Peer And Parental Influences Associated With Adolescent Marijuana Use

An analysis of relationship influences on adolescent drug use, coming from research examining how family and friend relationships affect resistance to drugs. For scholars in communication, social psych, health psych, family studies

The purpose of this study was to examine the impact of parental factors on annual (past year) marijuana use in African American youth within the Greater Cincinnati area. The following research questions were examined for this study: 1) Do family structure, parental education and parental job status impact African American adolescent marijuana use? 2) Does impact of all these variables differ based on sex and grade? Participants were African American students in 7th through 12th grade (N = 7488) in public and private schools within the Greater Cincinnati area. Logistic regression analyses revealed that annual marijuana use differed based on family structure, parental education, and parental job status. More specifically, annual marijuana use was highest among students not living with both parents, students who had parents with a high school diploma/GED or less, and students who had parents who were unemployed. This held true for males, females, junior high and high school students, with one exception family structure did not significantly impact male marijuana use. Findings should be considered by health educators and preventionists when developing programs and efforts to prevent youth marijuana use.

Peers and peer relationships play an important role in the development of marijuana and other substance use in adolescence. Although social influence (peer influence, social contagion) is frequently cited as a primary determinant of non-illicit substance use in adolescence (e.g., alcohol and tobacco use), it is unclear to what extent social influence plays a role in the development of adolescent marijuana use, or what social-environmental factors may moderate the effect of social influence on adolescent marijuana use. The present study utilized a novel longitudinal community detection method (DLSMM; dynamic latent space mixture modeling) and the latent space estimator of Xu, 2018 to examine the association between the formation of longitudinal peer groups in a high school social network, social influence, and the development of marijuana use. Utilizing three waves of high school social network data collected from the PROSPER peers project, DLSMMs identified three longitudinal peer groups over the high school period (10th - 12th grade). The peer groups varied on important demographic, substance use, family, and school characteristics across the three waves, and exhibited differing levels of membership stability across time. Membership in the identified groups predicted later marijuana use, and the centrality of marijuana users compared to non-users differed within the overall network and the identified groups. Over time, marijuana users were more likely to move-to the periphery of the network compared to non-marijuana users. Latent space adjusted linear-in-means models identified time-varying social influence effects. In particular, the marijuana use of an adolescent's social connections in 11th grade increases the probability of adolescent marijuana use in 12th grade. Furthermore, the effect of social influence on marijuana use in 12th grade was stronger for students who were members of peer group 1 in 11th grade compared to students who were members of peer group 3 in 11th grade, controlling for important demographic characteristics of the sample. The results of this dissertation collectively highlight the importance of adolescent peer groups to both the development of marijuana use, and the strength by which social influence processes have an effect on marijuana use behaviors.

The purpose of the present dissertation was to examine risk perception as a potential moderator for the relationship between parental substance use involvement and adolescent marijuana use. This dissertation used data from the 1997 and 1998 National Household Surveys on Drug Abuse (NHSDA). In some households, an adult and an adolescent were included in the sample. A total of 2481 parent-child pairs were derived, so that adolescents' marijuana use involvement and perceptions could be analyzed in relation to parental use of marijuana, alcohol, and cigarettes. The parent and the adolescent were interviewed independently and in confidentiality, using the same standardized questionnaire. Substance use measurements used in the present study included lifetime; past year, and past year dependency (for both the parent and the adolescent); risk perception was based on the adolescent's perceived risk of physical harm from weekly marijuana use. Data analyses allowed for the control of potential confounders, including: age, gender, race, location, socio-economic status, peer marijuana use, and marijuana use parental attitude, accommodating statistically the complex sample design. As expected, parental use of marijuana, alcohol, and cigarettes was associated with adolescent marijuana use. In most of the models to test for moderation, "great" (high) risk perception was inversely associated with the influence of parental use on adolescent marijuana use. The odds of marijuana use among adolescents who perceived great risk in weekly marijuana use and who had a parent who had ever used marijuana were greater compared to adolescents with a parent who had never used marijuana (OR = 3.4, 90% CI = 1.18, 9.64), but these odds of marijuana use were lower compared to adolescents with a parent who had ever used marijuana but did not perceive great risk (OR = 7.6, 90% CI = 3.34, 17.33). An ancillary finding was that parental substance use was inversely related with adolescent risk perception of marijuana. Limitations notwithstanding, this research suggests that risk perception could function as a protective factor and could be a potential component of selective interventions, which involve children of substance users. However, more research is needed to better understand the underlying mechanisms for risk perception among children exposed to parental substance use.-- Abstract.