Is the benefit of the REF really worth the cost?

Simpler options are imperfect but perhaps no more so than the panels’ unavoidably cursory ‘peer review’ of submissions, says Dorothy Bishop

Published on
April 28, 2021
Last updated
April 28, 2021
Divergent paths - one direct, the other convoluted, symbolising two approaches to the REF
Source: Getty

POSTSCRIPT:

Print headline: Do the benefits of the REF really justify its cost and complexity?

Register to continue

Why register?

  • Registration is free and only takes a moment
  • Once registered, you can read 3 articles a month
  • Sign up for our newsletter
Please
or
to read this article.

Related articles

Reader's comments (9)

Getting grants used to be out of financial necessity - if a specific (ambitious) research project cannot proceed without funding, then we applied for it. Since grant awards have become a performance metric, all academics are now pressured to get grants for REF and promotions. How can this type of reward structure not lead to financial inefficiency of UKRI's funding???
I agree, I find this very frustrating. I'm a mathematician - other than my computer, I just need pen and paper. Why do I have in, with respect to grant income, the same requirements as someone who employs teams of research assistants, or someone in physics who runs a lab?
The intention of this article is good. More efficiency in the REF would be a good thing. But I am not sure the proposed cure would be suitable. Giving each and every institution the incentive to hire as many bodies as possible would increase the speed of the monopoly game that is the competition between universities. If my University is not assessed on paper quality and instead on Department size, they will do what they can to recruit more clueless student customers with poor education from developing countries just to get the income required to hire those new bodies for the REF. Instead, we need to maximise quality per Department, not quantity. The recent suggestion here on THE to use citation metrics of journals would do this trick. They are quick and easy to deploy, and while they may be off for each particular article, these effects will balance each other out statistically because there are many submissions and because the journal's impact factor is a consistent estimator of scholarly article impact.
I totally agree with Professor Bishop's premise that the costs of assessment now outweigh the insights gained. We're also now beset by the problems of "gaming" the system at everything from the individual researcher to the institutional level. My own view is that this isn't just about research performance and that similar dramatic simplification is needed in TEF and KEF so that academic staff spend more time doing academic work and less time accounting for the doing of academic work ... https://www.timeshighereducation.com/blog/radical-rethink-uks-excellence-frameworks-needed. If we don't pause and rethink now, another 5-8 years will pass and the opportunity cost will continue to grow.
I have a contact at another Russell group university who for the past two or three REFS did the following. With and friend/colleague, after work, he went through all of their depts. intended REF submission and using a simple Excel table calculated a score based on the Impact factor of each publication and the citation index. They then printed out the Excel sheets, sealed them in a envelope, which was dated & signed across the flap. These envelopes were not opened until the official score emerged and in both cases their score were only a second decimal place different. Cost of their effort - about 3hrs and £10 (six pack of 'refreshment' ; somewhat cheap and just as effective as the 10's of millions for REF, which as been a massive waste of both money and academic time!
But you see.... think about all those people who are employed, paid, and get reputational benefits from being appointed as REF panelists/lead?? What on earth will they do if they lose all those benefits??
I agree with the article completely. The current REF has built a massive internal and external bureaucracy for very little return. Peer review is not really a reliable method of judging articles that have already been through a review process. In addition, the difference between the top departments are marginal. It is just a huge waste of time and effort for a marginal outcome.
REF is a poker game with high stakes for VCs and research directors. The impact on academic life (publish 4* or perish) should not be underestimated. A simpler way of incentivizing academics to achieve research goals would be to ...
In some fields almost all of the papers that get into the top journals use US data. Partly to do with realiability and availability of data, but also becuase most of these journals are US based. Is that proper use of UK tax payers funds? Coud the REF not give more weight to research that is primarily aimed at studying issues that are relevant to the UK?

Sponsored

Featured jobs

See all jobs
ADVERTISEMENT