'All the Evidence Shows …': Reasonable expectations of educational research

After years of being criticised in Britain as irrelevant to the task of raising educational standards, educational researchers have been invited to join in the improvement of policy and practice. Welcome though the invitation is, some views of how they should deserve influence reflect unrealistic or narrow definitions of really useful research. This paper argues against exaggerating the prospects for a science of teaching; against tying the national research agenda tightly to what are currently identified as 'central issues'; and for a broader approach to relevance than direct applicability to improving educational practice.