Debra W. Armstrong and Dr. Renata T. Forste, Sociology
Many adolescents today are involved in high risk behaviors which have negative health consequences, and which reduce the likelihood of them growing up to be responsible adults.1 Past research of high risk behavior has been largely based on cross-sectional, area specific data. In this research, national longitudinal data are used to predict high risk behaviors by focusing on risk behaviors such as alcohol use, or smoking cigarettes. Few studies have considered these behaviors together and as noted by Jessor, these behaviors are more of a lifestyle than just separate events.2
According to social bond theory, juveniles engage in high risk behavior if they do not have strong ties to society.3 Hirshi identifies four types of social bonds or ties: (1) attachment, caring for other people; (2) commitment, willingness to conform; (3) involvement in conventional activities; (4) beliefs about following rules.3 In an effort to further understanding of the link between social bonds and high risk behaviors, the effect of family attachments, school involvement, and beliefs on high risk behaviors is modeled. Understanding the antecedents of a high risk lifestyle will promote the early identification of factors associated with high risk behaviors and facilitate intervention.
To estimate the effect of social bonds on high risk behavior, data from the National Survey of Adolescent Males is used. The NSAM was collected in 1988 and is a nationally representative sample of never-married, 15-19 year old males in the United States. A follow up survey was collected in 1991; both surveys provide background, attitudinal, and behavioral measures for a sample of 1,676 young men. The attitudinal and background measures from the 1988 NSAM are used to predict high risk behavior reported in the 1991 NSAM.
Models are estimated using logistic regression techniques. The coefficients represent the increase or decrease in the log odds of being involved in high risk behaviors (versus not) associated with a unit or category change in the independent variables.
The dependent variable is a dichotomous measure indicating whether or not the respondent is involved in multiple high risk behaviors. High risk behavior is defined as smoking cigarettes and drinking alcohol at least weekly. Independent variables include: gender attitudes, school involvement, family structure, religiosity, and self-efficacy measures. Social bonds regarding conformity and beliefs are measured by gender attitudes, religiosity, and an indicator of selfefficacy. Traditional gender attitudes, are hypothesized to be associated with high risk behaviors.
Religiosity is measured by looking at religious affiliation and attendance, as well as respondents reports of how important religion is to them. High religiosity is expected to decrease participation in high risk behaviors. Self-efficacy is measured by attitudinal questions concerning self-esteem. High self-efficacy is expected to decrease the likelihood of participating in high risk behaviors.2
School achievement, an indicator of involvement in conventional activities, is measured by reported grades and educational aspirations. Having low school involvement is expected to increase the likelihood of participating in high risk behavior. Social bonds of attachment are indicated by family ties. Family attachments are measured by respondents’ reports of parental influence, family structure at age 14, contact with nonresidential fathers, and whether their mother worked outside the home. Strong family attachments are expected to decrease high risk behavior. Finally, demographic controls for age, socioeconomic status, and race are also included.
Multivariate results indicate that adolescent males for whom religion is unimportant are more likely to be involved in high risk behaviors than those who feel that religion is very important. Those with non-traditional gender attitudes are less likely to be involved in high risk behaviors than those with traditional attitudes. Also those who report high self efficacy are less likely than those who report low to participate in high risk behaviors. Those with low grades, as opposed to high grades, are 2.4 times more likely to be involved in high risk behaviors. Being influenced often by what parents think significantly reduces the likelihood of being involved in high risk behaviors.
Based on social bond theory, Hirshi argues that if an adolescent has bonds to society, the adolescent will be less likely to engage in high risk behavior.3 The importance of social bonds in predicting high risk involvement is supported by this study. As shown by the results, adolescents with strong beliefs and conformity, school involvement, and strong family ties are less likely to be involved in high risk behavior than adolescents with weak social attachments. Social bonds, therefore, as predictors of high risk behavior, can help identify high risk adolescents and promote early intervention.
References
- Alexander, C. S. et. al. “A measure of risk taking for Young Adolescents: Reliability and
Validity Assessment.” journal of Youth and Adolescence, 19(6):559-569. - Jessor, Richard. 1992. “Risk behavior in Adolescence: A psychosocial Framework for
Understanding and Action.” In David E. Rogers and Eli Ginzberg (eds.), Adolescents at risk: Medical and social Perspectives, pp. 19-34, (boulder, CO: Westview Press). - Hirshi, T. 1996. Causes of Delinquency. Berkeley, CA: University of California Press.