Friday, 5 March 2010

Can there be generalisable research via community engagement?

The mainstay of the field research of ISCRI (and one of the constituent departments within it: Ethnicity and Health) is the community engagement model. Over the last ten years and more, UCLan researchers have used this to find the views of particular communities on particular issues (sometimes ‘hard to reach’ communities and taboo issues) especially when this has been needed to focus government policy and resources.

An integral aspect of the process has been the incorporation of social benefits. The research process involves the active participation of members of the target community who thus benefit from training (in research) and education (getting onto a degree programme) and thus an inflow of money into, and ongoing support for, the community in question. In return, local knowledge of the views of local communities can be more easily accessed by people who have local credibility.

Although ISCRI has a good reputation in this area, the approach does carry some essential tensions (about which I had an interesting conversation with Alastair Roy; in fact, warming to his critical theme, he almost seemed intent not so much to praise the community engagement model as to bury it). One obvious problem is that the local researchers are trained for only limited amounts of time. So the expertise ISCRI markets is at one stage removed from the coalface. Or, to put it another way, it is an ongoing task to ensure the quality of that local research.

Second, the focus is narrow and concerns local communities. So the claim has to be that the research approach or methods used for particular communities is bespoke and locally appropriate. But this suggests a regress: in order to measure the validity of research methods in particular local contexts, one would already need a valid meta-level research tool to assess the ground level tool. Sadly that too would need to be local. I’m not sure how this challenge could be addressed in principle although there is a practical fix.

That emphasis on locality suggests a third challenge: the approach potentially yields an understanding of communities in their own local terms. And that may well resist handy generalisation. In a crude slogan (with which I’m unhappy), an idiographic approach resists codification in nomothetic terms.

I suspect that the solution for these problems includes a weakening of the idiology of the community engagement model while retaining its practical virtues. Whilst it might work well for merely local knowledge (if the genuinely idiographic were possible), its broader validation needs an essentially general element. Ali, for example, suggested a model of partnership between experienced researchers and local trainees. The compromise would be that the research agenda would be couched in broader rather than merely local terms. The results would be more portable and more generalisable. But there would be less to the Winchean idea that communities expressed themselves in their own terms. But perhaps giving that up is more of a sentimental than a substantial loss.