31 January 2019
“The REF is a process of expert review.” This statement at the start of the 'Panel Criteria' document is an important reminder that the academic community is responsible for making the assessments of research quality that are central to REF.
In undertaking the task of revising and updating the key documents, especially the 'Panel Criteria', we have therefore been mindful of the need to provide criteria that are as clear and consistent as possible, but which also recognise and reflect disciplinary differences as appropriate. The sub-panel members have thus spent a lot of time discussing the fine details of issues such as defining ‘research independence’, which can mean different things in different research cultures. In this case, we were able to agree guidance across Main Panels C and D, with Main Panels A and B taking a slightly different approach.
We also wanted to provide reassurance on some key issues, especially where there were changes from 2014, following the Stern Review. For example, the Stern report recognised that interdisciplinary research was not disadvantaged in the scoring in REF 2014 but that there was concern that there may be caution in submitting such work. There are therefore more formal systems in place in REF 2021 intended to reassure the community that interdisciplinary research will be assessed in an appropriate and equitable way. This includes academic colleagues on sub-panels and main panels who will specifically advise on the assessment of interdisciplinary outputs. But it is important to be aware that the sub-panels themselves are rarely single disciplines. They include people with a range of disciplinary expertise, and theoretical and methodological approaches. So we are confident that the final membership of the sub-panels (there will be a further round of appointments in 2020) will have the expertise to assess interdisciplinary research outputs.
Another area where we wanted to provide some reassurance concerned the criteria for double-weighting of outputs. It was felt that in REF 2014 there may have been some reluctance to request double-weighting, in case this was not agreed. So we have set out some guidance – again in common with Main Panel D – to help colleagues decide whether to request double weighting. There is also an opportunity to put forward a reserve output, so the risk is minimised.
As noted above, the sub-panel structures reflect and include multiple disciplines. But some areas of research cut across these panel boundaries. In Main Panel C, the area of criminology was identified as potentially located across three sub-panels in particular – law, sociology, and social work & social policy. We have therefore strengthened, and made more transparent, the processes that were in place in REF 2014 to support research in this area, and will report on this in the post-REF review.
The consultation exercise was very important in alerting us to these and other issues, and we look forward to further discussions with academic colleagues. REF 2014 showed that research excellence can be found in a variety of different contexts and we hope colleagues across the social sciences community will feel confident in presenting their submissions to this process of expert review.