A metric in the UK¡¯s teaching excellence framework that scores universities on graduate earnings appears to heavily reflect how far an institution is from London, data have revealed.

Newly released data on English universities that would form the basis of TEF assessments this year suggest that institutions in London and the south east are much more likely to be flagged as performing well for graduate salaries.

Meanwhile, most of the lowest scores on the metric ¨C which uses statistics from the government¡¯s Longitudinal Education Outcomes dataset on graduate outcomes ¨C are recorded by universities based in the north of England.

Critics said that the data prove that the metric ¨C which currently is only ¡°supplementary¡± information to judge universities¡¯ teaching performance but which is set to be a core part of assessments in future ¨C is flawed and should not be used without some adjustments for regional wage variations.

³ÉÈËVRÊÓƵ

The analysis was made possible after the Office for Students released .

It shows that out of 43 English institutions that get a ¡°double positive flag¡± for the share of graduates earning above the national median wage for 25- to 29-year-olds, well over half (28) were in London or the south east and none were in the north west. At the same time, no universities in London or the south east score a ¡°double negative flag¡± for the metric, but more than half that do are in the north.

³ÉÈËVRÊÓƵ

Separate graduate destination data suggest that the only two institutions in the north of England that receive the double positive flag on the metric ¨C the University of Leeds and Durham University ¨C simply have relatively high numbers of graduates going on to work in London soon after leaving.

Tim Thornton, deputy vice-chancellor of the University of Huddersfield, one of the universities that has a double negative flag and where data show that about 70?per cent of graduates work in the local region after leaving, said that using such a metric would simply be ¡°ranking local labour markets¡±.

¡°There¡¯s a real danger of damaging unintended consequences,¡± he said. ¡°Higher education providers are effectively being incentivised to send graduates to London and the south east, rather than working with local employers.¡±

Paul Youngson, head of planning and timetabling at Huddersfield and an expert on TEF data, said that he was ¡°surprised¡± that the metric ¡°makes no attempt to take the large, known variations in earnings in local labour markets into account¡±.

³ÉÈËVRÊÓƵ

¡°We know that the majority of graduates work in the local region of their university, and we know there is a difference of about ?4,000 per year between the highest-paying region and the lowest; and to take no account of this doesn¡¯t make sense,¡± he said.

An OfS spokesman said that the median salary used in the metric ¨C which according to is ?21,000 ¨C was ¡°below the starting salary of professional, socially valuable graduate jobs such as nursing, midwifery or teaching¡±.

¡°Earnings do vary across the country, reflecting differences in job opportunities, productivity levels and the local cost of living, and this will have some bearing on earnings outcomes,¡± he added, noting that the Department for Education was ¡°exploring whether and how this can be taken into account of in some future presentations of the data¡±.

simon.baker@timeshighereducation.com