Last-minute REF changes ‘show how far things drifted off course’

Major last-minute changes to the UK’s Research Excellence Framework should raise questions about how the controversial reforms were adopted in the first place, according to experts.

UK Research and Innovation (UKRI) has announced a major rolling back of changes to the 2029 audit, bringing it more in line with the 2021 iteration.

In a move heralded by politicians and research-intensive universities, organisers will increase the importance of the main “quality” component of REF 2029, known as contributions to knowledge and understanding, to 55 per cent.

The component was due to be downgraded to 50 per cent from the 60 per cent weighting in 2021, with the people, culture and environment (PCE) section increased to 25 per cent.

It follows a concerted campaign to undo changes made in June 2023 that were intended to use the REF to address long-held concerns about research culture within UK universities.

“We have always been clear that REF should maintain a core focus on excellence, so we’re pleased to see strong recognition for outputs and impact in the new weightings,” said Chris Day, chair of the Russell Group and vice-chancellor of Newcastle University, welcoming the changes.

“It’s right that REF evolves, but that it does so in a way that builds on previous successes and maintains rigour, reduces complexity and limits unnecessary burden.”

John Womersley, a former chair of the Science and Technology Facilities Council, said the late-stage updates should prompt questions about why the REF had adopted reforms that were clearly politically unacceptable.

“This is about as late as changes can be made without causing a delay to the whole REF,” said Womersley, now a senior adviser at the University of Edinburgh.

“These changes are going to be presented as a mid-course correction, but the reality must be that this work stream had drifted really far off course for this kind of dramatic action to be felt necessary so late in the process,” he said.

Womersley said he could “only speculate” as to what had happened “but it seems plausible that initially generic and well-intentioned ideas about research culture became captured by those with strong views about the need to use the REF as a way to engineer social change in universities in a particular direction”.

“No doubt they thought this was the morally correct thing to do, and believed that they had UKRI’s management’s support in doing this, but I can only assume they were working inside something of a bubble of like-thinking people,” he continued.

“Events suggest to me that at some point not so long ago someone senior outside the bubble, either inside UKRI or [the Department of Science and Technology], discovered what was going on in detail and reacted strongly and negatively.  We also know that universities have been voicing concerns too,” said Womersley.

There was also disappointment among some academics that the PCE – now to be known as “strategy, people and research environment” (SPRE) – had retained a higher weighting than in the 2021 exercise, with little clarity on what the “strategy” part will entail.

Alice Sullivan, professor of sociology at UCL, said she welcomed the reframing but it was “disappointing that Research England have made such a minimal concession in the face of opposition to the proposed PCE element of REF 2029 from everyone from vice-chancellors to grassroots academic freedom organisations”.

“The content of SPRE remain completely unclear, and it is not obvious how Research England can reasonably decide how much weight should be given to this without first deciding exactly what it is and how it will be measured,” commented Sullivan, noting the proposed lack of consultation over which metrics and methods will be used to assess this area.

“These are difficult times for the higher education sector, and we need to be able to have a meaningful and fully-informed discussion on REF,” said Sullivan.

Speaking at a Universities UK conference after the changes were announced, science minister Patrick Vallance defended the importance of including SPRE in the exercise.

“Diversity and inclusion is not some ‘nice to have’ – it is fundamental to having better outcomes and how research systems work…It’s not a ‘nice-to-have’ – it’s a requisite for the system we want to have,” he said.