Social network characteristics and alcohol use among young Swedes with different ethnic backgrounds

Abstract

Abstract Background The study explores how social network determinants relate to the prevalence and frequency of alcohol use of members of social networks. In a so-called dyadic design we study how similar alcohol habits co-exist among individuals (egos) and their peers (alters), when variables such as ethnic background, network composition, and other socio-cultural aspects are considered. Methods The data were derived from a Swedish survey entitled “Social Capital and Labor Market Integration: A Cohort Study.” The study participants (egos; n = 1989) were around age 23 at the time of the interview. A so-called dyadic design was applied, which means that all components of the analysis refer to ego-alter pairs (n = 7828). The outcome variable considered how alcohol prevalence and frequency of binge-drinking co-exist between egos and their alters. The independent variables also measured mutual attributes and behaviors - whether egos and alters were at the same age and sex, had same ethnic background, were relatives or friends, had similar religious affiliations, or intensely interacted with friends. Results The analysis revealed that ego-alter similarity in terms of age, sex and ethnic background predict ego-alter similarity in alcohol use and binge-drinking. For example, if egos and alters shared a similar ethnic background, their risk of alcohol use was at least 30 percent higher as compared to those with different ethnic backgrounds. Relative to ego-alter pairs with mixed ethnic backgrounds, the odds of binge-drinking were highest for ego-alters pairs with Yugoslavian background (OR 1.76; 95% CI 1.27-2.42), followed by those with Iranian (OR 1.57; 1.04-2.35) and Swedish background (OR 1.28; 0.84-1.95). Conclusions We conclude that network similarity (i.e., homophily) is an important explanation for the co-existence of alcohol use among members of peer networks. Alcohol use is more common in homogeneous peer dyads representing population groups with higher use. Key messages Peer similarity predicts alcohol use and binge drinking. Ethnic similarity of peers is associated with increased alcohol use and binge drinking.