Mistaken claims that nearly one-third of seven-year-olds were unable to recognise three letters of the alphabet in the 1991 tests were not corrected by Kenneth Clarke, then Secretary of State for Education, and gained widespread currency, she said.
This led the public to think that something was terribly wrong with primary teaching, she told the conference.
The government also suppressed until July this year a Leeds University report which found that the results of the SATs (Standard Assessment Tasks) were unreliable, Ms Gipps, of London University's Institute of Education, said.
The Leeds report found that schools with a large number of ethnic minority children, children from deprived backgrounds or even younger rather than older pupils would not appear in a good light when test results were published.
This would have little to do with the quality of their education. Ms Gipps asked whether this incident demonstrated gross incompetence by Mr Clarke's advisers or political handling of unpalatable evidence.
Policy-making in education based on research evidence had been 'cut off at the knees', said Ms Gipps, expressing the frustration of an educational establishment which has seen its influence diminished by right-wing think tanks such as the Adam Smith Institute and the Centre for Policy Studies.
Policy was based on principles and gut reaction, not argument and evidence, she said. The decision to reduce GCSE coursework when it was proving successful in motivating pupils was a prime example.
'In the suppression of unwelcome research reports, the rubbishing of academics' arguments and the marginalising of unproductive pupils and schools, we see a further erosion of democracy and will see an increase of the underclass by virtue of the type of education system we are developing,' Ms Gipps told the conference.