Skip
To Content
|  | |||||||||||
 |
|
|||||||||||
|
Parental Influences on Adolescent Marijuana Use and the Baby Boom Generation |
||||||||||||
The increase in marijuana use by adolescents, particularly young teens, during the nineties (SAMHSA, 1998a, b; 1999; Johnston et al., 1998) is striking and remains to be understood. Following a steady decline from the high prevalence levels observed in the late seventies, sharp increases beginning in 1993 through 1997 were observed among young people, especially adolescents (see Figure 1.1). A slight decline has been observed among adolescents between 1997 and 1998. Among youths aged 12-17 in the United States, the annual use of marijuana reached a high of 21.3% in 1979, decreased to a low of 6.9% in 1992, increased to 15.8% in 1997, and was 14.1% in 1998 (SAMHSA 2000).
1 Estimates for 1979 through 1993 were adjusted for changes in question format as of 1994.
Sources: SAMHSA (1991; 1998b); University of Michigan.
Several explanations for the increase have been proposed: the coming of age of a new generation of youths who have had little direct exposure to the negative consequences of drug use in an era of declining drug use (Bachman, et al., 1998); a decrease in drug prevention efforts; and reduced governmental and media attention to the drug problem. Another relevant factor might be the influence of parents who, as members of the baby boom generation, belonged to cohorts heavily involved in marijuana use in their own youth. Because of their cultural and historical experiences in adolescence, these parents may be more likely to have children who use marijuana themselves.
Various individual and social factors have been related to drug use by young people. Individual factors include unconventional attitudes, lack of religiosity, inadequate school performance, and poor relationships with parents (Hawkins et al., 1992). Drug use in the immediate social environment of young person, especially drug use by peers, has consistently been identified as a most important factor (Bailey and Hubbard, 1991; Bauman and Fisher, 1986; Dishion and Loeber, 1985; Flay et al., 1994; Hawkins et al., 1992; Iannotti and Bush, 1992; Kandel, 1980; Urberg et al., 1990). Drug use by parents has also been found to be related to and to predict drug use by adolescents. Moreover, the influence of parents relative to the influence of peers may be stronger than is generally thought. Because of methodological issues, the influence of peer drug use relative to the influence of parental use may have been overestimated by as much as a factor of five (Kandel, 1996).
Relatively few studies have examined the impact of parental marijuana use on child use. To the best of our knowledge, we have identified eight studies. Most were conducted more than a decade ago and are about evenly divided among those that relied on perceived parental behavior (Fisher et al., 1987; Forster, 1984; Huba and Bentler, 1980; Johnson et al., 1984; Newcomb and Bentler, 1986) and independent parental reports of their marijuana use (Andrews et al., 1993; Brook et al., 1985; Gfroerer, 1987; Hops et al., 1996; Kandel, 1974; Kandel and Andrews, 1987). One study examined both types of reports (Newcomb et al., 1983). Parental effects of marijuana use have been documented for mothers and fathers, sons and daughters, with no differences by parent or child gender (Gfroerer, 1987; Hops et al., 1996; Huba and Bentler, 1980; Kandel, 1974; Kandel and Andrews, 1983; Newcomb et al., 1983). Only one study (Newcomb and Bentler, 1986) examined the association between perceived parental use and self-reported child marijuana use as a function of ethnicity. The association was stronger among whites and Hispanics than African-Americans. Similarity between parents and children in drug use has also been documented on the basis of independent reports from each respondent for the use of cigarettes (Kandel and Wu, 1995; Bauman et al., 1990; Gfroerer, 1987; Needle et al., 1986; Rittenhouse and Miller, 1984), alcohol (Barnes et al., 1986; Gfroerer, 1987; Green et al., 1991; Lau et al., 1990), and cocaine (Gfroerer, 1987). Gender and ethnic patterns of association with these other substances differ from those observed with respect to marijuana. Maternal influence for smoking appears to be more powerful than paternal influence (Kandel, 1974; Kandel and Wu, 1995), particularly among daughters (Charlton and Blair, 1989; Chassin et al., 1986; Clayton, 1991; Kandel and Wu, 1995). A sex-specific impact of parental alcohol use has been reported, particularly for sons (Cadoret et al., 1980; Cloninger et al., 1981) and to a lesser extent for daughters (Bohman et al., 1981). The influence of parental smoking on child cigarette smoking is greater among white than African-Americans and Hispanics (Griesler and Kandel, 1998; Sussman et al., 1987).
The National Household Survey on Drug Abuse (NHSDA) provides an unusual opportunity for investigating parental influences on children's marijuana use, particularly as a function of parental membership in the baby boom generation. In a number of households, up to two respondents were selected for participation in the surveys. In most of these households, one respondent was an adult, providing independent data on parental patterns of drug use in dyads of parents and children. The multiple surveys spanning close to a twenty-year interval provide variations in the adolescent drug experiences of parents, variations in the historical context of the parents' adolescent experiences, and variations in the historical periods in which the children reached developmental periods of risk for initiation into marijuana use.
