Professor Murphy, the new president of the British Educational Research Association, the co-organisers of the conference, said education policy-makers and politicians had been frustrated in the past by the independence of researchers and their relative freedom to look at what they wanted and to draw whatever conclusions the data suggested.
The past 15 years had seen more commissioned research and calls by such key figures as Margaret Thatcher and Sheila Browne, the senior chief HMI, for the research agenda to be more closely controlled.
Professor Murphy said there is a place for commissioned research that is policy directed, but, as others had warned, there are inherent dangers if contracts become more and more prescribed. Educational researchers risk being used to support educational dogma rather than putting it to the test.
The British Educational Research Association is continuing to address the ethical issues which face researchers, he said. The association "needs to strive to monitor the arrangements relating to contracted research". Researchers needing to stay in work can easily become ensnared in double-bind situations with funding agencies which expect to pay the piper and call the tune.
"All our best research should, in a sense, be subversive," Professor Murphy said. It should have the ability to challenge the status quo, where evidence was assembled to justify that. "It should, if it is any good, at times shock people, surprise people, get people out of their seats to debate educational issues, consider change, explore and experiment with new ways of doing things." Education, he said, needs research and stands to benefit enormously from it.
Research, he said, should also raise the quality of public debate. Writing sensational headlines based on the annual GCSE results had almost become a national sport, and it appeared to matter little whether the results were better or worse than those of the previous year.
The professor said he had looked at the demographical changes among GCSE candidates over the past eight years, a period when the birth rate plummeted to an all-time low and the proportion of candidates from middle-class families increased.
He said: "It is crazy to draw conclusions from national changes in GCSE results without attending to such issues, when we know that social class is a potent indicator of educational success. This does not tell the whole story about what has caused the changes to occur and I am not suggesting that it does. But it is a pretty important feature that needs to be brought into the debate."