Young Adults' Exposure to Alcohol and Marijuana-Related Content on Twitter

Background:  Exposure to substance use content via social media is an emerging area of inquiry, and research suggests that social media use may influence young adults’ substance use behaviors and norms. Most such studies, however, have focused on older social media platforms like MySpace and Facebook. Twitter is among the most popular social media platforms used by young adults, and evidence suggests it may yield richer data vis-à-vis young adults’ substance use behaviors and norms compared to older platforms. Nevertheless, Twitter remains underutilized in substance use research. Given Twitter’s popularity among young people, its unique attributes as a social media platform, and because alcohol and marijuana are commonly used substances among this age group, it is important to better understand interactions about alcohol and marijuana use behaviors that occur on Twitter. Informed by selective exposure theory, we conceived of exposure to pro-substance use content on Twitter as occurring (a) “passively” through the viewing of Tweets about peer alcohol or marijuana use or (b) “actively” by following Twitter handles that promote alcohol or marijuana use behaviors and/or posting Tweets about one’s own alcohol or marijuana use behaviors. We hypothesized that (1) those with active exposure to pro-alcohol or marijuana content on Twitter will be most likely to be current binge drinkers or marijuana users, and (2) those with active exposure to pro-alcohol and marijuana content on Twitter will have the lowest perceived risk of binge drinking and marijuana use.

Methods: Data were collected for this pilot study using a cross-sectional survey design. Participants (N = 587) ages 18 to 25 (M=21.8, SD=2.2) were recruited from SurveyMonkey® Audience—it has been previously used for health attitudes and behavior research—to complete an online survey during February 2014. The majority of the participants were female (n=333) and 80% (n=471) were Caucasian. Our dependent variables were current binge drinking (Y/N), current marijuana use (Y/N), and perceived risks associated with binge drinking and marijuana use, respectively (i.e., no-slight, moderate and great). Our independent variables were demographics and level of exposure to pro-alcohol and pro-marijuana content on Twitter (i.e., none, passive, and active). Multivariate binomial logistic models were used for dichotomous outcomes (i.e., current binge drinking, current marijuana use) and multinomial models were used for polytymous outcomes (i.e., perceived risks). Missing data were handled via multiple imputation using chained equations. All analyses were conducted using Stata 13/MP.

Results: Current binge drinking and lower perceived risk of binge drinking were associated with higher levels of exposure to pro-alcohol content. Similarly, current marijuana use and lower perceived risk of marijuana use were associated with higher levels of exposure to pro-marijuana content.

Implications: Our findings underscore the need for further research regarding young adults’ exposure to pro-substance use content via Twitter. They also highlight the importance of developing and testing prevention messages that can be used in social work and public health practice to effectively target and engage young people by countering online Tweets that promote alcohol and marijuana use behaviors.