Tuesday, 31 August 2010

Guest Post: Setting the Record and Sex Trafficking - A Critique by Dr Belinda Brooks-Gordon

I am delighted and privileged to host a guest post by Dr Belinda Brooks-Gordon, Reader in Psychology and Social Policy at Birkbeck College, the University of London.

Belinda is a leading academic authority on sex work policy and the author of The Price of Sex: Prostitution, Policy and Society. She was a Liberal Democrat candidate at the last General Election, and she is also a Patron of Westminster Skeptics.

Setting the Record is an attempt to work how many trafficked migrant sex workers there are in England and Wales, and where they come from.

Police visited 142 premises meeting 254 sex workers in the process. These comprised 210 migrant women and 44 British Nationals. The data show that 24 sex working women were ‘trafficked migrants’, 113 were ‘vulnerable migrants’ and 73 women were not either the trafficked or vulnerable.

Nobody had been kidnapped or imprisoned and the research identified few individuals subject to violence.

There is therefore a drop since 1998 when 71 trafficked women were known to the police. The hard data shows little has changed despite hysteria over trafficking numbers. The report aims to be ‘more nuanced’ with the new ‘vulnerable’ category and there is laudable recognition of the diverse reasons for choosing sex work and volition of migrants. It states: ‘for some, selling sex is the result of a conscious, rational and independent assessment of the available options, and working in the UK is seen as a relatively safe and profitable place’ (p32).

The rest of the report however has data collection methods and figures so weird and wonderful they could be at the end of the pier.

It is an example of the perils of lack of peer-review, and contains the most nonsensical multipliers and whacky extrapolations I have read since the unintentionally hilarious attempt produced by the Home Office.

There is no critical review of previous research to expose flaws in previous research and avoid making the same mistakes.

Unfortunately flawed assumptions are repeated, for example the discredited assertion that 56% of trafficked women suffer PTSD.

The authors attempt to work out the number of premises and sex workers from newspaper, internet and telephone kiosk adverts, presumably copying flotsam containing the same daft data collection, which has been extensively criticized.

Past research to estimate the sex worker population as a Fermi problem is looked to for ‘corroboration’ only after the worst is done.

From the adverts they tried, unsuccessfully, to remove duplicates. The number of businesses identified was multiplied by the number of beds and multiplied again because of shift patterns. It might have been more logical, given that women may work at two different establishments on different days, to divide rather than multiplying here.

Attempts to get data from ten regions failed, leaving only six so essential data on nearly half the country is missing. There was no data from the North West or North East region. There is distorted data from the South East and Eastern region where duplicates could not be removed. Because of the variety of systems in law enforcement areas across England and Wales different data collection methods occurred in each region and it was wholly unsystematic.

The methods lack reliability, as the police were allowed flexibility to make a judgement on their observations as to the likely presence of vulnerability indicators, this introduces bias and there is no inter-rater reliability on any of the measures.

Representativeness of the sample is claimed but the variables or demographic factors on which the women are claimed to be representative of others in the sex industry are not given.

There are no concrete examples of any of the vague categories or indicators given.

Dangerous unsupported assumptions about the uniformity of coercion levels are made in the absence of any data: ‘there is insufficient volume of data to determine whether regional differences exist in coercion methods. It has therefore been assumed that coercion methods and levels are uniform across the country’ (p25).

The statistical test Pearson’s Product Moment Coefficient was used to correlate the number of premises and police officer numbers and the arbitrary removal of inconvenient outliers and all the London data would ring alarm bells in the editorial office of most academic journals.

Despite claims of objectivity there is bias that: ‘this report examines organised, and thus illegal, forms of prostitution’ (p10). However, recent acquittal shows that organized sex work is not inherently criminal.

The ethics of Setting the Record as a piece of research have to be questioned, as well as the storage and use of sex workers’ personal details, coming as it does just two weeks after the sex workers were ‘named and shamed’ by the Metropolitan Police website publishing sex workers’ personal data.

