Analyses next examined the moderating effects of the popularity of the target teen's closest peer and the duration of the friendship using the same approach described above. Results are presented in Table 3. Peer popularity was found to be a significant moderator of the relation between peer substance use and relative changes in teen substance use over the following year. Duration of friendship did not display any moderating effects. Figure 2 depicts the moderating effect of peer popularity, and shows that relative increases in teen substance use are much more likely to be predicted by high baseline levels of peer substance use when that peer is popular among other adolescents.
