Three-D Issue 29: First signs of how REF will work

Well, it’s called REF 2021, but already in 2017 many feel suitably terrorised and daunted by the research assessment exercise now called REF. The straws in the wind which begin to set out how the exercise will work have come so far in the form of two sets of decisions from HEFCE, the first dealing mainly with the panel structure (and with appointment to panels, which is in train as we go to press), and the second with selection of staff and ‘outputs’ to be submitted for assessment.

The panel structure will be much as for 2014. There are two fewer panels (or sub-panels to be pedantic) than last time. HEFCE have ignored all advice from our neck of the woods and have retained library and information management in the same unit of assessment as communication, cultural and media studies (in sub-panel 34). Film and screen studies, however, though explicitly named, will be included along with music, drama and performing arts in sub-panel 33. This structure will pose many problems for assessors and submitting universities alike, as the statement appearing alongside this article explains. Not least it means that sub-panel 34 will either be very large, or will have too few members on the CCMS side of things to cover the sheer diversity of research in the field, while there may be an unfortunate tendency for sub-panel 33 to be seen as a more natural locus for practice based research.

MeCCSA will be having discussions with both Dinah Birch, who will chair Main Panel D (embracing both 33 and 34) as well as with the chairs of these sub-panels, whose names were published shortly before we went to press. They are Professor Maria Delgado (Royal Central School of Speech, Sub-Panel 33) and Professor Justin Lewis (Cardiff University, Sub-Panel 34). Prof. Delgado, who was then at Queen Mary, University of London), chaired the equivalent sub-panel in 2014, while Prof. Lewis was a member of his equivalent sub-panel (36) in 2014. We will, among other things, be urging them to make the remit and criteria of the sub-panels very clear on these and related issues, when guidance on submissions is published.

This first HEFCE paper also puts great stress on interdisciplinary research, though most of us probably feel this is now an orthodoxy in much of the field we till in; indeed they envisage one member of each sub-panel having primary responsibility for the assessment of interdisciplinary work. The paper makes clear that the Government Department for Business, Energy and Industrial Strategy has been heavily involved in the formulation of the assessment of impact, though, reassuringly, this will comprise 25 per cent of the overall assessment (more than in 2014 though less than might have been the case) along with 60 per cent for outputs, and 15 per cent for environment, (protecting the latter at the expense of the former).

The second HEFCE paper (issued in November) addresses proposals about staff and outputs. The decision arising from the Stern review last year that all staff should be submitted, was broadly welcomed, though with much unease about the manner in which such an apparently simple regulation could be implemented. The HEFCE paper muddies this water considerably. All staff will be submitted who have “significant responsibility for research”. How this is to be defined or regulated remains to be seen. A distinction is made between Category A eligible staff (who meet the core eligibility criteria), and Category A submitted staff (who have been identified as having “significant responsibility for research”). Clear? No, we thought not. It will be up to institutions to operationalise these categories, for example using work allocation formulae, or other indicators (eligibility to be grant holders, supervise doctorates…), and we can expect much further debate and dispute about this. Not surprisingly, in anticipation of this HEFCE coyly say that they will “provide further guidance” on how staff are to be identified after waiting for Main Panels to discuss the matter. The census date for staff will be July 31st 2020.

The decoupling of staff from outputs is also fleshed out a bit further. The number of outputs to be submitted will be based on the total FTE of Category A submitted staff. The number of outputs required will be this number times 2.5. Any work seeing the light of day (“first made publicly available”) between Jan 1, 2014 and December 31st. 2020 will be eligible. A minimum of one output per submitted staff member will be required, and a maximum of five outputs from any one member of staff (not including any on which they are co-author). On portability of outputs (i.e. can a university claim credit for work by someone they appointed the day before the census date – one extreme outcome of the much feared transfer market), HEFCE have now opted for the kind of balance that results in endless complexity. They propose to have a “transitional approach” whereby outputs may be submitted by both the institution employing the staff member at the census date, and also the originating institution where they were when the output was “demonstrably generated”. Once again Main panels are faced with the task of resolving these matters, and we can expect further clarification (or not) in the months to come. Impact is much like last time. One case study plus one more for every 15 staff submitted will be required – in other words a minimum of two case studies per submission.

More guidance will be published in the summer of next year, with the final guidance to be issued in January 2019. MeCCSA will be consulting with and informing members regularly as decisions are made, and as we discuss the preparation of the Guidance with the relevant panel and sub-panel chairs. Dinah Birch will be attending and speaking at our annual conference in January, and we will keep members informed of developments via the website, and where necessary, the jiscmail list.

Posted by Einar Thorsen