Two earlier analyses of parent-child associations in drug behavior were conducted on the NHSDA. In analyses of parent-child pairs in the 1974, 1976 and 1977 surveys, Rittenhouse and Miller (1984) found that current maternal cigarette smoking and alcohol use were significantly associated with child lifetime and current marijuana use. Paternal cigarette smoking and alcohol use had no impact on youth substance use. Parental marijuana use was not examined. In subsequent analysis of dyads in the 1979 and 1982 surveys, Gfroerer (1987) examined same- and cross-drug associations for parent and child cigarette, alcohol, marijuana and cocaine use. Many more significant associations in substance use occurred for mothers than fathers. Although parental use of all classes of substances was associated with child marijuana use, the lifetime use of marijuana by mothers and fathers was strongly and uniquely associated with child lifetime and current marijuana use, controlling for sociodemographic characteristics and parental current cigarette smoking and alcohol use. The association between parental and child drug use was stronger for marijuana than for cigarettes, alcohol or cocaine.
We undertook research in the NHSDA to investigate the role of parents, especially members of the baby boom generation, on the marijuana use of children. We investigated the association of marijuana use between parents and children, the differences among parental birth cohorts, and the determinants of child marijuana use. We addressed five major research goals:
Develop a strategy to define parental exposure to the marijuana epidemic.
Assess the strength of the association between parental and child marijuana use according to pattern and extensiveness of use, by sex of parent, and age, sex and ethnicity of child.
Assess the impact of membership in the baby boom generation and parental exposure to the marijuana epidemic on child marijuana use. In particular, determine whether parents who were members of baby boom birth cohorts at the highest risk for marijuana initiation in their adolescence influence the behavior of their children differently from parents from other birth cohorts, i.e., baby boom cohorts that did not experience the explosion in marijuana use in the 1970s, or cohorts that preceded or followed the baby boom generation.
Determine the unique influence of parental marijuana use on the child's marijuana use, controlling for other determinants of the child's use, in particular, parental use of drugs other than marijuana.
Identify important predictors of marijuana use by young people in addition to parental marijuana use. This analysis is constrained by the limited data other than drug use available in the NHSDA.
This report is based on parent-child dyads available in NHSDA surveys conducted from 1979 to 1996, i.e. the 1979, 1982, 1988 and 1990-1996 surveys. Although dyadic data were ascertained in the 1974, 1976 and 1977 surveys, they are not available for analysis. Dyadic data were not collected in 1985. The report addresses the research goals outlined above through descriptive and multivariate analyses.
1.3 Organization of Report
Chapter 2 outlines the central hypothesis of the study. It proposes an empirical definition of exposure to the marijuana epidemic that is based on a differentiation between incidence and prevalence rates, identifies five different periods of the marijuana epidemic from 1963 to 1996 and identifies the developmental period of highest risk for onset of marijuana use. On the basis of these criteria, the chapter delineates nine groups of birth cohorts with different experiences of the epidemic in a crucial phase of adolescent development. The assessment of the impact of differential parental exposure to the marijuana epidemic on child marijuana use is presented in Chapter 6.
Chapter 3 briefly describes the NHSDA methodology and the dyadic samples available for analysis, and discusses the strengths and weaknesses of the NHSDA for achieving the aims of the research. The chapter also describes the measurement of variables used in the analysis. Complete details about the measurement and construction of the variables are presented in the Technical Appendix.
Chapter 4 presents the sociodemographic characteristics of parents and children in the dyadic sample. Characteristics of children and adults in dyads and those not in dyads are compared to assess the impact of non-systematic sampling on individuals selected for inclusion in dwelling unit pairs.
Chapter 5 reports the rates of marijuana use observed among parents and children and the associations between parent and child use through cross-tabulations and odds ratios. The chapter also includes a comparison of drug use patterns among youths and parents in dyads and those not in dyads.
Chapter 6 assesses the impact of differential parental exposure to the marijuana epidemic on children's marijuana use. The chapter also examines predictors of child marijuana use other than parental birth cohort and membership in the baby boom generation. Limited variables are available in the NHSDA data set to implement such analyses, since the content of the surveys focused almost exclusively on patterns of drug use. Two sets of multivariate analyses were implemented. One set consisted of multivariate logistic regressions, which provided an overview of the relative importance of different factors, including parental marijuana use, on the child marijuana use. In addition to membership in the baby boom generation, parental exposure to various periods of the marijuana epidemic and parental marijuana use, the predictors included parental use of other substances (cigarettes, alcohol, cocaine), and sociodemographic, attitudinal and personality characteristics of parents and children. The second set of analyses consisted of structural causal models, which provided a more dynamic understanding of the direct and indirect effects of selected parental and child factors on the child marijuana use.
The Technical Appendix provides details about the construction of the drug use and other selected variables.
Appendix tables present survey-specific data for the multiple surveys that are aggregated in most of the tables presented in the main body of the report.
This page was last updated on June 03, 2008. |
* Adobe™ PDF and MS Office™ formatted files require software viewer programs to properly read them.
Click here to download these FREE programs now
| Highlights | Topics | Data | Drugs | Pubs | Short Reports | Treatment | Help | OAS |
