Reader "Steve Dildo" has alerted us to the client list of the "Desert Divas," an Arizona prostitution ring that was busted three years back. Mr. Dildo was most interested in the presence of a "Kevin Pitsoogle" on the list, but we've determined that this name is fake, unconnected to the similarly named former West Virginia basketball player. That left us with the data set itself: the list contained the price paid for each encounter, along with the specific prostitute involved, the phone operator who set up the appointment, and the way that each client had been referred to the ring (by a website, by a friend, and so on).
While the plurality of encounters cost $375, many clients paid more or less than that. We looked at the 1,370 first-time transactions that deviated from the standard price—the average price among this subset was $510—to try to find the pattern behind the pricing.
Our investigation produced a model that explains 71 percent of the variance in cost. It's not just the identity of the prostitute that influences the charge: the identity of the phone operator is important as well. About 18 percent of the change in price can be explained by who the prostitute was, while 12 percent of the change can be explained by who the phone operator was.
That first figure seems self-explanatory, but why are the people who handle the setup so important? One possible explanation would be that a given receptionist tends to send out the same prostitutes, which would make the operators' effect a meaningless ghost of the prostitutes'. But that isn't the case: There's no significant interaction between the two groups. Our remaining best guess is still an unknown variable: Perhaps certain operators are assigned to clients expected to pay more, or are better at upselling clients to more expensive services.
The referral method was important as well: The interaction between the means of referral and the specific phone operator accounted for 7.2 percent of the total variance, and that between the referral and the prostitute explained 6.7 percent. This means that certain pairings of methods and employees explain a decent portion of the change in cost.
The cheapest clients came from "The Erotic Review," an online escort directory; our model predicts that clients who found the Desert Divas via that site paid just $400, compared to the $510 average for the non-standard-price crowd. The big spenders came to the Divas from Eros Guide, another site — clients from there would be expected to pay $679.
The moral of the story: be careful what prostitution rings you frequent, lest you become the subject of statistical hypothesis testing. Here are the test results, for those interested: