The RAE may be replaced by a metrics-based system - but how will it work? How should you play your research hand when you don't know the rules? The simple answer is not to try, says Harriet Swain
Ka-ching! Another article in an obscure academic journal, another notch in your research assessment exercise rating, and another few pounds for your university. You've finally got the hang of this RAE business... just as it's about to change.
The Government's announcement in this year's Budget that it wanted to scrap the RAE in favour of a system that relies on metrics is likely to have caught anyone skilled at working the RAE system on the hop. How do you beat a system when you aren't sure exactly what it will be?
Stop trying too hard. Tom Sastry, a senior researcher for the Higher Education Policy Institute and joint author of a paper on the use of metrics to allocate research funds, says: "'Don't try' is still good advice. Setting a strategy now for the next ten years isn't a good idea."
It is advice echoed by the Higher Education Funding Council for England, which tells researchers to "continue to carry out research of the highest quality within the strategic framework of your higher education institution".
On the other hand, says Gary McCulloch, dean of research and consultancy at the Institute of Education, University of London, any change in the system will eventually affect how people run their research. "Many people have said that the RAE has had consequences on people's behaviour and so will the metrics system."
His main piece of advice, particularly for those in social sciences and the humanities, which he says are likely to be less suited to metrics, is to influence debate early on about the kind of system that will be introduced.
"They should find out how they can effectively have a say, rather than having to respond to what is coming," he says.
This will also help ensure academic ownership of the new system. "Whatever the problems of the RAE, it is at least a peer-review system," McCulloch says.
Stuart Palmer, chair of the Russell Group's research group, says that the practical steps universities can take to improve their chances under the new system will depend on the type of metric or combination of metrics used. If one measure used is the success of grant applications, universities will need to work harder with individual staff to help them make applications. This would include having internal scrutiny panels, he says. Managers may recruit academics with an experimental rather than theoretical background, as they are likely to bring in more grant income.
The Government's likely encouragement of applied research means that universities should probably focus on getting industrial grants and exploring industrial collaborations through a dedicated industrial links office, he says. This does not mean neglecting people who will get big grants from research councils because the system may be weighted more heavily towards them. Rather, universities will need "a portfolio of people" - some with industrial, some with research council pulling power.
Sir Gareth Roberts, author of a 2004 review of the RAE for the funding councils, says academics will need to think harder about the impact of their research, either in terms of citations or use by industry or business, and of how this impact can be measured.
Ian McNay, professor emeritus of higher education and management at Greenwich University, recommends defining research widely, so that research and development consultancies are included in income metrics, and the base for matching funds is maximised. Networking will still be important in the new system, he says, but the key networks may be different. "Get known among research users who are potential funders. Build on networks from other activities - professional development work or alumni, for example."
He also suggests shifting your publication strategy. "The RAE discouraged publishing in professional journals and drew complaints from professional bodies and employers/users/practitioners," McNay says. This is likely to change under the new system. "If practitioners see that you do useful work and talk their language, they will be better disposed towards you than if you write only arcane stuff for obscure journals."
But Palmer says that staff must be encouraged to publish in a more focused way in the top journals. Under the RAE's peer-review system there is more scope for work to be judged on its quality wherever it is published, he argues, whereas under a metrics-based system the standing of the journal will be all important.
Another possible measure under the new system is the number of PhD students who complete their degrees. Palmer says that will mean developing a graduate school support system.
Sastry says that whatever system is introduced be influenced by the 2008 RAE because performance in that is likely to be used as a future benchmarke. The temptation could be for managers to blow the next ten years' recruitment budget this year to ensure they have as many star researchers in place as possible. It is an unwise approach, he says: while RAE rules limit the number of publications submitted per person, a new system based on research income or citations would mean no limit to the amount of money an individual could bring in.
He warns that the danger of thinking too tactically about what is likely to happen is that the Government could throw in a new metrics comparator at the last minute that would destroy the best-laid plans. "The best way of trying to do everything is to give good researchers freedom to try the best they can," he says.
Further information
Higher Education Policy Institute:
TOP TIPS
- Concentrate on managing research rather than the research assessment process
- Contribute to the consultation on metrics
- Think of research impact
- Build on links with industry
- Do as well as you can in 2008