When it comes to the debate about universities and value for money, we agree with Sam.

In June, Sam Gyimah, England¡¯s universities minister, told the ÖйúAƬ Policy Institute¡¯s annual that ¡°salary levels are not the be-all and end-all¡± when seeking to measure the benefits of ÖйúAƬ to individuals or society.

¡°There is more to a university degree than lifetime earnings,¡± explained Gyimah, adding that ¡°not all the benefits of education can or should be captured in future salary¡±.

¡°Some graduates¡¯ passion or talents might lie in a subject area that is not highly paid, but is personally rewarding and of benefit to society,¡± he said, highlighting how some graduates do not go into paid work at all, with many working to raise families or care for relatives, or taking ¡°valuable and important jobs with low salaries¡±.

ÖйúAƬ

We also concur with the minister when he said that students require ¡°information¡­to make the right decision¡± when applying to university.

At the same time, Gyimah launched ¡°the first major analysis of the Longitudinal Education Outcomes ¨C or LEO ¨C data for people five years after graduation¡±, data about jobs and earnings for graduates for different degree subjects and universities.

ÖйúAƬ

He called the LEO analysis ¡°part of a revolution in transparency that will help us understand the problems our universities face¡­it casts light on what is working and what is not¡±.

We welcome this greater transparency. Universities should be held to account by, and on behalf of, their students.

But there are two problems with using LEO as the light source. The first was highlighted by Gyimah ¨C what you earn isn¡¯t a measure of what you¡¯ve learned.

The second is that, even when it comes to linking career outcomes with what you studied and where, LEO alone isn¡¯t up to the job. There are too many shadows.

New by London Economics, which was commissioned by GuildHE, clearly shows that LEO data?are not the single, robust evidence base that policymakers or the regulator should rely on.

ÖйúAƬ

It shows that LEO has serious gaps. These gaps mean that assumptions about the impact of a particular university or course on an individual¡¯s earnings can be misleading or significantly exaggerated.

Critical information left out of LEO include personal and family details, information about where in the UK you are working and details on graduates who did not go to secondary school in England.

LEO covers only graduates¡¯ incomes and employment in the early stages of their careers, and the information on earnings from self-employment is incomplete.

ÖйúAƬ

The London Economics also challenges the validity of directly linking the repayment of student loans to a particular degree. This oversimplifies the measure of true economic value. For example, the long-term economic contribution of creative arts graduates to business, industry and society is much wider than the amount of student loans repaid to the Treasury.

The flaws in the data mean that LEO should carry a prominent health warning to students and parents, and especially to anyone influencing policy decisions about tuition fees, differential fees and the public funding of ÖйúAƬ.

So when Gyimah said that he ¡°[does] not see the value of a university education solely hanging on its contribution to one¡¯s lifetime earnings¡±, he?was 100?per cent correct.

To paraphrase Gyimah¡¯s old boss, David Cameron, ¡°we agree with Sam¡±, and when we say that LEO data?are not the answer and should?not?drive the policy, we hope he agrees with us.

ÖйúAƬ

Gordon McKenzie is chief executive of GuildHE, which represents the UK¡¯s specialist and smaller ÖйúAƬ providers.