Key Points:

Two top health researchers have accused the Families Commission of "ideologically driven" bias in presenting domestic violence as a problem of men battering women.

Professor David Fergusson and Associate Professor Richie Poulton said their respective long-term studies of people born in Christchurch and Dunedin in the 1970s showed that most domestic violence was mutual.

"In a high proportion of these couples, we are seeing mutual fighting. It's brawling," said Professor Fergusson.

In contrast, the commission is backing White Ribbon Day on November 25, which asks men to wear a white ribbon to show that they do not condone "men's violence towards women".

The commission, chaired by former Race Relations Conciliator Rajen Prasad, was set up by the Labour Government in a deal with Peter Dunne's United Future party after the 2002 election. It has a budget of $8.2 million a year.

The private spat between the professors and the commission began after last year's White Ribbon Day, when commission chief executive Paul Curry said: "Almost all family violence is carried out by men on women and children."

The two professors wrote to the commission in March objecting to this claim.

Commission principal policy analyst Radha Balakrishnan said Mr Curry now accepted that he had made a mistake but stood by the claim that the worst domestic violence was perpetrated by men.

"We are talking about the most serious and lethal cases where perpetrators are predominantly men and the sufferers are predominantly women and children," she said.

"The gendered nature of intimate partner violence is really important."

But in an email to the Herald, Professor Fergusson said: "It is my frank view the commission's stance on domestic violence is not being guided by a dispassionate and balanced consideration of the evidence.

"Rather, it is being guided by an ideologically driven model that assumes on a priori grounds that domestic violence is a male problem and that female-initiated domestic violence does not exist or is so trivial that it can be ignored in the commission's policy focus."

The country's longest-running study of a birth cohort, covering 1037 people born in Dunedin in the year ending March 1973, found that 37 per cent of women and 22 per cent of men who had partners by the age of 21 had perpetrated acts of violence against their partners ranging from "pushing, grabbing or shoving" (29 per cent of women, 21 per cent of men) up to "beating up" (1 per cent of both men and women).

At age 21, 360 of the young people in the sample agreed to bring their partners to be interviewed too, providing what was said in 2001 to be the world's "largest study of abuse in a representative sample of couples to date".

The results showed that both partners abused each other in most couples where any abuse occurred.

Only 6 per cent of men committed abuse when both partners agreed that the woman did not commit any abuse, but 18 per cent of women committed abuse where the man did not. Male and female abusers shared "the same history of childhood conduct disorder and adolescent juvenile delinquency long predating their partner abuse".

The researchers concluded that women were not simply defending themselves against male attackers but that both sexes' violence stemmed from deep-rooted personality traits such as distrusting other people and being prone to anger, arising from a mix of genetics and upbringing.

They therefore recommended therapy for men and women, possibly including joint counselling for couples - an approach that is strongly opposed by anti-violence agencies.

These results were mirrored last year by Professor Fergusson's study of 1265 people born in Christchurch in 1977, of whom 1003 were re-interviewed at age 25.

Again, similar numbers of men and women reported violent acts against their partners - 6.7 per cent of men and 5.5 per cent of women said they had carried out minor assaults such as pushing or shoving, and 2.8 per cent of men and 3.2 per cent of women reported severe assaults such as punching, kicking or beating up their partners.

Ms Balakrishnan said both studies used a wide definition of "violence".

"Most people would consider it family violence where there is physical violence, where there is fear, where you are afraid for your safety," she said.

She pointed to a national Justice Ministry survey of 5300 households in 2001 which found that 21.2 per cent of women, but only 14.4 per cent of men, said they had ever had a partner who "used force or violence on you, such as deliberately hit, kicked, pushed, grabbed or shoved you, or deliberately hit you with something, in a way that could have hurt you".

Police statistics also show that men dominate the worst cases of family violence, including 31 out of 35 family homicides last year.

Professor Fergusson agreed that the homicide figures showed that the worst family violence was perpetrated by men. But that was such a small group that it did not show up in his sample of 1003 people.

He said the commission was "trying to have a bob each way" by saying that it was focusing on this tiny proportion of severe violence, yet also suggesting that domestic violence affected a fifth of the population.