Women are more aggressive partners than men.....

I'd be prepared to argue that (on average) men are quicker to forget times they've behaved aggressively, as they might be more inclined to 'dismiss' the effect their behaviour had on others. And women more likely to 'dwell on it' long after the event when they have been aggressive, because they are more aware of it being 'a problem' and more aware of how others (including their partner) might perceive their 'unfeminine' behaviour.

But that's obviously just guesswork on my part, and projecting how I respond to things onto the subjects. I don't often lose my temper, and never get physical, but when I do it sticks in my memory far longer than it probably deserves.

Sorry the bolding went astray.

Participants were all students recruited via e?mail and
undergraduate lectures - so we have a fairly unrepresentative sample of men and women for a start. They are younger, more educated, more 'motivated' in terms of education and career. Are the men typical of all men, or are they less likely to be aggressive and controlling because they don't think women are making them feel inferior? Are the women more likely to be 'competitive' and 'driven' (sorry for the choice of words, I can't think how better to put it).

all participants were required
to be in a romantic relationship, or have been in a
romantic relationship, of at least 1 month’s duration - That's very early days for relationships, is it typical of most relationships? Are student relationships typical anyway?

And what about the 'stages' that relationships go through? They are unlikely to be living together, very unlikely to have children, unlikely to have one partner (generally the woman) no longer in paid employment and financially vulnerable. Unlikely to have money or job worries in general - or at least not the kind that involve car loans and mortgages.

The 'controlling behaviour' indiicators sound like a lot of things that immature people are likely to do in their earliest relationships. “Want to know where the other went
and who they spoke to when not together,” “Use nasty
looks and gestures to make the other one feel bad or
silly,” “Try and restrict time one spent with family or
friends” and “act suspicious and jealous of the other
one.” People tend to grow out of those!

There are also things like self-reporting - are they actually being 'fair' to themselves? Many people agree that they are 'as bad as each other' in a relationship, when an outsider would actually say differently.

Good points about age, length of relationship and children, Buffy - because (as far as I can tell, reading stuff at second hand) violence often begins or escalates dramatically when the woman in the relationship becomes pregnant for the first time.

Bloomin - any comments on my questions re. methodology - i.e. selection methods for sample, possible confounding factors (class, educational level), whether the d-values are indicative of any significant difference between populations?

(NB I am not offering these as "flaws" because I'm not a social scientist so not qualified to comment, but I am very interested to hear your thoughts on it).

BN, I didn't comment on the validity of her research in either of my posts about lectureships, and made it clear I wasn't trying to in them.

I don't know anything about psychology and don't really know anything about inter personal violence, either from an academic point of view or a general activism point of view. I'm not capable of commenting on her work. I can only look at her abstract and see if she was mis-reported. I don't think she has been.

I still think it is exceptional for anyone to be given a lectureship based on no publications, and a lectureship is usually about what you then are going to research; teaching is a side issue.

If you don't want to discuss academic careers, don't! Respond to the posters posting about the content of the study.

Okay, here's my quick plot for a pair of generic normalised gaussians. Apologies for actually showing you some diagrams and talking maths here, but I think this is a case where a picture is worth a thousand words.

Both assume a d-value (crudely, a measure of how different the means of two populations are given the spread within each population) of 0.2.

I've looked at two possible pairs of distributions, one where the two sigmas are the same (1.0) and the other where the sigmas are different (1.0 and 5.0), assuming a mean of 0 (entirely arbitrary - this bit doesn't matter - just slide sideways to whatever value you're actually interested in!) for one distribution, then calculating the other mean according to the formula for the d value (which is our starting assumption, because that's the value we are given in the paper).

The plot on the left shows both sigmas equal to 1. I'd argue that, to use a technical phrase, to a close approximation, there's sod all difference between these two distributions.

The plot on the right shows a bigger difference between means - but ties up nicely with Dadwashere's suggestion that we may be looking at a situation where there could be more low-level violence in one population, but the other population (the broader one) shows much more violence out at the extreme end of the spectrum.

In other words, I think the d-values in this study are so low that we're not looking at a difference that is of any interest, even if other groups were able to replicate it.

(NB, would concur with the comment that given how cut throat academia is, it is interesting to be given a lectureship with no publications behind you).

Dad, I think other things like age come into it. I would have felt very differently in that situation if I was 16 and the other person in their twenties than I would have done at 25.

I also think what is and is not abuse is not a series of broad simplistic statements.

'hitting with an object' is abuse.
Hitting your partner with a baseball bat on purpose so that require medical attention and not done in self defence - abuse.
Hitting your partner by throwing a piece of jam on toast at their knee in the middle of a mutual heated argument in a one off, not repeated act of childishness - not abuse?

