“Inhumane”: Quantitative textual analysis of a qualitative survey on gender violence among victims of international human trafficking

Focus groups are often employed in research because they offer insight on perceptions and experiences from stakeholders. Qualitative interviews also allow researchers to get more in-depth and pragmatic information on the phenomena of interest, thus providing breeding ground for novel interpretations of results coming from solely quantitative analyses. Furthermore, several computational techniques, including machine learning and generative pre-trained transformers (GPTs), have helped develop quantitative analysis to process massive textual data sets. In this context, text mining has been used to examine and find key themes in group interviews’ transcripts to give objectivity to their analysis. Natural language processing (NLP) techniques can be used to first gather information on common themes, as well as to inform on the feelings risen from the text via more advanced sentiment analysis tools. The aim of this work is to examine the transcript of a focus group about gender violence among victims of international human trafficking, using several text mining techniques. The focus group involved two researchers and four key stakeholders working in the third sector with expertise on the matter – either by working with victims of gender violence, or by working with victims of human trafficking, or both – as well as a moderator. The stakeholders work in the province of Palermo (Italy) and refer to their experiences in this specific territory. Preliminary results show the reality of human trafficking: the harrowing experience of violence, the consequent loss of identity, coming from systemic and inhumane practices of women’s exploitation in the streets. Education and independence, while recognized as essential, are rarely sought after. This also highlights that the naturalization of gender violence makes reporting their perpetrators particularly difficult.