Combining web surveys and general population surveys to improve data on people who use drugs in France: reflections on recruitment strategies and the generalisability of results

This paper looks at the implementation and recruitment strategies of the 2016 European Web Survey on Drugs (EWSD) in France. It also explores the possibility of generalising the EWSD results by using a matching procedure that links EWSD respondents with respondents of the 2017 general population survey (GPS) in France. Two non-probabilistic recruitment strategies for the EWSD were tested. The first strategy sought to reach drug users through specialised websites related to drugs, while the second strategy targeted a sample of French Facebook users through a paid advert that was posted on around 177 000 Facebook pages. As the paper shows, different recruitment strategies yielded different samples of respondents, in terms of their drug use patterns and demographic characteristics. Further, the paper uses a matching procedure, namely propensity score matching, to explore if inferences could be made to the general population based on responses to questions in the EWSD. In this way, the paper shows how it was possible to estimate the prevalence of use of several drugs in the general population, even though these questions were not asked in the GPS. This example contributes to the ongoing work on generalising web survey results to larger populations.