Alan Smithers' concern that the "education-research industry" should produce material which is "useful" to policymakers and practitioners is commendable (TES, September 8). But he is surely misguided when he asserts that this should be the defining feature of the enterprise, rather than one of its constituent parts. In this regard, we might want to distinguish between "educational research" and "research in educational settings". The former is geared to informing explicitly and directly policy and practice; the latter, to the development and testing of theories, concepts and ideas which are neither necessarily defined nor confined by such matters.
On this understanding, the efforts of the "education-research industry" are more in tune with the demands of the research assessment exercise.
DR BARRY TROYNA Reader in Education University of Warwick Coventry