Poor correlation between metrics and REF ratings, review finds

There is poor correlation between the quality rating of outputs submitted to the Research Excellence Framework derived using metrics and their ratings from peer review, an analysis has found.

The correlation analysis was conducted as part of a review of metrics commissioned by the Higher Education Funding Council for England and led by James Wilsdon, professor of science and democracy at the University of Sussex.

The review, which is due to be published in full on 9 July, shows that for individual authors metrics give significantly different outcomes when compared with the results of the REF peer-review process. “Metrics cannot provide a like-for-like replacement for REF peer review,” Wilsdon told the Data for Policy conference at the University of Cambridge on 16 June.