Alienated fathers' rights groups and the paternity testing industry are responsible for urban myths about paternity fraud, a university study has found.

Professor of Sociology at Melbourne's Swinburne University of Technology, Michael Gilding, said figures suggesting that up to 30 per cent of paternity tests showed that the nominated father was not the parent of the child in question were based on unreliable sources or studies.

He said the correct figure was closer to one per cent.

"What I found was that there were a couple of offhand statements which have been repeated so often they've been accepted as fact and when you actually look at the evidence, it's really weak," Professor Gilding said.

"In a way the evidence we have to work with is indirect but I've used every bit of public information I could find in the world for the last 30 years," he said.

"I thought for medical textbooks to be saying 10 per cent and for labs to be saying 20 per cent, there must be a pretty good basis. But there wasn't."

"On the whole in western societies I think it's somewhere between one per cent and three per cent and I think in Australia it's around one per cent."

Co-director of the Men's Rights Agency, Sue Price, said Professor Gilding's claims were "rubbish".

She said an American study of 44 blood bank testing clinics worldwide gave a figure of 28 per cent.

She gave further examples of a 1972 test in England, which she said found 30 per cent of husbands could not have been the fathers of their children and a 1997 test which revealed paternal discrepancies of up to 20 to 30 per cent.

"These are cited studies, I don't think you can just dismiss these," she said.

"I don't think there's any evidence whatsoever to prove that there is only three to four per cent from these figures. You have to look at these early tests that have been done."

A spokeswoman for Genetic Technologies Limited, a group which conducts DNA paternity tests in Australia, said their figures showed a 10 to 20 per cent exclusion rate of tested fathers.

However she said the sample of people they test is a skewed group.

"That means it's not within the general community, it's because there is suspicion or doubt over who might be the child's father," said Sue Lang of Genetic Technologies.

"So we're not talking general population, we're talking members of the population who already suspect that the parentage is in doubt."

Professor Gilding said in Australia, only 0.025 per cent of the population was tested for paternity each year and that public interest in it had grown since federal government frontbencher Tony Abbott was tested earlier this year.

"It's a much more exciting headline to say one in three kids are products of paternity fraud whereas me saying one in 100 is a pretty boring headline," he said.

Editors Comment: Take a look at some of the credible sources cited on this website such as the American Association of Blood Banks, University of Manchester U.K., University of California at Los Angeles, California, U.S.A., The Hospital for Sick Children in Toronto, Ontario, Canada and many more. click here