A leading education academic has warned against the growing trend of using evidence-based research to decide which classroom practice should or should not be sanctioned.
Martyn Hammersley, professor of education and social policy research at the Open University, claimed that relying on research to inform education policies would result in valuable teacher expertise being sidelined.
He said while there is a need for high-quality research, the type of extensive trials carried out in medicine - the model that some have called to be introduced in education - are expensive and not necessarily suitable in education.
Some experts have called on the education world to replicate healthcare's National Institute for Health and Clinical Excellence, a state-funded organisation that interrogates the efficacy of medical treatments and decides whether they the NHS should fund them.
Last month, the Commons science and technology committee said the decision to roll out Every Child a Reader was made without considering other interventions.
Its report concluded that a controlled trial of Reading Recovery, the programme at the heart of Every Child A Reader, was feasible and necessary. It said trials should be put in place to identify the most effective and cost-effective alternatives.
But Professor Hammersley said there are dangers in trying to ape medical research: "Policy-makers are ambitious to improve education, but have to accept the nature of life that it doesn't fit (healthcare's) model and they have to adjust policymaking to deal with that.
"But because one of the problems of the anti-professionalism that started around Thatcher's time is that there is a distrust of people who work within the relevant sector; it is assumed they have their own interests at heart.
"The notion of evidence-based practice feeds that prejudice - that you can't trust what practitioners say, that you have to look at the research," Professor Hammersley went on.
"You do, but you also need to talk to people with relevant experience. Randomised controlled trials are an excellent method for particular purposes, but there is no single research method for all purposes and no method is perfect."
The literacy hour, in which primary teachers were told in minute detail what to teach and how to teach it, has come to be seen as the epitome of centralised diktat. The Government has now moved away from this approach and looks set to increase autonomy further with the abolition of the national strategies and a new primary curriculum due in 2011.
Professor Hammersley said research could have a valuable role as one of a number of factors to be taken into account when new policies are developed - but the notion that it was possible to find a single policy or practice that is clearly best overall is unrealistic.