Once again garbage in, garbage out as a method for counting sex-trafficking victims, from the New York Times

I found myself looking up an old quarrel at Slate, between Jack Shafer and Peter Landesman. Landesman had written a trafficking story for the New York Times Magazine, and Shafer had debunked it at Salon.com. During the back-and-forth about trafficking statistics, Landesman cites Kevin Bales (who founded Free the Slaves) as the source of numbers of ‘sex trafficking victims in the US’. The argument stretched from 2004 to 2005; probably those involved would give different numbers now; what I’m pointing out is the gall of anyone, much less a social scientist, presenting this technique for estimating victims as an algorithm.

a very complex algorithm

Bad enough that Bales begins with an estimate for which no methodology has ever been given, but then the social scientist uses media reports as evidence. Did the scientist check into the sources mentioned by the media reports themselves? Did he distinguish at all between media who just copy and paste each other’s news and those who do any actual research? Did he exercise any scientific judgement at all as to reliability of any given media source? Just how circular can a process get?

… The estimate of 30,000 to 50,000 people being held in forced labor in the United States for purposes of sexual exploitation was arrived at in this way: firstly, we used the State Department’s estimate of 18,000 to 20,000 people being trafficked into the US each year. (Admittedly, the State Department has not explained the methodology by which they arrived at this estimate, so we use it in the hope that they will soon make their research methods clear.) Secondly, we adjusted this estimate according to two surveys we have recently conducted. The first survey was of all media reports of trafficking cases in the US over the past four years. These reports covered 136 separate cases of forced labor, 109 of which noted the number trafficked totaling 5,455 individuals. As with most crimes, the number of known and reported cases is a fraction of the actual number of cases occurring. To the best of our understanding the proportion of known to actual cases for human trafficking is low. In this survey 44.2% of cases involved forced labor in prostitution and 5.4% involved the sexual abuse of children, totaling 49.6%. As this is a rough estimate I rounded this up to 50%. In a second survey of forty-nine service provider agencies in the United States that had worked with trafficked persons, we asked how long each trafficked person they had worked with had been held in forced labor. The minimum reported time was one month, the maximum was 30 years. The majority of cases clustered between three years and five years.

So, if 9,000 to 10,000 of the people trafficked into the US each year will be enslaved for sexual exploitation (50% of 18-20,000), and they are likely to remain in that situation for three to five years, then the number of people enslaved for sexual exploitation at any one time in the US could be between 27,000 and 50,000 people. Since a number of people working in the area of human trafficking have stated that they believe the State Department’s estimate is low, I chose to make our estimate based on the upper end of the State Department figure, thus giving an estimate of 30,000 to 50,000.

Why should Garbage in, garbage out characterise nearly all efforts to estimate the number of trafficking victims? There is no straightforward way to count workers in the informal sector and undocumented migrants, whether they are suffering terribly or not and whether sex is involved or not. Estimating them can be carried out in various ways laced with caveats, but wild guesses are not even estimates. Anyone interested in serious work in this field can check out, for example, a report from the Economic Roundtable, in Los Angeles, entitled Hopeful workers, marginal jobs. No, there are no mentions of sex trafficking victims or, indeed, victims in general. Instead you will find methods not braggingly called algorithms for estimating numbers working in informal jobs in LA. You can also read about this in Harder Times: Undocumented Workers and the U.S. Informal Economy.

The fact that garbage is so prevalent in trafficking rhetoric demonstrates how little actual information proponents have. No one responsible would resist the arguments if there were real substance to back them up! Wake up, oh ye of too much faith!

— Laura Agustín, the Naked Anthropologist

4 thoughts on “Once again garbage in, garbage out as a method for counting sex-trafficking victims, from the New York Times

  1. Pingback: Twitter Trackbacks for Trafficking Statistics | Garbage | Sex | Shafer | Landesman | Border Thinking on Migration, Trafficking and Commercial Sex [lauraagustin.com] on Topsy.com

  2. Dave

    And those ridiculously inflated guesses that are commonly stated as facts are what get reported in the mainstream press while skepticism and more objective studies are largely ignored.

  3. Pingback: Held Together With Lies « The Honest Courtesan

  4. Alan Sussman

    This is just another example of the degree of scientific illiteracy our schools have failed to correct. It adversely affects all areas of our social/political discourse.


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.