'Asking who they have been with' is controlling behaviour.

'Have you been out with Janice again? that bitch. And she is just using you. She's not really your friend. Stop seeing her.' - controlling.

Have you been out with Janice lately? Is her mum feeling any better. - probably not controlling.

Women being more aggressive than men sounds bollocks.

Are we allowed to talk about this or do we have to compare the size of our phds before we do that?

Hrmmm. I don’t doubt that women could very easily be more aggressive in terms of quantity but certainly not in terms of the commonness and extent of severity. I doubt the study takes into account perceived threat. I mean, technically, I was sexually assaulted by a teenage girl I barely knew back in my 20's, woken up finding her straddling me, grinding her crotch against mine as she worked on getting my pants off. It was a huge WTF experience and I pushed her off but I felt no threat, since I was stronger and she was not armed with anything except raging lust for my cock. I did not feel assaulted or damaged by the experience so while it would show in a scientific study the damage done to me was nothing really, not comparable to what a woman would probably feel in a gender reversed situation.

Sorry, 'there aren't groups of women.'

BloominNora, I am not criticising her level of experience in terms of the validity of this research. There is plenty of good research published from work done by research students. I have no issue with that at all.

What I have issue with is this - if anybody said to me that they were going to get a degree, then a second degree, then a PhD and then were going to immediately get a job as a lecturer, I would say maybe that is possible if you have had numerous research papers published and/or you have previously had a succesful career in a field related to your research area.

I would say the more likely outcome for somebody with a PhD, no previous career and not a single publication is that they might get a post doc research position if they were both good and lucky, on a temporary contract, and then they might get another such contract and eventually they might become a lecturer once they established a publication record.

If somebody was in that position and they were researching violence against women, I would think their chances of even getting a post doc were reduced.

So somebody like Finn Mackay, an award winning activist with multiple publications and work behind her has gone from PhD student (research area - violence against women) to research fellow. And maybe that was her choice, because she wanted to prioritise activism (I don't know). But very many highly qualified people with PhDs do not get given lectureships and careers in academia. There are groups of females with doctorates sitting around going, 'what shall I do with all my knowledge of violence against women? I've not published anything but I can't decide if I should take up a job as a lecturer at York or Manchester Met.' There aren't people sitting around in an any academic research area with those options.

So when I read that somebody with far less experience and no publications has been given a lectureship in the research area of inter personal violence, they must have that job based on the value given to their particular analysis, presumably over other more experienced candidates with other research interests. I think it is worth considering what their particular field of expertise is and what particular analysis of violence their research is advocating to make them the best person to have that job. And it would seem that the analysis valued is the alleged problem of female's greater violence towards men.

And that speaks volumes about academia as a system, not about Dr Bates as a person or an academic.

Okay, quick skim later, I have a some questions about methodology (I am not a social scientist by the way, so would be interested in the thoughts of social scientists on this). The participants were selected through responses to an e-mail request to undergraduates at a university. This suggests to me that:

1. we need to think about educational level as a confounding factor;
2. we need to think about social class as a confounding factor;
3. the participants were in a sense "self-selecting" - they chose to respond (or not) so may not be representative of the population at large (even the population of university students).

How much of an impact are these things likely to have on the results?

On the plus side, the issue of d values and t values for the results seems to be quite carefully calculated.

Another question for the social scientists out there - is a d value of 0.2 (or thereabouts) indicative of a difference between populations that actually matters? I just tried out a test case, with mu1=0.0, mu2=0.2 and sigma for both distributions of 1.0 (assumption!) and the results are two gaussians which are only fractionally displaced relative to one another, so not something I'd be inclined to think of as a particular interesting difference between a couple of populations (eyeballing it, I'd say that if one looks at the mean for distribution 1, about 55% or thereabouts of the population for distribution 2 is greater than the mean for 1 - in other words, there ain't much difference... but then, I'm a physical scientist, so I stand ready to be corrected on this one).

I think there is far better and more reliable data on this subject than Dr Bates's study.

The facts and crime statistics speak for themselves.

I cannot find a copy of the original research paper to comment on.

There have been a lot of discussions on here recently about the difficulty of women advancing careers in academia, getting academic jobs, getting grants to study women's issues or being able to study women's issues at all.

Without wishing to diminish her achievements, Dr Bates primary research interest (according to her) is female interpersonal violence against men. She has moved straight from PhD to a lectureship, with no publications at the time she was given a lectureship in 2011 and no prior career in a related field to explain her success.

That is a rare, exceptional and incredible situation.