Postema, Dagmar (2024) The Oldest Profession in the Modern Age: Developing a Revised Risk Score Assessment Model to Predict the Possible Presence of Human Trafficking and Illegal Prostitution in Online Escort Advertisements. Research Master thesis, Research Master.
![[img]](https://gmwpublic.studenttheses.ub.rug.nl/style/images/fileicons/text.png) |
Text
PostemaS4076370MasterThesis.pdf Restricted to Repository staff only Download (1MB) |
Abstract
The rise of the internet has created a novel platform for the advertising of sexual services, particularly through online escort advertisements. However, this digital shift has also been exploited by criminals in organizing and facilitating human trafficking and illegal prostitution. This thesis investigates potential risk indicators of these illicit activities within online escort advertisements in the Netherlands, a task that has become increasingly challenging for (municipal) law enforcement due to the sheer volume of advertisements and a decentralized approach. In response to this challenge, a risk assessment model was developed in 2015 to aid in identifying suspicious advertisements. This research evaluates the indicators used in the existing model through a comprehensive literature review and proposes new potential risk indicators. These form the basis for the quantitative analyses using a dataset of 2,746 online escort advertisements that were active in Municipality X in 2023. 100 subsamples of 125 advertisements, including 34 that were inspected by law enforcement and 91 randomly selected, were used in a bootstrap procedure in which multiple logistic regressions were performed to predict the possible presence of unwanted circumstances in online escort advertisements. The outcome of the inspection served as the dependent variable in multiple logistic regressions to assess the effectiveness of each model. Key findings include the improved performance of logistic regression optimized weights over human-determined ones. While the revised risk indicators alone do not outperform the current risk indicators, a combined model including all risk indicators significantly improves the predictive capabilities of the revised risk assessment model. The estimated models are employed to construct new risk scores, which are then compared with the current ones. The revised risk score model is then utilized to create a practical traffic-light model, which intuitively categorizes advertisements in one of three risk categories: low-risk (green), medium-risk (orange), and high-risk (red), ultimately aiding municipalities in the triage of which advertisement to select for an inspection.
| Item Type: | Thesis (Research Master) |
|---|---|
| Supervisor name: | Huitsing, G.E. |
| Degree programme: | Research Master |
| Differentiation route: | Lifespan Development and Socialization [Research Master] |
| Date Deposited: | 06 Aug 2024 07:10 |
| Last Modified: | 06 Aug 2024 07:10 |
| URI: | http://gmwpublic.studenttheses.ub.rug.nl/id/eprint/4231 |
Actions (login required)
 |
View Item |