Adolescent peer influence in the social media context: literature review

Relative to research on offline peer influence processes, research on online peer influence is in its infancy. In this section, we provide an overview of the extant research on online peer influence processes. It should be noted that this section is not meant to serve as a systematic review of the literature on adolescent social media use and peer influence; such a comprehensive review is beyond the scope of this chapter. Rather, we aim to capture the state of the literature: what we currently know, what is implied, and what research questions remain unanswered or ambiguous given the methodological limitations of the current work.

Longitudinal studies

Relative to studies of offline peer influence processes, very few studies have followed adolescents over time in order to understand longitudinal peer influence processes online. Below, we highlight some of the few exceptions, which examine two domains that may be subject to peer influence effects: substance use and sexual behaviours.

Substance use

Several recent studies have examined longitudinal associations between exposure to peers’ alcohol-related social media content and adolescents’ alcohol use. Alcohol-related social media content may include photos of peers drinking or being drunk or hungover, or other posts (e.g., status, links) about alcohol use. For example, in a sample of U.S. high school-aged adolescents, exposure to friends’ alcohol-related content on social media predicted adolescents’ initiation of drinking and binge drinking one year later, controlling for other risk factors for adolescent alcohol use (Nesi, Rothenberg, Hussong, & Jackson, 2017). These associations were mediated by injunctive peer nonns, or adolescents’ beliefs that their peers approved of drinking, suggesting that exposure to alcohol-related content may influence both attitudes and behaviours (Nesi et al., 2017).

Another study of U.S. adolescents suggests a complex relationship between online activity and offline smoking and alcohol use. Specifically, in a short-term longitudinal study of 10th grade students, adolescents who reported a greater number of close friends posting pictures of themselves “partying or drinking alcohol” on social-networking sites (SNSs) were more likely to report drinking alcohol and smoking six months later (Huang, Unger et al., 2014). No peer influence effects were found related to friends’ simply “talking about partying” online, highlighting the role of social media’s visualness in contributing to peer influence effects. In addition, an interaction effect was found such that adolescents whose friends did not use alcohol were more likely to be affected by exposure to risky online pictures (Huang, Unger et al., 2014). In a separate analysis of this data using stochastic actor-oriented models (Huang, Soto, Fujimoto, & Valente, 2014), the authors found that exposure to risky online pictures increased adolescents’ subsequent smoking, but not drinking, behaviours. On the other hand, this study found evidence for selection effects: adolescents with similar SNS-use habits, including similar exposure to risky online posts, were more likely to form and maintain friendships over time.

Finally, another study examined exposure to friends’ alcohol-related posts as predictors of first-year U.S. college students’ alcohol use, based on average number of drinks consumed per week (Boyle, LaBrie, Froidevaux, & Witkovic, 2016). Exposure to peers’ alcohol-related content predicted the emerging adults’ drinking behaviour six months later, and this effect was found to be stronger among males (Boyle et al., 2016). These effects were partially mediated by beliefs about the significance of alcohol to college life, nonns regarding other students’ drinking behaviour, and motives for drinking to enhance positive affect. Importantly, this study controlled for the effects of offline alcohol-related peer influence — specifically, perceptions of friends’ actual drinking behaviours (including drinking, getting drunk regularly, and drinking primarily to get drunk). These findings suggest a unique and additive effect of online peer influence, preliminarily supporting the transformation framework’s proposition that these processes are amplified online.

Sexual behaviour

Relative to longitudinal studies of online peer influence on alcohol-related behaviours, fewer studies have examined online influence on adolescents’ behaviours or attitudes related to sexuality. In one notable exception, a large sample of Dutch adolescents aged 13—17 was followed for six months (van Oosten, Peter, & Boot, 2015). Specifically, adolescents were asked how often they had been exposed to “sexy online self-presentations” through others’ photos on social media sites; examples included photos that included “sexy” gaze, appearance, posture, or clothing (pp. 1082—1083). Controlling for key demographic variables (e.g., sexual orientation, socioeconomic status), and for perceptions of friends’ injunctive norms regarding casual sex, higher exposure to others’ sexy online self-presentations was associated with higher probability of engaging in oral sex and sexual intercourse over the subsequent six months. However, no longitudinal associations were revealed for changes in sexual attitudes, and complex reciprocal interactions were revealed, in which some attitudes and sexual behaviours (but not all) predicted increased exposure to sexy online self-presentations over time (van Oosten et al., 2015).

In another study of Dutch adolescents aged 12—17, descriptive and injunctive peer nonns were longitudinally associated with the following “online sexual risk behaviours”: searching for someone on the internet with whom to talk about sex or to have sex offline, sending a partially nude photo to someone not known offline, and sending an address or telephone number to someone only known online (Baumgartner, Valkenburg, & Peter, 2011). Using autoregressive cross-lagged panel models across four waves, six months apart, the study found that peer norms were predictors, but not consequences, of these risky online sexual behaviours, and that descriptive nonns were stronger predictors of behaviour than injunctive nonns. Notably, these findings align with the transformation framework by highlighting opportunities for “novel” behaviours — such as online sexual risk behaviours — made possible by the social media environment.


In conclusion, several longitudinal studies provide preliminary evidence that online exposure to social media content related to alcohol or sexuality may predict changes in adolescents’ drinking and sexual behaviour over time. However, the extant body of work is limited in several ways, including that few longitudinal studies have assessed behaviour at more than two time points or over a follow-up period greater than one year, precluding the modelling of longer-term developmental trajectories related to online peer influence. Additionally, there is a significant need for more longitudinal studies that examine the unique effects of online peer influence, controlling for offline peer influence. Moreover, we are aware of no longitudinal studies that have examined peer influence on disordered eating, depression, non-suicidal self-injury, suicide, aggressive or deviant behaviours, or positive behaviours.

< Prev   CONTENTS   Source   Next >