My friends, I'm #SOTALLYTOBER: A longitudinal examination of college students' drinking, friends' approval of drinking, and Facebook alcohol-related posts. (May 2019)

• Record Type: Journal Article
• Authors: Steers, Mai-Ly N.; Neighbors, Clayton; Wickham, Robert E.; Petit, Whitney E.; Kerr, Bradley; Moreno, Megan A.
• Is Part Of: Digital health. Volume 5(2019)
• Publication Date: 2019-05
• Subjects: Social media -- alcohol -- peer influences -- college students; Medical care -- Data processing -- Periodicals; Medical informatics -- Periodicals; 362.10285
• DOI: 10.1177/2055207619845449
• Languages: English
• ISSNs: 2055-2076

Abstract

Background: Literature has consistently shown a positive relationship between young adults' social media alcohol-related posts and drinking outcomes; however, the reasons for this association and the psychosocial influences behind students' posting of alcohol-related content are still unclear. Peer influences have been robustly shown to predict students' drinking such that students' perceptions of their friends' drinking is positively associated with their own drinking. Objective: Although research has demonstrated that online and offline peer influences are robust predictors of drinking among college students, perceptions of friends' approval and students' drinking in relation to alcohol-related posting have yet to be explored longitudinally. Methods: The current multi-site, multi-method study examined students ( N =316; 58.7% female) from a Midwest (58.8%) and Northwest university over a 4-year period. All Facebook alcohol-related posts were coded each academic calendar year and perceived friends' approval of drinking and students' alcohol use were assessed annually. A lagged, random coefficients negative binomial model was specified to examine between- and within-person effects. Results: After controlling for perceptions of friends' alcohol-related posts, results revealed that time, drinking more, and perceiving friends as more approving of drinking were significantly and positively associated with posting alcohol-related content at the between-person level. Moreover, a significant interaction of Time X Drinking, with drinking at the between-person level, emerged such that heavier drinkers tended to post less often over time. Conclusions: Increases in alcohol-related content posts are likely to over-inflate students' drinking norms and their drinking. Thus, it is plausible that social media networks containing more alcohol-related content may contribute to cyclical increases in drinking for individuals within that network.