Staff Record

All five of the case study institutions used their institutional HR systems to source the majority of the data for the staff record. The HR systems in use were Agresso, Trent, Bond, and Birmingham's own in-house development.

As anticipated, meeting the majority of the information needs for the REF Staff was straightforward but three fields were problematical to all five case study institutions. In the main this was because the HR databases did not hold:

• Early Career researcher
• Previous Institution (for eligible staff arriving during the survey period)
• Destination Institution (for eligible staff leaving during the survey period)

In the case of Previous/Destination institution, universities had not previously had a clear operational reason for systematically collecting these data and therefore had to rely on sources such as academic CVs and recruitment documentation. If these fields are included in the final REF specification it may be that HEFCE will make some provision for incomplete data for the first collection. However, it would be prudent for institutions to put mechanisms in place to capture these data as soon as possible.

Three of the institutions voiced the opinion that the definition of 'Early Career Researcher' was ambiguous or hard to track through an individual's employment history and a simpler definition would be welcome.

It was also noted by several institutions that the staff record had a significant amount of commonality with the HESA staff return and that there may be scope for future integration of the REF and HESA staff returns. It may be worth investigating this possibility with senior staff at HESA. However the current census date for the staff return is 31st July. This falls immediately before the major period of staff turnover and any attempt to use staff returns to reconcile with another return whose census date was (even shortly) after the 31st July would inevitably lead to inconsistencies, missing staff data and probably prove impractical. The only completely sure way to tie HESA staff data to an REF return would be to have a common census date. Whether this would be practical and agreeable to all stakeholders would require further investigation.