Recent reports, like the one from the World AIDS Foundation, suggest from a pragmatic harm reduction stance, that violence against women can only be addressed in a fully decriminalized system.

Setting the Record is not as woefully inadequate as previous ACPO-related reports such as Brain et al, critiqued by Helen Self but it falls far short of basic research stardards.


No purely anonymous comments will be published; always use a name for ease of reference by other commenters.


@mazzawoo said...

I had to read this report for work. I found its elaborate explanations for the data confusing and I was also dismayed at the data gaps; I don't see why they couldn't have collected that data in the other regions.

What perturbed me most was that the researchers categorized the women interviewed but didn't publish any of the anecdotal evidence. Working in the violence against women field, we need to know what women need in order to develop our services and this was a missed opportunity to hear their stories.

Retired said...

Well way back in 1972 the predominantly West Indian pimps of Ipswich had a saying "Capture".

This was showing 15 year old girls in residential care a bit of kindness (involving older prostitutes) and recruiting girls from care into prostitution. "Capture" had no force.

Meanwhile a local chicken gutting and packing factory was offering a cash payment to dole clerks who could coerce unemployed women to work for them. Free transport to and from work etc. All a dole clerk had to do to earn the payment was suspend benefits in an attempt to force unemployed women into work at the chicken packing station.

Well the chicken packing factory always had staff shortages but there was never a shortage of girls entering prostitution to top up their benefits.

Panda car patrolmen were on a fiver a week from the pimps. CID on a lot more.

Pimps recorded the index numbers of cars picking up their girls.

And there was no serial murdering.

And so postwar madness begat social workers and their ilk. With the enthusiasm of the zealot seeking self justification they set about "Saving" the prostitutes.

"Come gals there is a better life bidding you. The chicken gutting line has vacancies."

And verily did the girls question what use are these do gooders.

"Help you get benefits if you tell them what they wanna hear"

And thus was born sociology statistics.

Matt Greenall said...

Thanks for this useful analysis. One of the challenges (which I am sure you are aware of!) with critiquing reports like this one, is to be accused of denying trafficking exists or denying harm within the sex industry. These criticisms are not on the whole valid; nonetheless I think it would be useful to put forward a new framework for research on sex work, migration and trafficking that avoids many of the ambiguity traps that research like "Setting the record" fall into. Are you aware of any work in this direction? Just for information, I work on HIV and sexual health programmes in low-income countries, where we also struggle with getting good quality, ethically produced data in particular in relation to sex work. These are also contexts where law reform in relation to sex work is increasingly a hot topic - often driven by HIV programmes which in some countries are the only "international" programmes coming into systematic contact with sex workers.

Belinda BG said...

Yes the ‘data’ does seem so divorced from any real human being doesn’t it?

One of the problems with the lack of regional data is that the report should have been more explicit about the representativeness of the regions and not tried to extrapolate so widely. It is also strange that the regions for which there was no data, generally showed high incidents of vulnerability without any reason.

The authors should have focused on the data they did have – ie 24 people, and explored those cases to explore how transferable the data was. There is a difference between ‘transferability’ of findings (ie to a similar population) and ‘generalisability’ (to a wider population) - which is what they tried to do but not sustainable with the data they had or the statistical tests they carried out. I mean why did they correlate the number of businesses with the number of police officers? Just what precisely was the research question?

I have to admit that I would not want to read anecdotal evidence, but if you mean qualitative data, grounded in accepted theory, collected in an ethical way, with full statements as to how representative the selected data to the overall dataset, and how it was collected, that would have been really valuable.

Unfortunately so often ‘anecdotal’ is confused with ‘qualitative’ and so many of the reports on sex work and trafficking are just anecdote masquerading as qualitative research. The latter is more time consuming, expensive and requires more skill to do but there are well-respected, systematic ways of collecting and evaluating qualitative evidence. We just we seldom see them used in research on trafficking. This hampers the development of good policy.

