Too many education researchers are producing partisan studies that are based on flawed methodology and biased samples.
They are also too ready to work "largely in a vacuum, unnoticed and unheeded by anyone else" and often pepper their work with unnecessary references to the theories of "great thinkers".
The criticisms are contained in a long-awaited assessment of academic journal articles that the Office for Standards in Education published on Wednesday.
The study is seen as less important than the report on research that is about to be published by the Department for Education and Employment. Nevertheless, researchers have been apprehensive about its publication since last September when it was announced that James Tooley, a right-wing academic, had been commissioned to carry out the evaluation by chief inspector Chris Woodhead, an ardent critic of education research.
The report that has emerged is in some respects more temperate than had been expected. It also includes a surprise - the revelation that it was co-written by a Manchester University researcher, Doug Darby, who describes himself as an Old Labourite.
The most strongly-worded criticism of researchers in the report is, in fact, to be found in Mr Woodhead's foreword: "Much that is published is, on this analysis, at best no more than an irrelevance and distraction."
James Tooley, the recently-appointed professor of education policy at Newcastle University, and Mr Darby are more guarded in their conclusions.
They hold back from endorsing the accusation made by Professor David Hargreaves of Cambridge University - that much education research is "second rate". But they find fault with 26 of the 41 articles they selected at random from four journals: The British Journal of Sociology of Education, the British Educational Research Journal, the British Journal of Educational Studies, and the Oxford Review of Education.
However, their criticism is hardly scathing: "Given the seriousness of many of these weaknesses, the tentative conclusion is that there are rather worrying tendencies in a majority of the articles surveyed.
"These conclusions may be disturbing, in particular in terms of the general health of the academic research community, and its potential influence in terms of the training and education of future teachers."
Tooley and Darby also acknowledge that all the articles they read were relevant to educational policy, practice or theory. Their report is, however, certain to generate controversy because it names the academics who failed their test of impartiality.
One researcher singled out for criticism, Andrew Sparkes of Exeter University, described the experience s of a lesbian PE teacher in the British Journal of Sociology of Education. "The paper sets out to show moments in Jessica's life to illustrate the oppression under which she lives," Tooley and Darby say. "However, if readers had expected many graphic details, they will be disappointed. Hardly anything is related which could possibly qualify as homophobia and oppression."
Educational research : an OFSTED critique is available free from the OFSTED Publications Centre, PO Box 6927, London E3 3NZ. Tel. 0171 510 0180