The international literature suggests that people are more likely to admit to using psychoactive substances by checking a box on a self-administered questionnaire—especially when the questionnaire is completed on a computer—than by answering a researcher’s questions over the phone. Based on a comparison of two French surveys conducted in 2005— the Life Events and Health Survey (EVS) and the Health Barometer, this chapter analyzes the potential impact of the data collection method on the reporting of such practices by individuals aged 18 to 64. Unlike in the U.S., where there is unanimous agreement on the superiority of self-administered questionnaires, the results of the two surveys appear to be highly consistent here: the prevalence of illicit drug use, reported using a computer-administered self-administered questionnaire with headphones (Audio-CASI) in the EVS survey, is found to be particularly close to that reported in the Health Barometer, which was conducted via telephone interview. Audio-CASI, on the other hand, seems particularly well-suited for reporting alcohol consumption. Analyses based on logistic models show that Audio-CASI appears to be better suited for young men, while telephone interviews yield more convincing results for women and older adults. With some nuances, these two data collection methods appear, at first glance, to be well-suited for sensitive topics. [author’s abstract]