff development workshops. Universities go to great lengths to make sure students are happy, through various gimmicks and social events.

The Graduate outcomes for all subjects by university (LEO) survey is undertaken by the Office of National Statistics (ONS). It surveys the employment and earnings of higher education graduates using matched data from different government departments. It is interesting that the ONS describes the survey ads “experimental.”

Although interesting the sheer number of variables impacting on graduate earnings after finishing at university render the findings meaningless when compared to subject sample sizes. After Donelan’s speech, former universities minister Jo Johnson tweeted salary data is about as useful a guide to course quality as an MP’s majority.

Of course, one of the “experimental” findings is that students undertaking STEM subjects have higher earnings that those doing humanities and arts. And the strong suspicion it is humanities and arts courses that Johnson and Donelan are firing at.

Many would probably argue that earnings are not the best proxy for judging course quality in any case. But it is interesting that the Graduate Outcomes survey, through a series of reflective questions. found that graduates of creative arts courses are more likely to be using skills learned during their course in employment than their peers who studied maths, biology, or physics.