Logo image
Back
Development of a tool to accurately predict UK REF funding allocation
Journal article   Peer reviewed

Development of a tool to accurately predict UK REF funding allocation

S. Al-Janabi, L.W. Lim and L. Aquili
Scientometrics, Vol.126(9), pp.8049-8062
2021
DOI: https://doi.org/10.1007/s11192-021-04030-w
url
Link to Published Version *Subscription may be requiredView

Abstract

Understanding the determinants of research funding allocation by funding bodies, such as the Research Excellence Framework (REF) in the UK, is vital to help institutions prepare for their research quality assessments. In these assessments, only publications ranked as 4* or 3* (but not 2* or less) would receive funding. Correlational studies have shown that the impact factor (IF) of a publication is associated with REF rankings. Yet, the precise IF boundaries leading to each rank are unknown; for example, would a publication with an IF of 5 be ranked 4* or less? Here, we provide a tool that predicts the rank of each submitted publication to (1) help researchers choose a publication outlet that would more likely lead to the submission of their research output(s) by faculty heads in the next REF assessment, thereby potentially improving their academic profile; and (2) help faculty heads decide which outputs to submit for assessment, thereby maximising their future REF scores and ultimately their research funding. Initially, we applied our tool to the REF (: Institutions Ranked by Subject (2014). https://www.timeshighereducation.com/sites/default/files/Attachments/2014/12/17/g/o/l/sub-14-01.pdf.)) results for Neuroscience, Psychiatry, and Psychology, which predicted publications ranked 4* with 95% accuracy (IF ≥ 6.5), 3* with 98% accuracy (IF= 2.9–6.49), and 2* with 95% accuracy (IF= 1.3–2.89); thus indicating that researchers wishing to increase their chances of a 4* rating for the aforementioned Unit of Assessment should submit to journals with IFs of at least 6.5. We then generalised these findings to another REF unit of assessment: Biological Sciences to further demonstrate the predictive capacity of our tool.

Details

Metrics

35 Record Views
5 Times Cited - Web of Science

InCites Highlights

These are selected metrics from InCites Benchmarking & Analytics tool, related to this output

Collaboration types
Domestic collaboration
International collaboration
Citation topics
6 Social Sciences
6.238 Bibliometrics, Scientometrics & Research Integrity
6.238.166 Bibliometrics
Web Of Science research areas
Computer Science, Interdisciplinary Applications
Information Science & Library Science
ESI research areas
Social Sciences, general
Logo image