Guy Nason of the Royal Statistical Society has set out concerns about the teaching excellence framework in his opinion article “In terms of statistical prowess, the TEF does not even get bronze” (7 June), although in practice his concerns are about the technical consultation document on the subject-level TEF. Unfortunately, the argument that he sets out confuses the subject-level TEF proposals – which are in consultation and pilot stage, and thus not yet final – with the TEF overall and, by extension, the year-three results.
Readers of Nason’s piece will not necessarily understand the relationship between the subject pilots and the technical consultation: both are designed to help develop a framework for the robust assessment of the relationships between institutional practices and student outcomes, drawing not, as the article sometimes implies, on raw data but rather on data that control for differences in institutional subject and student composition. Through the pilots and the technical consultation, we are seeking to develop a model that is robust and sensitive to subject and institutional differences.
It is worth pointing out that the TEF draws on a range of data sources, of which the metrics – the statistics – are only one. The TEF assessors, both at institution level and in the subject-level pilots, use the statistics as a starting point for the exercise of judgement: this has been clear from the outset of TEF.
Sir Chris Husbands
Chair, teaching excellence framework
Vice-chancellor, Sheffield Hallam University