The analysis has thus far been confined to exploring the functions of knowledge in immigration policy. The focus on a single area of policy has allowed us to compare patterns of use between different types of organization, operating in rather different political contexts. Nonetheless, this has inevitably implied a relative neglect of how these patterns might vary according to the type of policy area involved. Might other types of policy be associated with more interest in the instrumental, as opposed to the substantiating and legitimizing, functions of research? If so, what sorts of feature might shape knowledge use of this kind? Can we make any general claims about correlations between features of policy areas and the deployment of knowledge?

In this final chapter, I shall attempt to extend my theory of the functions of knowledge to see what it can reveal about the uses of research in different areas of policy. As we shall see, there are a number of policy domains that are likely to be more dependent on instrumental knowledge than was the case with immigration policy. However, the discussion of these dependencies raises further questions about the notion of instrumental knowledge. I shall draw on recent literature on the ‘co-production’ of science and policy to question whether it makes sense to talk about research as instrumental to policy, even in highly technical areas that rely extensively on research.