When Junk “Science” is Used to Attack Sex Work
As you might recall, Craigslist removed the adult services category from the site last September, under intense political pressure, which was founded on the argument that the classifieds were being used to promote child sexual trafficking and underage sex work.
Deborah Richardson, the chief program officer of the Women’s Funding Network, was one of the primary movers and shakers behind this claim, which was repeated in the media and led to her cross-country tour:
“An independent tracking study released today by the Women’s Funding Network shows that over the past six months, the number of underage girls trafficked online has risen exponentially in three diverse states,” Richardson claimed. “Michigan: a 39.2 percent increase; New York: a 20.7 percent increase; and Minnesota: a staggering 64.7 percent increase.”
But it turns out that these numbers are junk science, at best.
The Village Voice is reporting that when they interviewed people who actually understand statistics, research methods, and science, they had plenty of things to say about these numbers.
A University of Minnesota professor who teaches students how to design research said that the study is fundamentally flawed. And the chair of the Anthropology Department at John Jay College of Criminal Justice in New York thinks that the study “seems pretty bogus.” Given that he led a Justice Department-funded study on juvenile prostitution in New York City in 2008, he has grounds for his skepticism.
When these and other experts looked a little more closely, they found that the folks behind these claims used shoddy methods, faked their data, and lied to Congress. All so they could get publicity and money. The numbers were the result of a project by the Schapiro Group, which had no academic researchers. They also didn’t have any prior experience studying prostitution. In order to “study” the phenomenon of under-18 sex workers, they simply went to online-classified sites and counted the number of pictures that looked like they were of minors.
Why don’t journalists link to primary sources? Whether it’s a press release, an academic journal article, a formal report or perhaps (if everyone’s feeling brave) the full transcript of an interview, the primary source contains more information for interested readers, it shows your working, and it allows people to check whether what you wrote was true. Perhaps linking to primary sources would just be too embarrassing.
–from the Guardian
Anyone who knows the first thing about sex work will know that, just as in Hollywood, the fashion industry, or (for that matter) the mainstream job market, women consistently strive to appear younger than they are in order to be marketable. Whatever your personal feelings about that might be, I’m sure you’ve noticed that, too. That’s going to skew the data.
There’s also the question of what markers make someone look underage. Sometimes, it’s a choice of clothing. Or a hairstyle. Or the fact that they’re skinny. Or that they have very smooth skin (perhaps, thanks to Photoshop.) So how do you make that work in this kind of research? Well, the Schapiro group asked 100 people to look at photos of teens and young adults and guess which ones were minors. They then came up with this:
“The study showed that any given ‘young’ looking girl who is selling sex has a 38 percent likelihood of being under age 18,” reads a crucial passage in the explanation of methodology. “Put another way, for every 100 ‘young’ looking girls selling sex, 38 are under 18 years of age. We would compute this by assigning a value of .38 to each of the 100 ‘young’ girls we encounter, then summing the values together to achieve a reliable count.” (taken from the Voice article)
This is pure nonsense, of course. For example, were the photos that these folks looked at comparable to online sex ads? Were the people wearing similar clothing and in similar poses? How consistent were the results? Did each person clock in at about 38%, or were some people higher and some lower? Can they duplicate these numbers with a different group of participants? And how do the demographics of the participants influence their perception of the age of a person in a photograph? For that matter, does the context make a difference? Do people assess the age of someone in a photograph differently when it’s on an escort’s ad?
But that didn’t stop the folks behind this from having six new observers count how many “young-looking” women were selling sex online and claiming that 38% of them were minors. This was such good news for them that the Women’s Funding Network paid Schapiro to dramatically expand the study to include Michigan, Minnesota, New York, and Texas, and they’d like to include the entire US.
Their next step was to repeat the survey of online ads every three months, which is how they came up with their claim that Minnesota had a 64% increase in child prostitution in such a short time. But that doesn’t address the possibility that seasonal fluctuations in sex work can affect things. Or that maybe some sex workers were posting multiple times, rather than more individuals doing so. Taking two data points and charting an “exponential” curve (as Deborah Richardson, the chief program officer of the Women’s Funding Network, described it) is simply wrong. I learned that in 8th grade math.
It also turns out that when they were asked, the folks at Schapiro couldn’t explain how they got their numbers, where they found the images they used to “calibrate” the 38% that everything rested on, or how the researchers knew the ages of the women in the photos. For that matter, when a Village Voice reporter spoke with Beth Schapiro, she flipped from saying that “this is not a precise count of the number of girls being prostituted” to being “specific numbers.”
Even though these numbers were so badly generated that they wouldn’t have been accepted for a student paper, much less a peer-reviewed study, the Women’s Funding Network shouted them from the rooftops and got both the funding they wanted and shut the erotic services section of craigslist down.
It’s worth mentioning that the Village Voice also owns Backpage.com, one of the websites that sex workers advertise on, so there’s a potential conflict of interest. On the other hand, their reporter interviewed people who conduct rigorous academic research, as well as experts in the field of sex work, and child abuse and they have serious doubts about the report. I also notice that Beth Schapiro’s response attacks the objectivity of the reporter without actually addressing any of the methodological questions that these experts raised.
Unfortunately, these sorts of imaginary claims work really well if you want to create a panic. For example, we see all sorts of inflammatory headlines about thousands of sex workers flocking to cities that host the Olympics or the Super Bowl and when the dust settles it just doesn’t happen like that.
I’m 100% in support of protecting and taking care of children and teens. And I strongly believe that nobody should be forced into sex work (or for that matter, any kind of labor. But let’s not get distracted by the issue of trafficking and domestic or agricultural labor.) But adults who are choosing to enter sex work deserve the same respect and protection as anyone else. I’m getting tired of people using the issue of children to create a moral panic about what consenting adults do. And I’m getting tired of people making things up in order to scare folks into giving them money. Lying and screaming aren’t justified if the facts don’t support your claims.
Sex workers deserve better than that. And so does everyone else.
Careless use of the word ‘exponential’ like that is an immediate red flag when in comes to supposedly research-based claims.
Came in here to say the same thing. Misuse of ‘exponential’ makes me want to HULK SMASH!