---
title: The Stern review of REF
layout: post
---
<a href="https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/541338/ind-16-9-ref-stern-review.pdf">Lord Stern's review of the Research Excellence Framework</a> is out today in the UK. Not as exciting as the fact that <a href="https://www.martineve.com/2016/07/28/my-book-password-is-out-now/">my book is also out today</a>, I know, but still a marginally important publication, I suppose.

The biggest recommended change in the report is that institutional submissions be decoupled from researchers. In other words, institutions must submit _all_ research-active staff BUT not every researcher has to submit four outputs. An _institution_ has to submit a number of outputs in relation to the number of its staff on research contracts.

This comes with a number of potential behavioural changes, especially when coupled with the impending TEF, and a few areas where things are unchanged:

* Institutions may take less-productive researchers off research contracts to lower their expected outputs.
* Institutions may hire ultra-productive researchers on research-only contracts for submission.
* It _could_ reduce bullying of those without their four outputs if an institution is already feeling secure in its return (although I am unsure of this).
* It doesn't seem to reduce the institutional burden of selectivity in quite the way it would like (i.e. massive internal gaming/mock REFs/critical friends etc.) at the level of unit of assessment. Institutions will still conduct internal appraisals etc.

Also, and this will change HE hiring practices, outputs are _not_ to be portable. In other words, researchers cannot take their publications to their new institution, thereby preventing "poaching".

The good news is that peer review remains and fears about a potential recommended wholesale metricised REF have not come to pass. Metrics are to be used to _support_ panel members' assessments.

Also, impact now features at the institutional level. This could actually be quite positive for interdisciplinary research. I suspect that the recommendation that impact could be "impacts on teaching" will get watered out as there's too much scope for institutions to bend the rules there.

Other recommendations include: slimming environment statements at the unit of assessment level and moving to the institutional level; making REF data and metrics open and accessible; and ensuring that TEF alongside REF doesn't add too much to the burden.

Stern has also, of course, had to contend with the sudden shufflings of the White Paper and the recent BIS demolition by Theresa May.