@ Retired
I tend to think of Lazarsfeld being the progenitor of empirical sociology

What you say about Ipswich is interesting – and raises the point about the historic and enduring nature of ‘beats’ or red light areas.

Kimpatsu said...

The lie that Asian men in Ipswich groomed underage girls for prostitution has long since been debunked. Your sources are a bit suspect, at least.
The real point of this report is political; ACPO wants to generate moral outrage so the craven and innumerate politicians, desperate to be seen doing something, give the police tough new powers to crack down on all prostitution. It's a power grab by a private organisation that serves its own members. personally, I think membership of ACPO should be illegal for any serving police officer, but that's a discussion for another day.

paulathomas said...

@Mike Greenhall

I haven't been looking at this research for long and haven't had time to read the report on which Belinda comments here so what I say will be more general.

I am a mathematician by academic training. As such I have been underwhelmed by the quality of the evidence gathering and analysis in the literature on trafficking.

It does no service whatsoever to the victims of trafficking to massage the figures. It does no service either to use the victims of trafficking for an agenda which has nothing to do with them. out of naivete, desperation or any other motivation. It also does a huge disservice to the victims of other forms of human trafficking to concentrate solely on sex trafficking.

The way to tackle this is by pointing out the flawed methodologies that litter the literature in this debate. Ben Goldacre has shown us how to do this. Let's do it!

Matt Greenall said...

@paulathomas: I'm totally with you, and we should hold people to account as the blogger has here when they make claims based on dodgy data. But I think we also need to come up with alternatives: more ethical and acceptable standards for research. Or else a strong rationale for not doing research. A few of us are trying to push for this from the HIV angle (internationally, global HIV work is pretty much the only place paying attention to sex work outside of the trafficking paradigm)...
Matt (not Mike!)

Michael said...

The problem with all the disputes about numbers, a game which has been going on for some time, (1) is it provides a very poor understanding of the phenomenon.

If people are seriously concerned about human trafficking for sexual exploitation and not simply going on a moral crusade (2) they need to understand this as a market economy not a monolithic moral panic. People leave their places of origin for a variety of reasons, and receive varying degrees of assistance along that trajectory which may include inducement or coercion.

The economies of both countries of origin and destination and their varying regulation of markets, immigration and sexuality are all important variables. The 2010 United Nations Office on Drugs and Crime (UNODC) report on the globalisation of crime (3) points out that the financial incentives for such traffickers are low, and that a well regulated legal market provides few opportunities for such activity.

Hysteria about numbers such as seen in the highly diverse media coverage of this report actually victimises the victims of trafficking further, to say nothing of others caught up in the xenophobia this arouses. It deflects from our understanding of the causes and solutions, which is fundamentally not a law and order problem.

Sources cited:
1. Washington Post 2007

2. Weitzer 2005

3. UNODC 2010

paulathomas said...

@matt greenall

Sorry for getting your name wrong!!

I agree up to a point. As I understand it, and perhaps an academic in this field can help us out here, there are high quality ethical standards for this type of research. They are followed by most researchers. It is just that Governments choose to listen to those who do not follow the ethical guidelines.

Phone Sex said...

As a sex worker, I'd say this piece mostly rings true to me. Definitely the issue of sex trafficking seems to be exaggerated out of all proportion to reality these days. I think opponents of sex work have realized that the moralistic arguments don't carry as much weight as they used to, so now they try to keep us criminalized by scaring people with the bogeyman of trafficking, and pretending that they are interested in our welfare.

Bigger Penis said...

The authors should have focused on the data they did have – ie 24 people, and explored those cases to explore how transferable the data was. There is a difference between ‘transferability’ of findings (ie to a similar population) and ‘generalisability’ (to a wider population) - which is what they tried to do but not sustainable with the data they had or the statistical tests they carried out. I mean why did they correlate the number of businesses with the number of police officers? Just what precisely was the research question?

Phone Sex said...

I believe the monolithic conception of human trafficking is problematic and I suggest that trafficking occurs as a result of women’s willingness to immigrate –for work, for a better life etc.
Phone Sex