Informatics Approach to RAE-2008 Preparations

In order to minimise the amount of busy-work and time-wasting associated with RAE 2008 preparations, we need to see it as an opportunity, not a burden. For instance,

1. Rather than invent new, one-off mechanisms, solely for RAE purposes, we should reuse and extend existing mechanisms, or, if there are no suitable existing mechanisms, we should invent new mechanisms that will be of long-term benefit.
2. Where important, but non-urgent, tasks are being neglected to our detriment, we should use the RAE to raise their priority.

• Our School database is a valuable resource that contains a lot of the information required for our RAE 2008 submission. Some of the information that it does not yet contain, e.g. uptodate information on our grant portfolio, would be a useful addition, for instance, it would enable summary information on grant income, distribution, etc to be displayed for those in a research leadership role. I intend to use the School Database as the main repository for collecting RAE relevant information. Ken Dawson and Rosemary Soutar are adapting it to allow additional information to be stored and are providing a variety of ways to view this information. Jean Bunten will be inputting RAE specific data to it.
• It is important to the academic health of our staff that we regularly appraise them, but this is a non-urgent task that is indefinitely postponed. By using the appraisal mechanism in our RAE preparations, we can raise its priority. I intend to use appraisal to identify RAE borderline cases and provide advice on improving their performance. This will be of benefit both to the career of the individual and to our RAE preparations.
• The online availability of our publications is patchy and non-uniform. We have a technical report series, but it is seen as unwieldy and is dramatically underused. By simplifying the use of the technical report series and using it to record our RAE 2008 submissions, we avoid inventing a new device to add to the confusion and encourage people to use the technical report series. It is likely that there will soon be a requirement to make all publications arising from research council grants freely available online. The technical report series could adapt to address this requirement by providing a uniform portal to all our publications. This would simplify access to our publications and present a much better image to the rest of the world.
• College ran a mini-RAE exercise over Summer 2004. This required all AT staff to fill in a pro-forma giving a large amount of RAE-relevant information. Informatics successfully resisted taking part in this exercise on the grounds that we planned something less onerous to staff but more useful for our RAE preparations. Our alternative plan is now underway. Forms similar to the College ones are automatically generated from the School Database. Much of the information required on this form is already in the Database and most of what isn't yet there will be input by support staff. There will remain a residue of material that must be provided by individuals -- mainly publication and esteem indicator information. This will enable those preparing for RAE 2008 to get snapshots of the situation, for a varying pool of potential Cat A candidates, whenever they want. If we had taken part in the College exercise, some of the incentive to do this information gathering properly and some of the goodwill of School staff would have been squandered.

For a variety of pragmatic reasons, having a few large research groups is optimal in an RAE submission. Panels will tolerate quite loose groupings, as long as there is some reality to them. Our six research institutes provide exactly the right level of granularity and community, so we will use them as the basis of our 2008 submission.

However, it is appropriate to have some sub-group structure beneath the institute level. For many kinds of information gathering, it is the research sub-group that is the most appropriate source. I am identifying leaders of Informatics research sub-groups within each of our research institutes, and intend to ask for their help in the RAE 2008 preparations. Here are some of the areas I intend to ask them to help me with.

• Eligible Staff: We need to identify the staff who might be eligible for submission to RAE 2008. We can then start to monitor those staff to identify those on the borderlines who might benefit from some mentoring. The DoIs and sub-group leaders will both be consulted to identify borderline cases. We can then alert the appraisers of the identified staff to hold an appraisal as a matter of urgency. The RAE webpage for the appraisee would be one of the inputs to that appraisal. A joint action on appraiser and appraisee would be to estimate the current and the target RAE star grade for the appraisee. The appraisal report should contain a plan for moving from the current to the target grade. To assist with this process, I have developed a discussion document on what the star rankings might mean. Human Resources have confirmed that this is an appropriate use of the appraisal mechanism.
• Impact Indicators: In 2001, the RAE 5(a) played a central role in describing the research of each department. A common mistake was merely to list the research interests of each group. The RAE panels were much more interested in the impact of the research. The Edinburgh 2001 submission was focussed around our research impact and will provide an exemplar for 2008. To ensure that we follow this lead, I intend to collect a few key impact indicators from each of our sub-groups and construct the 2008 RAE 5(a) equivalent around these. I will be asking sub-group leaders to identify the impact of their sub-groups. To this end, I am drawing up a list of impact indicator exemplars.
• Publication Outlets: Appraisers, sub-group leaders, PIs and DoIs will want to advise their mentees on publication outlets. It is important for both the RAE and the individual's career development that papers should be published in outlets with the strongest reputation possible for that paper. Coming to a collective view on the relative merits of the outlets in each field will be a useful resource in this process. Sub-groups probably define research fields that are sufficiently small and coherent for this exercise to be carried out. In some sub-groups the situation will already be so clear that the exercise is trivial; in others, it may require more work.

I welcome suggestions for improving our RAE preparations and for using RAE 2008 as an opportunity rather than a burden.