Methodological problems associated with an Empirical Study of Stigmatised Workers

We reflect on major methodological problems that we encountered in our empirical study of 109 streetworkers, conducted in 2008-2009 in Melbourne, Australia (Cregan, Kulik & Salinger, 2013). The problems arose from the stigma attached to streetworking and the subsequent marginalisation of the workers. We discuss and evaluate the effectiveness of the means we used to overcome the problems. The first problem was concerned with access and conduct of the enquiry. It was difficult to directly approach streetworkers. Workers are suspicious of strangers, fearful of arrest, and concerned with attracting clients. There was also a possibility of creating difficulties for the workers, causing them physical risks from clients and procurers, bringing about harassment from police, and provoking abuse from passers-by (Sanders, 2004). In response, we contacted the workers via four 'drop-in' centres for adult sexworkers. We administered our survey in a private, safe room in each centre. The second problem consisted of the choice and use of an appropriate method of enquiry that employed statistical analysis of a representative sample, while eliciting the views and perceptions of an understudied group of workers. In response, we used a multi-methods approach (Teddlie & Tashakkori, 2009). We conducted an inductive content analysis of the workers’ responses to open-ended survey questions (Neuendorf, 2002). We then carried out a deductive multivariate analysis of variables constructed from valid and reliable psychometric scales that comprised the items in the closed survey questions. The themes derived from the content analysis were used to inform our selection of these variables. We concluded by discussing the problem of assessing the trustworthiness of data collected from workers, most of whom were heroin addicts, and many of whom engaged in coping strategies to create a false identity (Grandey, 2003).