Nearly a fifth of academics submitted to the 2014 research excellence framework were early career academics with fewer than the standard four outputs.
The figures are contained in a breakdown of submission data for the REF, published by the four UK funding councils on 13 February.
As previously announced, 155 institutions made submissions to the REF by the 29 November deadline. A total of 52,077 full-time equivalent staff were submitted, compared with 52,401 to the 2008 research assessment exercise.
The REF figure amounts to 28 per cent of the 185,585 academics who were in post in 2012-13, while those submitted to the last RAE represented 30 per cent of the 174,945 staff in post in 2007-08, according to figures from the Higher Education Statistics Agency.
The biggest decline in volume of academics submitted, 5 per cent, was to main panel D, which covers the arts and humanities. The number submitted to main panels A (biomedicine) and C (the social sciences) both fell by 3 per cent. However, submissions to main panel B, which covers the physical sciences, rose by 9 per cent.
The highest number of staff, 14,415, was submitted to main panel C. The smallest number, 10,698, was submitted to main panel D.
Meanwhile, the total number of research “outputs” – mostly papers or monographs – submitted to the REF fell by 11 per cent, to 191,232. The largest fall, of 16 per cent, was to the biomedicine panel. Outputs to the arts and humanities panel fell by 15 per cent, those to the social sciences panel shrank by 11 per cent and those to the physical sciences panel slipped by 3 per cent.
These declines are reflected in the fact that the average number of outputs submitted per member of staff declined from 3.75 in 2008 to 3.41 in 2014. As in 2008, the lowest number of submissions per staff member (3.32) was to the arts and humanities panel, while the highest (3.53) was to the physical sciences panel.
One possible explanation for the falls could be the introduction to the REF of criteria that made it easier for staff to submit fewer than the standard four outputs owing to various “individual circumstances”, such as a period of illness or maternity leave, or having only recently begun academic careers.
In total, 16,361 of the total staff headcount of 56,089 – rather than full-time equivalents – were submitted with reduced outputs: 29 per cent of the total. Of those, 10,099 – or 18 per cent of all academics submitted – were early career researchers. Another 5 per cent of staff submitted reduced outputs, because of maternity, paternity or adoption leave.
The new element of the REF was the requirement to submit impact case studies. In total, 6,975 were submitted. In line with the volume of academics submitted, the greatest number, 2,040, were submitted to the social sciences panel.
However, main panel D (arts and humanities) received slightly more than main panel A and only slightly fewer than panel B, despite submitting substantially fewer academics. This reflects the fact that humanities departments tend to be smaller than other departments, and small departments are required to submit a proportionally higher number of case studies.
The results of the 2014 REF will be announced in December.