Abstract

Recent studies of youth drug use have increasingly focused on the influence of perceived risk of harm and individual disapproval in determining use. Prominent studies suggest that increases in perceived risk of harm from marijuana use influence decreases in use directly and indirectly through an individual's disapproval of use. This hypothesized causal relationship has yet to be fully explored at the individual level. The current study examined the social sources of beliefs about the risk of harm from marijuana use and tested the hypothesized causal relationship between perceived risk of harm, individual disapproval, and past month frequency of use in a nationally representative sample of 17,166 youth ages 12 through 17. Data were from the 2002 National Survey on Drug Use and Health. Through structural equation modeling, a social influence model of youth marijuana use tested the hypothesized causal relationship noted above. As part of the social influence model, the strength of perceived disapproval of parents and friends for use was examined relative to the strength of perceived risk of harm in determining past month frequency of use. As perceived risk of harm was theorized as a social product whose interpretation could vary by social location, the social influence model was also tested for subsamples by gender and total family income. Results of the analysis for the full sample suggest that though the social influence model was a good fit to the data, the influence of perceived risk of harm on past month frequency of marijuana use was largely direct and not mediated through an individual's disapproval of use. Additionally, individual disapproval had a very weak effect on past month use, controlling for other estimated effects. The latent factor disapproval from friends had the strongest total effect on individual disapproval for the full sample, while disapproval from parents had the strongest total effect on past month use. When analyzed separately by income level, individual disapproval had a very weak, statistically non-significant effect on past month use for both low and higher income youth. Controlling for other estimated effects, disapproval from friends had the strongest total effect on use for low income youth, while disapproval from parents had the strongest total effect on use for higher income youth. When analyzed separately by gender, individual disapproval had a very weak effect on use for both males and females, though statistically significant for the female subsample. The total effect of perceived risk of harm on use was stronger for males than for females. Disapproval from parents had the strongest total effect on past month use for both males and females. Overall, results suggest that though an important variable, perceived risk of harm from marijuana use was not a primary determinant of individual disapproval in light of controls. Informal mechanisms of social control such as perceived disapproval of parents and friends were often stronger influences on past month use than both perceived risk of harm and individual disapproval. These results suggest that drug prevention policies should focus on the influence of more informal mechanisms of social control in deterring use. A discussion of the current study's limitations and recommendations for future research are provided.