In your articles on evidence-based policy ("I can't hear you ..." and "From the lab to the lobby", 26 March), there was a strong sense of frustration with politicians for failing to give sufficient weight to research findings in political decision-making.

A fundamental problem with this perspective relates to the underpinning model of the policymaking process, commonly conceived as an abstract, rational, technical exercise in which difficult value-laden choices can be resolved by objective analysis. If we recognise policymaking as a practical "craft" in which judgments are based on many sources of knowledge and many legitimate voices, we obtain a more realistic perspective on the contribution of scientific evidence and a healthy scepticism about the concept of "evidence-based" policymaking.

Politicians must listen to many voices and balance a range of interests. Rarely will decisions be reducible solely to evidence - and rarely is this appropriate. What is the basis for the privileged influence of scientific evidence as one form of knowledge among many? And how definitive is our scientific evidence in providing specific policy recommendations? How often do our reports provide useful pointers for policymakers? How often can academics step up to the plate when ministers ask them: "What should we do?"

However, there are indeed many examples where strong political commitments override evidence, especially when the "political heat" is raised - but this is not "irrational", it is politics in a democratic system. And there is a problem with the manifesto commitments of incoming governments. But to generalise without investigating the whole picture risks committing the very crime that is being condemned, namely failing to respect the evidence.

ÖйúAƬ

In fact, much policymaking is informed by academic evidence, but this happens through "routine" processes of government, so it doesn't make the headlines. We understand that research often influences the perception and definition of policy problems and thinking about policy responses over time - look at the climate change debate. There are many examples of policy frameworks and strategies being influenced by syntheses of research and evaluation studies, and of the implementation of policy initiatives being informed by evaluation studies.

It is ironic that much fashionable polemic on evidence-based policymaking is not evidence-informed. Good research has been undertaken and presents a complex picture of a range of ways that evidence impacts upon policy, dependent upon the political context and circumstances. But this is not the stuff of headlines. So, let's be careful that we don't commit the same sins that we condemn in our politicians - of selecting evidence to fit our prejudices and interests.

ÖйúAƬ

The answer is not to withdraw from the game but to raise our game - recognise that there are many legitimate players, that the evidence will not "speak for itself", and act more forcefully as "advocates" for the evidence. Until we are confident that we have given full voice to our interests, let's be cautious about condemning our politicians for not listening.

Ian Sanderson, Leeds Metropolitan University.