On Sexed-Up Statistics

In yesterday's Guardian, Nick Davies, author of seemingly every British blogger's favourite book, Flat Earth News, delivered a pair of remarkable articles that confirmed him as one of the country's most important journalists.

In the first, Davies reported that a recent nationwide police initiative, Operation Pentameter, did not convict anyone of the crime of forcing women into prostitution after illegally trafficking them into the country.

This is rather surprising because, as he explains in a companion comment piece, forced sex trafficking has been widely reported as rife in Britain. The government has been telling Parliament and the nation that there are no less than 25,000 victims across the country. Anti-prostitution groups and charities agreed. Davies goes on to describe how this startling statistic was constructed through a process of exaggeration, misunderstanding, and plain invention.

In 1998, two academics identified a total of 71 trafficked women in the UK, and this did not refer specifically to forced or coerced trafficking. They suggested that the true figure could be anywhere between 142 and 1,420, but admitted that this was speculation, based on the assumption that for every confirmed case, there might be 2 to 20 in reality. A Christian charity quoted this as "an estimated 1,420 women", and others quoted them. The snowball had begun.

A second study estimated 4,000 victims of trafficking, but the researchers noted that this figure was "subject to a very large margin of error", "should be treated with great caution" and "should be regarded as an upper bound", as it was based on many assumptions. Heedless, another major charity quoted this as "4,000 trafficked women ... this figure is believed to be a massive underestimation of the problem". The government started repeating 4,000 as a fact.

Not to be outdone, a tabloid headline then reported no less than 25,000 sex slaves on the streets of Britain! Politicians started quoting this as a fact, although the newspaper provided no evidence for this figure at all. Asked why they believed it, a government minister said he used to work for the tabloid in question, and he trusted them to be accurate.

*

I have no idea how common forced sex trafficking is. I'd imagine it's not an easy thing to detect, let alone prove in court, so it could be going on behind closed doors and never make it into the statistics. It does happen, and obviously, every case is one too many.

But what certainly is true is that statistics have been greatly exaggerated, and then repeated, by the government and by various campaigning organizations. For more informed commentary on the issue by workers in the field, see Dr Petra Boynton's remarks here and the ongoing discussion here featuring Boynton and Belinda Brooks-Gordon.

Politician Dennis McShane MP "responded" to the criticisms of the 25,000 figure in an almost unwatchable TV interview and unconvincing article in which, amongst other things, he claims that 25,000 came from Amnesty International statistics. This is an outright lie. In fact, the tabloid did quote someone from Amnesty who commented on trafficking in general, but they didn't mention about numbers at all.

*

Attentive Neuroskeptic readers may well be experiencing a sense of déjà vu at this point. I have often written about the statistic - ubiquitous in Britain and elsewhere - that "1 in 4 people suffer mental illness". That number is made up, rather like the inflated statistics on forced sex trafficking.

Why are such statistics made up, and why are the made-up numbers usually shockingly high ones? It's no coincidence. This is what happens when the only people with an interest in talking about a statistic also have an interest in making it seem as high as possible. This is not to say that anyone deliberately fiddles the numbers, but rather, people naturally focus on the ones that suit them best.

In the case of mental illness, those who research mental illness know that their funding depends on the idea that it's a widespread problem. The more common people think it is, the more important studying it seems. Meanwhile, charities representing the interests of the mentally ill like high statistics because they make mental illness seem more "normal", thus destigmatizing it. It can't hurt their donation rates either.

With sex slavery, the inflated statistics were produced and repeated by organisations opposed to prostitution on moral grounds (including Christian charities and feminist groups), and by the government. The government's interest in the matter seems to be that they are currently trying to pass a law further restricting prostitution and the sex industry. The 25,000 supposed sex slaves must have helped convince Parliament about the importance of this move...

There must be many other examples of inflated statistics out there. It's inevitable, because in order to be taken seriously and to attract money, media attention and political support, campaigning organisations need to make their cause sound important. We can hardly blame charities for doing this, and as for politicians, we know not to trust them about anything. To expect an activist group or a political party to deal with evidence in a neutral and objective way is just naive.

What we'll always need, therefore, is people to scrutinize claims about social problems to keep the campaigners and the politicians honest. This is, or should be, the job of the media, but as Davies points out, the British media completely failed to do this for years. There will always be sexed-up statistics. What we need is more journalists like Davies to sex them back down again.

[BPSDB]