STANDING STILL OR MOVING FORWARD? A RESPONSE TO THE COMMENTARIES

The commentaries on my paper 1 are thought-provoking. Simons-Morton & Kuntsche 2 state that the paper provides new perspectives on an old ‘truth’, while Perkins 3 seems to suggest that it adds nothing at all. Borsari and Carey 4 think that I raise some interesting questions, but reject my conclusion that adolescents' overestimation of peer substance use (SU) has been exaggerated. Conversely, Hannigan & Delaney-Black 5 argue that this conclusion is understated because research on biological markers of SU indicates that young people's reports about their peers' SU are more reliable than their self-reports. Borsari & Carey state that ‘notably absent from this review is the recognition that international studies suggest that the phenomenon of exaggerated substance use norms is highly generalizable’ (p. 886). However, referring to research from 11 different countries, I point out that norm misperception in relation to alcohol use seems to be ‘widespread across nations and drinking cultures’ (p. 878). It may be added that almost all studies in the field are confined to young people, but a Canadian study 6 concluded that adults also overestimate others' drinking. The existence of the phenomenon thus appears to be well documented, but its prevalence and magnitude are probably far smaller than the research literature indicates. To ensure self–other comparability, I suggested that data on own and perceived peer SU could be collected in a real-time group setting using ‘the other survey participants’ as a comparison target. Borsari & Carey's response is that LaBrie et al. 7 have already applied this approach. The study in question evaluated a group-specific normative feedback intervention and data on SU were indeed collected in a group context. However, ‘the typical student’ was the only target, and the respondents' perception of the other group members' SU was thus not assessed. To my knowledge, the suggested group-based data collection strategy has not been employed in any study. Perkins also describes some studies incorrectly. He refers to two studies based on objective measures [i.e. blood alcohol concentration (BAC)] rather than self-reports, and claims that both showed that students typically exaggerate peer drinking. However, one of these studies 8 indicated that the majority (62%) believed that other students had about the same or lower BAC than themselves and that the modal tendency was to hold ‘accurate’ beliefs about the issue. The other study 9 included no measures on the perception of peers' BAC. There are several reasons why the research literature is likely to be biased in favour of demonstrating that youth overestimate peer SU, and many of my arguments are equally relevant with respect to studies on (mis)perceived attitudinal norms. While I focused mainly on methodological issues, Simons-Morton & Kuntsche suggest that publication bias also may have been at work. Indeed, when one considers all the weaknesses that characterize this literature, the paucity of critical voices is remarkable. The assumption that perceived SU norms influence adolescents' drinking and drug use (cf. 2-4) is worth some final comments. Few well-conducted studies have addressed the issue, but due to its longitudinal design, high response rate and the inclusion of several important confounders in the analyses, the study by Juvonen et al. 10 stands out. It revealed that adolescents' perceived SU norms did not predict subsequent drinking and drug use independently. High-quality studies demonstrating the effectiveness of the social norms approach to SU prevention are also lacking 11. Moreover, it should be noted that to the extent that young people exaggerate peer SU, their misbeliefs are mainly about distal social groups 12, 13—which are the least likely to have an influence on behaviour 14. None.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.