Over the past decades, the social sciences have paid great attention to socially undesirable behavior among adolescents, such as substance use, unsafe sex, or reckless driving (for reviews, see e.g., Dahl, 2004; Reyna & Farley, 2006; Steinberg, 2007, 2008). Although the various disciplines differ in their specific foci, they typically rely on longitudinal surveys for data gathering, usually due to obvious ethical and practical constraints in the experimental manipulation of socially undesirable behaviors. However, the importance of longitudinal surveys for the study of socially undesirable behavior among adolescents is at odds with our knowledge about a pressing, but under-researched issue in that type of research, the question-behavior effect.

The question-behavior effect refers to the possibility that the mere asking of survey questions about particular behaviors changes these behaviors (Fitzsimons & Moore, 2008; Sprott et al., 2006). Research on the question-behavior effect has predominantly focused on socially desirable behavior, such as donating money to charity, and socially neutral behavior, such as purchasing goods (for a review, see Dholakia, 2010). Lately, however, researchers have also started to investigate socially undesirable behavior. For example, a survey-based experiment has shown that college students who were asked about their intention to use illegal drugs consumed more illegal drugs in the 2 months after the survey than students who had not received that question (Williams, Block, & Fitzsimons, 2006).