One of the things I do is look at violence in health settings. To do this I have to define terms, such as threats. There is a whole discussion about thresholds of seriousness. Take for a second this statistic.

The Otago Women’s Health Survey results released in 1991 found one in three girls before the age of sixteen are likely to be sexually abused. Also, one if four women are likely to experience some form of unwanted sexual abuse as adults (ie over 16 years) (Mullen et al., 1991).

From the Otago Study women who reported child sexual abuse in a postal questionnaire were invited to an interview, along with an equal number who had reported no abuse. This study found that 32 per cent of the women had experienced some form of sexual abuse such as genital contact, attempted intercourse or actual intercourse before the age of 16. Of these, 16% were under 12 years old, with the greatest age risk being eight to twelve years old (Anderson et al 1993).

I know this literature. I’m aware of the study. Now… the astract reads differently.

Nearly one woman in three reported having one or more unwanted sexual experiences before age 16 years. A significant number of these experiences (70%) involved genital contact or more severe abuse, and 12% of those abused were subjected to sexual intercourse. The abusers were usually known to the victim, being family members in 38.3% of cases and acquaintances in another 46.3% Stranger abuse accounted for 15% of all abuse experiences. Most of the abusers were young men, disclosure of the abuse was infrequent, and only 7% of all abuse was ever officially reported. Prevalence rates showed no urban/rural differences, no cohort effect with subject age, and no age differences in disclosure rates.

Forced child intercourse at a young age by an older man is quite different from genital touching of a teenager by her boyfriend (ie. “petting”). The authors of the original paper correctly call these unwanted sexual experiences. Now that is not abuse. Some of it is, but not all of it.  But the “one in three are harmed” meme is out there telling men that they are evil.

There is a continuum here from unwanted acts to actual rape. Lumping everything together overstates the rate and leads to a minimisation of the impact on those women who have been really traumatised. (Bad events are not trauma. Trauma is when you are serverely injured, or think you might be severely injured).

Now, this is in today’s Herald…

Almost half (46 per cent) said current or previous partners had insulted them or made them feel bad about themselves, 30 per cent had been “belittled or humiliated in front of other people”, 26 per cent said partners had “done things to scare or intimidate [them] on purpose”, and 19 per cent said partners had threatened to hurt them or someone they cared about.

Asked about controlling behaviours, 18 per cent said their current or most recent partner “insists on knowing where you are at all times”, 11 per cent said he “ignores or treats you indifferently”, 10 per cent said he “gets angry if you speak to another man”, 9 per cent tried to stop her seeing friends and 5 per cent tried to restrict her family contact.

Dr Fanslow said some of these behaviours, such as “saying something that makes you feel bad about yourself”, might not be intentionally abusive. But they were still regrettable.

“I don’t think being insulted or made to feel bad is a good thing to do.

“It does say to me that these are things we need to be looking at quite seriously in our relationships,” she said.

Relationship Services’ national practice manager, Jo-Ann Vivian, said “put-downs” from partners often damaged people’s self-esteem and had long-term impacts on both emotional and physical health.

“Contempt is one of the hallmarks of a relationship which is in very deep trouble and likely to end,” Ms Vivian said.

She called for more education in schools about how to communicate with “healthy, respectful interaction” in relationships.

via ‘Half of women’ tell of emotional violence – National – NZ Herald News.

OK. lets find the paper and look at the methodology. The abstract is here, but I can find the paper

The survey was conducted by the School of Population Health at the University of Auckland, as part of the New Zealand Violence Against Women Study (a replication of the World Health Organization [WHO] Multicountry Study;  A random sample of women aged 18 to 64 years was obtained from (a) Auckland, the largest city (population 1.2 million), containing 26.8% of the total New Zealand population of women aged 15 to 64 and (b) north
Waikato, a rural region, containing 2.8% of the total population of women aged 15 to 64. Potential participants were contacted by door knocking at randomly selected households, following a population-based cluster sampling scheme. The starting point for each cluster consisted of a randomly selected street and dwelling number. From the starting point, interviewers in Auckland approached every 4th house and interviewers in Waikato
approached every 2nd house, until 10 households per cluster had been reached. Nonresidential (e.g., commercial, retail, or industrial buildings) and short-term residential properties (e.g., hotels, motels, boarding houses) were excluded

The method of surveying by household is difficult and well covered by the use of random cluster analysis (Kish wrote this up. In the 1960s). This could have been done better, but is OK.

The definitions they used came from the WHO. They are:

Physical violence was defined as having been slapped; had something thrown at them that could hurt them; having been pushed or shoved; had their hair pulled; having been hit with a fist or something else; having been kicked, dragged, or beaten up; having been choked or burnt on purpose; or having been threatened with a gun, knife, or other weapon or had one of these weapons used against them.
Sexual violence was defined as having experienced one or more of the following acts: being physically forced to have sexual intercourse when the woman did not want to, having sexual intercourse because she was afraid of what her partner might do, or being forced to do something sexual that she found degrading or humiliating.

OK, to the results. And here the bias in reporting comes out. The rate of violence is not 1:2 apart from Maori. It is more like 1:3 to 1:10.

Lifetime prevalence of physical and/or sexual IPV among M?ori women (57.6%, more than 1 in 2) was significantly higher than that of Pacific women (32.4%, 1 in 3) and European/Other women (34.3%, 1 in 3). Asian women reported significantly lower lifetime prevalence of IPV (11.5%, 1 in 10) compared with women of European or Other ethnic origin.

In the discussion it is fairly clear that this methodology gets similar results regardless of where you go.

The 58% lifetime prevalence for M?ori women is at the high end of the prevalence estimates found in the WHO Multicountry Study (Ethiopia: 71%; Bangladesh province: 61.7%; Tanzania province: 55.9%). The United States, Canada, and Australia have also reported a higher prevalence of IPV among indigenous populations compared with European groups

There is something about the agenda of these questions and the use of a sampling strategy that includes only women and (if you read the paper) assumes that the women’s attitude is always the correct one that is fundamentally flawed. We know from two child development surveys that the rate of violence  within relationships is roughly equal: men and women both get hit, assaulted, quilted and abused. If you only count women, you continue the women — vicitim, man — perpetrator (or the evil penis meme). which is clearly wrong.

Both men and women can be violent. Both men and women get hit. And the definitions used by the WHO are too broad. There needs to be a differentiation by type of aggression.

But it is better not to consider issues such as ethnicity (and poverty) and instead say that it is men;s fault.