Challenging demand for prostitution: international evidence review

3. Methodology

Evidence search

The search and analysis of the evidence was conducted between October 2021 and March 2022. The identification of relevant resources began with a desk-based library search on the Scottish Government Library Service, KandE, which has access to an extensive range of online search engines and databases. These have been summarised in Annex A.

To capture as much relevant material as possible the search included the following terms:

• For criminalisation models: criminalisation, prostitution, evaluation, implementation, evidence/literature review/synthesis, sex trade, sex trafficking, demand/purchase/buyer of sex, pimp.
• For decriminalisation/legalised models: decriminalisation, legalised, prostitution, evaluation, implementation, evidence/literature review/synthesis, sex trade, sex trafficking, demand/purchase/buyer of sex, pimp.

In order to supplement the 2017 evidence review and avoid duplication, findings were limited to literature published since 2016. Similar searches were conducted for New Zealand and the Netherlands, however, the search timeframe was extended to better capture the scope of research published since the adoption of current legislative approaches to prostitution (2003 and 2000 respectively). Research published since 2016 was, nevertheless, prioritised where possible so as to best reflect recent developments and lessons learned.

The first library sweep was supplemented with searches on legislation websites for each jurisdiction to provide an overview of the relevant legal provisions. These have also been included in Annex A. Search engines such as Google and Google Scholar were used to identify publications and grey literature that had not been identified in the initial search and to stay up to date on any new material that emerged in the course of the evidence review. Further resources that appeared relevant to the research aims were also sourced through references in the literature identified in the initial sift. Each resource was assessed on the basis of its relevance to the research aims through the application of the following inclusion and exclusion criteria:

Quality assessment

In addition to determining the relevance of the literature to the aims of the research, the robustness of empirical studies was assessed on the basis of the following quality criteria:

• clarity of the research aims/hypotheses and research questions;
• acknowledgement of related research and theories;
• declaration of funding and vested interests;
• suitability, transparency and evaluation of the methods used;
• suitability and representativeness of the interviewees/participants/sample;
• extent to which the conclusions were based on the information gained from the methods used;
• and the extent to which bias had been minimised throughout the study.

From this criteria a classification was calculated to capture the quality of the research, with "low" referring to studies where no or few methodological criteria had been fulfilled, "medium" where some of the methodological criteria appropriate for the study type had been fulfilled, and "high" where all or most of the methodological criteria appropriate for the study type had been fulfilled. These classifications should not be interpreted as definitive measures of the quality of the research but rather are supplied to provide an indication of the quality of the studies examined.

A total of 53 empirical studies were included in the review. The majority were classified as medium on the robustness scale, reflective of the lack of reliable and robust empirical evidence on prostitution (Wagenaar 2017). Given the inclusion/exclusion criteria specified above, no studies scoring low were included in the review. The most common limitations of the studies identified in the evidence were:

• Small sample sizes;
• A reliance on third-party data or convenience sampling limiting the representativeness of findings;
• Lack of transparency and evaluation of research questions and methods used;
• Lack of specification of the diversity of participant views;
• A lack of comparison group.

Studies were in the main observational, providing limited insights into contributing factors and the scale and validity of the phenomena identified. An overview of the research design, limitations for each study and scoring category given are provided in Annex B.

Evidence gaps

A number of gaps were identified in the evidence which hindered the assessment of the approaches adopted across jurisdictions in some areas. Of particular note was a lack of research into:

• "What works" with regards to the implementation of challenging demand approaches. There was a notable scarcity in empirical studies directly addressing the effectiveness of the interventions introduced across the cases examined and defining what "success" might look like in practice.
• The day-to-day implementation (i.e. operational practice) of challenging demand approaches both in terms of enforcement and support provision. Little was found which detailed the activities engaged in by frontline agencies to support challenging demand objectives or their effectiveness. Relatedly, there was scarce research examining regional and local variations with regards to implementation.
• Monitoring approaches used by government and non-government agencies to identify the numbers of those selling and purchasing sex online/indoor and outdoor. Beyond national evaluations, little research was found which detailed any ongoing monitoring mechanisms for measuring the numbers of individuals accessing support services.
• Up-to-date statistics on penalties and legal proceedings. The availability of overall figures for arrests, penalties issued and convictions related to challenging demand offences was limited across jurisdictions.
• Social attitudes around prostitution and the effects of the approaches adopted on deterrence. With the exception of two surveys and some qualitative research conducted with men involved in the purchase of sex, little research was identified which explored the deterrence effects of challenging demand interventions.
• The experiences of those involved in prostitution, particularly of men and migrants (who sell sex). The majority of studies which examined the views and experiences of those involved tended to focus on female support service users. The lack of insights into men's experiences and the views of women who do not access support services (such as some within seldom heard from groups, e.g. migrant and trafficked women and girls) means important insights into the needs of those most vulnerable may remain unreported in existing research.

Moreover, insights into evidence produced in languages other than English were limited to what was presented and described in the literature assessed which may mean some important features of the approaches here examined have not been adequately accounted for. The lack of research published in English was particularly marked in the cases of France and Norway contributing to a reliance on a few key texts.