Reflections on REF main panel B’s input into REF 2021

To everyone working on the research excellence framework, the publication of the final “Guidance on submissions” and “Panel criteria and working methods” will be something of a welcome relief. Those working on student recruitment, procurement, data protection, planning, research funding and many other areas will be able to attest right now that there is nothing more frustrating to getting on with the job than working in the dark.

I for one welcome the new REF rules; it is no secret that I was in favour of removing staff selection as a means of ending the “gaming” of the REF exercise. I am glad to see that in the final “Guidance on submissions”, staff circumstances have been retained and enhanced.

Although there was a persistent argument that individual circumstances will lead to an increase in burden for institutions, I believe this is a burden worth carrying to ensure that REF has robust measures to promote equality and diversity. My feeling is that the flexible approach to the output pool may well reduce the need for higher education institutions to submit requests for unit reductions and the increased burden of staff circumstances will be far less than anticipated.

That is not to say that the new rules do not come without challenges. There has been much discussion about how the new rules will increase the amount of outputs that will need to be assessed before output selection. This is nowhere more relevant than at my institution, UCL. For REF 2014 we submitted 9,416 outputs. 

As an institution that produces around 11,500 journal articles alone per annum, the decoupling of staff and outputs will require us to internally assess the vast majority of all outputs we have generated. 

Some might be tempted to fall back on metrics as an easy evaluation route, and the ongoing challenge to halt the practice of correlating journal impact factors to the merits of a specific author’s contributions continues. At UCL, we have signed up to the San Francisco Declaration on Research Assessment (DORA) and as panel B chair one of my challenges will be in ensuring that metrics are used only in a responsible and intelligent way that complements peer review. 

Another challenge for panel B will be the change from four to one unit of assessment for engineering. Managing units of such scale is always testing. The rule that there will be no multiple submissions in UoA12 was widely discussed both within the sub-panel and nationally (including at a town hall meeting at the Royal Academy of Engineering). I think it is a progressive and most welcome development. 

This is not to say that the REF is all about challenges. What excites me most is of course the research and getting up-close to the work that makes the UK’s university sector one of the best in the world. I look forward to reading those new impact case studies in panel B, and also to seeing how those statements presented previously have matured to provide new and even more impressive impact. Within the research environment, I look forward to seeing the results of institutional infrastructure investment, and progress in early career researcher development and equality and diversity practice.

Good luck to all involved in the next stages of the REF. November 2020 will be upon us sooner than we imagine!

This blog also appears on the UK Research and Innovation blog. More views from chairs of panel A: medicine, health and life sciences; panel C: social science and panel D: arts and humanities are also available.

David Price is vice-provost (research) at UCL and chair of main panel B: physical sciences, engineering and mathematics for REF 2021.