Star-studded future starts here

When it comes to the research assessment exercise, quality sells itself.

Still, a touch of rune-reading and a bit of self-publicity do no harm, says Harriet Swain

As everyone knows, the research assessment exercise is what being an academic is all about. And even though the next one isn't until 2008, you will already be wondering whether you could be doing more to prepare.

In some ways it is already too late. The research that will count is probably already well under way. In any case, insists Gareth Roberts, who conducted a review into the RAE in 2002, your priority should be to carry out the best quality research you can and trust the system to identify that quality.

Michael Prestwich, professor of history at Durham University, chaired the history panel for the past two RAEs. He says exaggerating your achievements will get you nowhere. "Panels are pretty well informed," he says. "You need to present the truth."

But how do you ensure it is the kind of truth they want to hear? Read the criteria, Prestwich says. "We followed the criteria very closely last time, and I'm sure the same will apply this time."

Ed Hughes, RAE manager at the Higher Education Funding Council for England, says it is important to pay attention to the ways the assessment criteria of each panel are likely to vary when they are published for consultation.

Main panel areas will weight the elements that make up the final quality profiles - research outputs, environment and esteem indicators - to different degrees.

Ian McNay, professor emeritus of higher education and management at the University of Greenwich, advises reading the rules and the runes. "The second is more difficult to penetrate and more risky," he says. "It would involve reading the output of panel members to see if there are dominant themes or approaches. Aim for a fit."

Larry Ray, professor of sociology at Kent University and head of the School of Social Policy, Sociology and Social Research when it was awarded a 6* in the 2001 RAE, agrees that it makes sense to find out about panel members before making submissions.

But Rama Thirunamachandran, Hefce director of research, advises against "distorting research". He says that the 2008 RAE has been designed to be flexible enough to measure whatever research is submitted - whether practice or research based. But while McNay concedes that the design of the 2008 exercise is an improvement on earlier ones, he remains cynical, advising sticking to mainstream approaches, at least in his subject, social sciences. "Challenges to the mainstream get lower grades, possibly because they imply that panel members have been mistaken in their work," he says.

Hughes says that this time some panels may ask for a brief description of why a particular output has been included. This might be especially helpful in identifying the significance of a piece of applied or practice-based research - for example, the extent to which something has contributed to policy development or a new industrial product or process.

He adds that panels are particularly keen to identify and support sustainability and vitality in the work and groups of people submitted.

"They want to get a sense that there is a long-term plan that can sustain the research activity reported in the submissions," he says. He hopes this will encourage inclusion of those starting out in academia.

Ray advises looking at what you said about yourself last time and to what extent you have continued on the same path. "If you have changed in any way, a story has to be told about that," he says. He attributes the success of his department in the last exercise to the fact that its submission conveyed a sense of momentum and was able to point to a number of theoretical and empirical areas it had opened up since the last RAE.

Roberts advises paying careful attention to the emphasis placed on the impact or significance of your work. He suggests that this may mean securing endorsements from the bodies that have benefited from it. This approach may even apply to books. "Most people can write a book - the point is what reviewers think of it, so you could have external endorsements of what you have written," he says. "You need to say to yourself, 'If I was on the receiving end, how would I know this was very high-quality research?'"

For McNay, the emphasis should still be on rigour in methodology rather than application. "Methodology can be international, applications are often local and so deemed not to be of international or world-leading quality," he says.

He also advises getting claims to excellence and originality high up in your articles. "Panel members are busy and first impressions count," he says. He suggests that there is likely to be more intense scrutiny this time for the star profile as opposed to sampling for the general grade, so more of the output should be read. This means that those managing the submission should be severe about anything that threatens the whole.

Meanwhile, he stresses the importance of getting round the conference circuit. The more you are out and about, the more likely that one of the panel members will have heard of you and have been impressed. Even if they insist they are immune to this, it is bound to influence their judgements.

"So get your best people out there and publicise, publicise, publicise," he says.

But beware of cliche. Prestwich recalls that in the 1996 assessment "an awful lot of departments were 'vibrant'". He says he mentioned this in his post-RAE report and it was striking how few departments described themselves as vibrant next time around.

Further information:
Guidance to Panels document instructing the main panels and sub-panels of the 2008 research assessment exercise on developing assessment criteria and working methods: www.rae.ac.uk/pubs/2005/01/

Sir Gareth Roberts' review of the RAE: www.ra-review.ac.uk/reports/roberts.asp

TOP TIPS

Do good research

Consult and obey the criteria of the panels

Show a long-term research profile

Remember, first impressions count

Get known