In an increasingly competitive external environment, research assessment and management is a necessary activity. The University of Bath undertakes research assessment and management at various levels including: whole university, department/school, research group, and individual researchers. These assessment and management activities require expert judgement and we believe that quantitative indicators can frequently inform but never replace this expert judgement.
The University has developed the following set of principles outlining its approach to research assessment and management, including the responsible use of quantitative indicators. These principles draw upon the Leiden Manifesto and the Metric Tide. They are designed to encapsulate current good practice and to act as a guide for future activities.
All research assessment and management at the University of Bath is:
Centred on expert judgement
Research assessment and management at the University of Bath is based on expert judgement. Criteria and/or indicators used must be carefully chosen in light of the purpose of the assessment and context (which may include external evaluations of the University’s research). Where appropriate, quantitative indicators can be used to inform judgements and challenge preconceptions, but not to replace expert judgement.
Set in the broader environment
Those undertaking research assessment and management must be aware of the possible consequences of these activities for behaviour. The potential for research assessment and associated metrics to reflect or introduce bias (e.g. gender) must be addressed. The University of Bath expects everyone involved in research assessment, whether as assessor or assessee, to behave with integrity and to conform to the University’s Code of Good Practice in Research Integrity.
Supported by reliable data
Where used, quantitative data sources are selected for their reliability (i.e. accuracy, quality, transparency and coverage). Any limitations inherent in data sources must be explicitly acknowledged. The aim is to avoid placing undue significance on quantitative differences out of context. Research quality is multifaceted and cannot be captured by a single indicator used in isolation.
Tailored: one size does not fit all
Disciplinary differences in research inputs, processes and outputs have to be taken into account. Any disciplinary biases in indicators used must be explicitly acknowledged and addressed.
Research assessment and management activities should also be tailored to the scale of the research activity being assessed. Particular caution is needed when interpreting quantitative indicators in small scale assessments such as the assessment of an individual researcher.
Assessment criteria and any quantitative data used must be transparent and available (on request) to those being assessed. Those conducting assessments must disclose the data sources used and ensure that researchers can review and correct data about their work.
Owner: University Research Committee
Approval date: February 2017
Approved by: Senate