[Update: in response to communication from Feminist Current’s Meghan Murphy, I am happy to clarify that the article critiqued below is not a “Feminist Current piece” but a Sam Berg piece which Feminist Current merely hosted.]

There’s been a bit of a social media buzz over this article on a radical feminist website, which claims that a recent Pro Sentret report from Norway – which you can read in English here – shows that “violence decreases under the Nordic model”. The author backs up her claims with an impressive array of graphs (and a fair smattering of ad hominems), and unsurprisingly receives glowing praise in her comments from people who were clearly predisposed to believe anything she said on the subject anyway.

I hate to burst their bubble. Well, actually I don’t.

The author kindly linked to one of my own posts on the report, though she seems not to have read it. If she had, she would have noticed that very near the start I referred to “methodological limitations” that made it unsafe to draw cause-and-effect conclusions from the study. At the time I didn’t feel it important to get into those limitations, but I will now. Apart from the health warning that always applies in studying a hidden population, there are two really massively important issues here:

1. Lifetime vs recent experience. The 2007-2008 study asked sex workers if they had ever experienced violence, throughout their “entire career in prostitution (which could be anything from one day to 50 years)”. The newer study asked about violence in the past three years alone. These are two very different questions, which can’t possibly give rise to comparable answers – at least without a detailed examination of the raw data, which we don’t have. We don’t know how long the respondents had been selling sex, in either survey; we don’t know how many in the first study had sold for more than three years nor how many in the second study had sold for less (though 16% of the latter group said they were not selling at the time the survey was carried out). You would expect, of course, that the more actual sex work-years covered by the survey, the more violence would be reported; and if we assume that the first study covered more actual sex work-years then we would expect to see higher rates of violence in it. I’m not even comfortable making that assumption on such flimsy data (which is why I didn’t make it in my initial post). But we certainly cannot make the implicit assumption that the Feminist Current post depends on, ie, that the two studies cover the same number of actual sex work-years.

2. Norwegian vs. foreign experience. Both surveys recorded sex workers’ experience of violence in prostitution wherever it occurred. For the 2012 study, we have a breakdown: 70% of respondents said it only happened in Norway; 12% said Norway and elsewhere; 10% said only elsewhere; 8% didn’t answer. There is also a breakdown of the venues within each country, but that is all. We don’t know, for example, which types of violence occurred in which country, or how many of the specific incidents occurred in which country. This makes it impossible to know how much of the reported violence even took place under the Nordic model. And we don’t have any of this data from the 2007-2008 study, so there’s really nothing for us to compare here at all.

Now, really, that ought to be enough to make it clear that Feminist Current’s claim is totally unsubstantiated. But just for the sake of argument, let’s say we really were comparing like with like. Would that justify their conclusions?

I’ll just address their headline statistic, namely, the claim that rape is down by half in the new study. That comes from here:

And indeed, the drop from 29% to 15% looks impressive. But wait a minute – look over in the left-hand column, halfway down. See that category of “threatened/forced into sex that was not agreed to”? Last I checked, that’s rape. And the number who said it had happened to them in the past three years was not 15%, but 27%.

The Feminist Current author didn’t miss that category of violence – in fact, she commented on it, but totally missed its significance. She also missed what the report’s authors had to say about it, which is as follows:

We have looked at how many checked both answers which could mean that they define both these categories the same way. Only 6 people have done this, which confirms our suspicion that many of the women would not characterize actual rape as rape. This also means that the actual frequency of rape is considerably higher than what is shown in table 10. If we combine the amount that checked these options and then subtract those that checked both we see that as many as 34%(25 people) of those that have experienced violence in the last three years have been raped/threatened into sex that was not agreed to.

So how does the 34% de facto rape rate compare with its 2007-2008 counterpart? To find that out, we’d need the same accounting exercise to be carried out on the earlier data. It could be that in 2007-2008 there was zero overlap between the persons who said they’d experienced rape, and the persons who said they’d been threatened or forced into having sex; this would give us a de facto rape rate of 64%, from which 2012’s 34% would still be an impressive drop. On the other hand, it could be that every person who said in the first study that they’d experienced “rape” also ticked the box for “threatened/forced etc”, which would mean the de facto rate in 2007/2008 was only 35%. In that case, the subsequent drop to 34% would be considerably less impressive, and probably statistically insignificant. The fact is, we simply don’t know.

Finally, since the Feminist Current argument rests entirely on the claim that “serious” violence is down in the 2012 study, I think there’s one other stat in that image worth highlighting:

This of course is subject to the same flaws as everything else in this study, and I’m not pointing it out to suggest that the number of sex workers threatened with a weapon actually has increased under the Nordic model, by 50%, from less than a quarter to approximately a third. I just think it’s kind of…curious that someone who takes the stats at face value, and accuses others of ignoring inconvenient data, doesn’t see any room for this in her analysis of how “serious” violence has changed under the law. But maybe she doesn’t consider “threatened with a weapon” to be serious enough.

There’s a lot more to criticise in that piece if I had more time, not least its contradictions with the radical feminist conception of sex work as inherently violent, inherently rape – and the way it almost mockingly dismisses certain forms of reported violence as not serious enough to be counted as violence for the purpose of this study, while then going on to insist that “any violence inflicted on them matters”. But I’ll let someone else unpack that one. As I said in my first post, we can’t safely draw any conclusion from the stats. The study’s significance lies in its qualitative findings – which are totally inconsistent with the idea of the law as a “success” and which are, unsurprisingly, totally ignored by Feminist Current.

I didn’t have to dig particularly deep into the study to find why the Feminist Current piece is wrong. Pro Sentret are careful to emphasise the lifetime/three years difference. They also highlight the fact that the number of sex workers who report being raped is much higher than the number who call it rape. There’s no reason why anyone who actually read the study wouldn’t be aware of these issues. If they choose not to share them with their audience, that’s a matter for them to explain.

(ETA: More on this subject here.)