A report just published by the Royal Statistical Society looks at the methods used (by newspapers and others) to construct league tables of universities and their departments, based on the results of the UK Research Excellence Framework (REF 2014).

The report's authors include David Firth (Warwick), as part of an RSS working group with Paul Fearnhead (Lancaster University) and Jon Forster (University of Southampton).

A conclusion of the report is that ideally a "quality per researcher" measure should be used for such tables, with all REF-eligible researchers counted and with "quality" being more closely connected to the HEFCE funding formula --- i.e., more closely than the commonly used GPA (grade point average) scoring of quality levels.

The RSS working group's report also makes some related suggestions for the funding councils themselves, about REF submission rules and reporting; these include elimination of the distorting “threshold” effect of REF 2014's formula for the number of Impact Case Studies in each submission.

The report is available via http://www.rss.org.uk/press, or as a direct download at http://www.rss.org.uk/ref .