Authors: Valeria Ivaniushina; Vera Titkova
The goal of our project is the meta-analysis of studies employing SAOM methodology to evaluate the effects of social influence on adolescent drinking behavior.
We collected academic journals, dissertations/theses, reports and conference materials that use SAOM methodology through systematic electronic searches in Web of Science, Scopus, PubMed, The Cochrane Library, EBSCOhost, ProQuest, PsycINFO (PsycNET), Embase and CINAHL. The search syntax extracted 252 articles. We identified 26 articles eligible for meta-analysis We extracted characteristics of the study and its results: country, year when the study was conducted, number and frequency of survey waves, model specifications and effect size data. We conducted meta-analysis to estimate the effect size of peer influence on alcohol consumption behavior among adolescents of different age groups.
There are different effects for testing peer influence (average similarity, total similarity, average alter) that are not directly comparable. Hence, we analyze different specifications of the peer influence effect separately. Parameter estimates collected from individual studies were analyzed using multilevel random-effects models [Gallupe et al., 2018; Viechtbauer, 2010]. Meta-analyses for different specifications of the peer influence effect were conducted separately.
The results show that the mean effect size was positive for average similarity, total similarity, and average alter. The mean effect size for peer selection was strong positive, too. Conversion log odds ratio values to Cohen's d give estimates as a medium to large effect. It demonstrates that alcohol consumption spread based on adolescents’ preferences to select similar friends and adjust their drinking behavior to friends’ behavior.
Add to my calendar
Create your personal schedule through the official app, Whova!Get Started