The results are out. They will be pored over by thousands of academics, administrators, journalists, policy makers, critics and friends of the REF. They will be interpreted in many ways and used for purposes beyond the original intentions of the four funding bodies that steer and sponsor the exercise. But what do the numbers really tell us? Within HEFCE we have analysed the results and here are some of the headlines.


Quality is up

REF expert panels found a significant improvement in the quality of submitted research since RAE2008.

The average proportion of outputs judged to be world-leading (4*) has risen from 14 per cent in RAE 2008 to 22 per cent in REF 2014.

Meanwhile, the average proportion judged to be internationally excellent (3*) has risen from 37 per cent to 50 per cent.

These results are consistent with independent evidence about the improved performance of UK research in international comparative terms. 

Excellence in most institutions 

Although our world-leading work remains concentrated in the most research-intensive universities, the panels have again found such world-leading work in most universities and in submissions of all sizes.

Three-quarters of all submissions were judged to contain at least 10 per cent and one quarter to contain at least 34 per cent world-leading work (4*).

Excellence in all forms of research

All types, forms and outputs of research were recognised as excellent and in many diverse institutions of all sizes.

Three-quarters of the institutions had at least 50 per cent of their output quality graded as internationally excellent (3*) or above.

One-quarter of the institutions had at least 74 per cent of their output quality graded as internationally excellent (3*) or above.

Importance of interdisciplinarity 

Panels observed increasing trends towards inter- and multi-disciplinary research and recognised excellence in these outputs.

Indeed, interdisciplinary research scored equally highly as outputs from a single discipline.

Initial analysis of the impact case studies undertaken by Digital Science and Kings Policy Institute also shows that around 70 per cent of the impact case studies stem from multidisciplinary research.

A real difference to people’s lives

Impact was evident in all domains, including the economy, society, culture, public policy and services, health, the environment and quality of life, within the UK and internationally.

Inclusion of impact in the REF has influenced the overall profile of institutions. Some demonstrated a particularly strong performance.

Most large research-intensive universities scored strongly on impact but outstanding impact was identified in institutions of all sizes.

Impacts reflect institutions’ productive engagements with a very wide range of public, private and third-sector organisations, and engagement directly with the public; impacts of the highest quality are being achieved through numerous pathways including intended as well as unanticipated routes.

Research environments and public funding 

Panels observed an improvement in quality of this element, although not directly comparable with RAE 2008. Research income has increased since the RAE, with increases in all main panels and almost all sources.

The correlation between environment and outputs was higher than that between impact and outputs, but there was still significant variability in the environment of strongly-performing universities. However the very best universities all had excellent environment scores.

This improvement comes in spite of relatively constrained levels of public funding for research experienced over the more recent years.

Our recently published review of QR funding shows the value of the funding that flows from the REF to UK institutions.  It shows how universities are coping with the current financial environment and gives a clear warning about the implications of any future cut to QR.

Progress on diversity

 We welcome the efforts made by institutions to meet new rules on special circumstances and early-career researchers. 

Allowances were made for individuals with special circumstances and early-career researchers to be submitted with fewer than four outputs.

In total 28 per cent of staff (by headcount) were submitted with fewer than four outputs, compared with 13 per cent in the RAE.

The results show that outputs from early-career researchers or those with special circumstances scored just as highly as those from more established staff and those without special circumstances.

So what next?

Once the dust has settled and Christmas dinners have been fully digested, here in HEFCE we will continue our programme of work on the REF.

We will look closely at the numbers to consider what the results mean for QR funding in 2015-16 and beyond.

We will think carefully about what the results show us about the quality and impact of research in UK institutions in order to make a strong and convincing case to the next Government in its post-election Spending Review.

And we will continue our programme of work outlined by Steven Hill which includes a comprehensive evaluation of REF 2014 together with a number of projects looking to the next REF and beyond.

Take a closer look at the numbers and analysis in this post on the REF web-site