Data sources and methods

Data from the Measures to Address Prostitution Initiative (MAPI)

All of the data in this study were collected from organizations who received funding from the MAPI program, during the period of 2015-2016 and 2019-2020. Two types of organizations received the MAPI funding: non-profit organizations and law enforcement agencies. The organizations who received MAPI funding are located in Nova Scotia, New Brunswick, Quebec, Ontario, Saskatchewan, Alberta, and British Columbia.

There are two key sources for the MAPI data in this study: client-based reports (2,289 clients from 13 non-profit organizations) and semi-structured interviews conducted with 18 MAPI organizations during the summer of 2021. MAPI organizations also supplied 129 narrative reports: 81 were from non-profit organizations and 48 from law enforcement agencies, which provided some qualitative information on their clients’ experiences. These reports include interim progress reports and annual summaries.

However, the majority of data in this study come from the client reports and interviews.

For the client reports, at the beginning of the funding period the 13 non-profit organizations were given an Excel spreadsheet with specific variables and categories covering three general areas: client information, characteristics at program intake and characteristics during program delivery. The variables and categories are listed below. These MAPI organizations were also given two documents with instructions for completing the Excel dataset, a coding guide and a user guide.

The following is a list of the pre-determined variables developed by the MAPI program staff at Justice Canada:

Client information variables

Characteristics at program intake

Characteristics during program delivery

The two open-ended variables, “Reasons for seeking help to exit prostitution” and “Describe changes/impacts”, were used to correct coding for some of the pre-determined variables and to create multiple new variables. For example, both open-ended variables were used to correct coding in the “Family status” variables, such as identifying clients who were pregnant or whose children had been apprehended by child welfare services. The open-ended variables were also used to code which respondents had exited or remained in/returned to the sex trade, as well as several variables on outcomes such as housing, mental and physical health, and substance abuse issues.

Data analysis was conducted without survey weights. Counts under 10 were suppressed to protect the privacy and confidentiality of MAPI clients. All percentages are presented in the tables and charts unrounded at the first decimal place and then percentages are rounded in the text.

Finally, semi-structured interviews were conducted with select staff at organizations who received from funding from the MAPI program between 2015-16 and 2019-20. Of the 23 organizations that received funding, 18 consented to and completed an interview between June 23, 2021 and August 12, 2021. Most interviews were conducted in English, a small number were done in French.

There are eight standard questions in the interview guide, which was sent to interviewees at least a day before the interview. Usually one question was modified with a quotation from a narrative report to get more context on the clients of the specific organization. The standard interview questions are listed below. Interviews did not strictly follow all of these questions and may have included follow-up questions based on the information being provided by interviewees.

Limitations

A few limitations with the data used in this study should be noted. There is some inconsistency in the reporting of client characteristics in the MAPI client dataset. During interviews, one MAPI organization representative noted that they were only able to select one characteristic per variable in the Excel dataset. For example, they only selected “Indigenous” instead of “Indigenous” and “Trafficked” which would have better represented their client. However, most of the MAPI organizations were able to provide multiple categories for the same variable, such as “Indigenous” and “Trafficked”.

All MAPI organization representatives explained in the interviews that their organization’s primary goal is relationship building with the clients. Staff do not enforce the idea that clients must exit the sex trade, they instead engage in safety planning and harm reduction strategies no matter what situation the client is in or has chosen.

In addition, some clients may not have fully disclosed all of their characteristics due to lack of trust. Over time, MAPI organizations have made efforts to update their client records when they became aware of new information. Therefore, when a client is present in several years of data for one organization, their characteristics from the most recent fiscal year were used for the dataset in this study. To the extent possible, duplicates were removed from the dataset. Clients with the same identifier but with vastly different characteristics and outcomes were kept in the dataset.

Finally, not all MAPI organizations offer the same programs and services. There is variability among the non-profit organizations in terms of reporting which programs their clients accessed and completed. For example, some organizations used the generic categories provided in the coding guide to code for programs and services, while others wrote in the exact name of the program, which required some online research in order to code to the generic categories.