Somehow, almost every educational fad dreamed up in the US eventually makes its way to the UK. Back in January 2001, newly elected president George W Bush announced that education would be taken more seriously than ever before. Science would provide the answers that had eluded public education throughout the previous century.

Accordingly, an Institute for Education Sciences was set up to promote scientific rigour in education research. Rigour became associated with studies conducted through randomised controlled trials (RCTs). “Once we have dozens or hundreds of randomised or carefully matched experiments going on each year on all aspects of educational practice, we will begin to make steady, irreversible progress,” said Robert Slavin, a leading advocate of evidence-based education.

Much of the rest of the world - including the UK - is now catching up with these American pioneers. “Randomised controlled trials will raise the bar in education and children’s services by increasing the use of quantitative evidence,” said England’s Department for Education in May.

Based on a report commissioned by education secretary Michael Gove and authored by science campaigner Ben Goldacre, the department is calling for a new age of RCTs. According to the report’s dogma of scientism, the future of schooling is contingent on the outcome of trials, rather than what teachers do in the classroom.

But despite its rhetoric of science, evidence-led policy is a dogma, and there is little evidence that it has actually made much of a difference to social policy or education. In the face of this failure, the “more research is needed” mantra is a constant refrain.

Studies of education trials indicate that the complexity of the variables influencing teaching and learning render RCTs pointless. Yet despite the dismal record of evidence-based policy, governments desperate for legitimisation have embraced the idea with reckless enthusiasm.

A medical model of research

The use of RCTs is often justified through an analogy with health: RCTs are an invaluable resource for testing the efficacy of different forms of medical intervention and are rightly valued by health practitioners. But clinical trials are unlikely to be effective in a non-clinical environment. Effective methods of research are context-specific, and emerge through interactions with the problems they attempt to confront.

There is a danger that the introduction of a medical model of research to education will encourage its medicalisation. The language of medicine is often used by advocates of RCTs, who justify their methods on the grounds that they will lead to the “interventions that work best”. However, a school is not a hospital; educators, unlike doctors, are there to teach, not to intervene. The integrity of a classroom relationship is compromised if it is perceived through the prism of an intervention.

The status of a student should not be confused with that of a patient, either - an intervention assumes that there is a deficit or illness that needs fixing. Regrettably, this assumption is frequently expressed by policymakers. In education, medicalisation often meshes with the pathologisation of the classroom.

An example of this trend is provided by British thinktank the Centre for Social Justice: its recently published report warns of an army of children starting school still wearing nappies and behaving like toddlers. Clearly, this diagnosis of classroom pathology represents an invitation for an intervention.

Advocates of RCTs and interventions hold the conviction that the principal question facing education is “what works?” So the UK’s Alliance for Useful Evidence, in its recent report Evidence for the Frontline, advocates its cause on the grounds that it will help teachers to know “what works”. Regrettably, its obsessive pursuit of what works subordinates the question of what children need to know to the dictates of narrow pragmatism. The imperative of “what works” invariably leads to the transformation of the curriculum in line with its instrumental ethos. So, if imparting a particular subject does not work easily, it can be pragmatically discarded.

Paradoxically, the “what works” model does not even work on its own terms. Despite all the resources invested by the Bush administration, RCTs have made little difference to the quality of US public education. Why? Because in education, RCTs produce evidence that has little relevance to distinct and continually changing contexts. Only a handful of initiatives have been identified in which the evidence base met its own criterion of rigorous research.

Advocates of the “what works” culture assume pedagogy is akin to a technology that can provide techniques yielding predictable outcomes, but education never has an entirely predictable outcome. Experience shows that so-called best practice cannot simply be reproduced in different environments - it invariably leads to a formulaic box-ticking culture that distracts teachers from developing their capacity for professional judgement.

Genuine education research raises as many questions as it answers. Such studies integrate the social and cultural influences that shape the dynamic of the classroom, and force teachers to pause and reflect on their own experience. What teachers need is not more evidence, but an intellectual space where they can reflect and discuss their experience and gain the insights that can guide them to experiment and learn.

Frank Furedi is professor of sociology at the University of Kent, England

What else?

Frank Furedi’s Authority: a sociological history, published by Cambridge University Press, is out now.

To hear his address to the ResearchED 2013 conference, go to www.tesconnect.comfuredi