The purpose of the study was to measure the size and structure of the underground commercial sex economy in the United States.
The study employed a multi-method approach, using both qualitative and quantitative data. Existing datasets documenting the market changes for illegal drugs and weapons were analyzed to measure changes in these markets and estimate the overall size of these markets. This was done by measuring changes in a series of "proxy" variables, which were assumed to be proportional to underlying activity. Thus, official national datasets that measured some sort of drug and gun activities over a period of time were collected to measure these changes.
Qualitative data was collected through interviews with 119 stakeholders and 142 convicted offenders, including local and federal law enforcement officers, prosecutors, pimps/sex traffickers, sex workers, and child pornographers. A total of 119 in-person, semi structured stakeholder and offender interviews were conducted about the structure of the underground commercial sex economy (UCSE), the profits generated through the UCSE, networking within the UCSE, and changes in the UCSE over time.
Researchers employed a targeted, purposive sample of urban areas as study sites. Sites were selected from a list of the 100 largest metro statistical areas by population as defined by the Census Bureau and then narrowed down to 17 sites based off the following criteria:
- Number of convictions in the Federal Justice Statistics Resource Center (FJSRC) data for UCSE-related offenses (at least 20).
- Recommendations from UCSE subject experts.
- Existence of a federally-funded human trafficking task force.
- Willingness of local law enforcement to work with researchers on this issue.
- Availability of gun and drug data to be used as proxies
- Geographic location.
- Where the city falls within known "pimp circuits" in the United States.
Participants were then sought out in the following ways:
- Researchers searched for online published news stories and press releases on individuals who were arrested, adjudicated, and/or convicted on UCSE-related charges in the eight cities and their surrounding metropolitan areas.
- Researchers reached out to the Human Trafficking Clinic of the University of Michigan Law School, since they had recently launched the Human Trafficking Law Project (HTLP), a database that documents cases of human trafficking in the United States.
- Used subscription to CourtLink to gain access to court documents for these cases.
- Asked stakeholders in each of the eight cities to suggest potential respondents who were convicted on UCSE-related charges.
All pimps, sex workers, and child pornographers identified in eight cities in the United States from 2003-2007.
Interviews with pimps prostitutes
Secondary analysis of data files listed in document: Dataset Guide 06.28.2013.docx, which is included in documentation files and publicly downloadable.
administrative records data
This study includes four excel files that contain information on interviews with prostitutes and pimps as well as secondary analysis of economic factors in the 8 cities.
- file14-SW_data_archiving.xlsx- This file contains 37 cases of prostitutes interviewed for this study. Questions include information on arrest/convictions, demographics, history of sex work, type of sex works (clients, preferences, etc.) safety, and violence while on the job.
- file7-Pimp_data_archiving.xlsx- This file contains 74 cases of pimps interviewed for this study. Questions include information on charge, sentence, demographics of accused, how they became involved in pimping, history with drugs, history of owning a business, rules and expectations while pimping, acts of violence committed, payments, advertising, and perception of risks while pimping.
- Proxy data (8-17-13).csv- This file includes the variables used for analysis from the secondary data files, which are those taken from secondary data sources in order to understand the market changes for illegal drugs and weapons.
- Raw proxy data (8-21-13).csv- This file contains numbers directly from the base datasets. It shows the original variables before they were recoded, transformed, or otherwise manipulated.