Perceptions of friends' substance use versus friends' self-reported substance use
Meisel, Matthew Kirwin
MetadataShow full item record
Peer substance use is one of the strongest predictors of one’s own substance use. Both perceptions of peers’ substance use as well as peers’ self-reported substance use independently predict individuals’ substance use. However, there is a lack of literature on whether social network factors influence these perceptions, as well as a lack of understanding in how substance use is associated with social network outcomes. The current study utilized an egocentric social network approach, in which focal individuals enumerated the most important individuals in their lives, answered questions about the substance use of these individuals, and then indicated whether the network members knew one another. The current study went beyond the traditional egocentric network design and gathered self-report substance use from the network members. 114 focal individuals and 243 network members participated in the current study. Focal individuals significantly underestimated the frequency of network members’ alcohol, heavy alcohol, and marijuana use. There was no significant tendency for focal individuals to overestimate or underestimate the frequency of network members’ tobacco use. Social network characteristics did not moderate the relationship between perceptions and network members’ self-reported substance use. Supportive of the false consensus effect, across all substances, the relationship between focal individuals’ perceptions of network members’ substance use and focal individuals’ substance use was stronger than the relationship between network members’ and focal individuals’ self-reported substance use. The moderating role of focal individuals’ alcohol use on the relationship between network members’ alcohol use and degree centrality did not differ when network members’ alcohol use was operationalized as focal individuals’ perceptions than when it was operationalized as network members’ self-reported substance use. The same effect was evidenced when examining heavy alcohol use, but different interactions emerged for marijuana and tobacco use. This suggests that when utilizing a network approach to study alcohol and heavy alcohol use, it is not necessary to gather self-report data from the focal individuals’ friends. However, for tobacco and marijuana use, because of the differential effects of perceptions and self-report on network characteristics, it may be useful for researchers to obtain tobacco and marijuana use from the network members themselves.