Adolescent Non-Involvement in Multiple Risk Behaviors: An Indicator of Successful Development?

Abstract

Based on the conceptualization of successful development as the joint maximization of desirable outcomes and minimization of undesirable outcomes (Baltes, 1997), the present study examined connections between adolescent non-involvement in multiple risk behaviors and positive developmental status. Results from a survey of 7290 high school students were used to define four profiles of risk behavior involvement (complete non-involvement, some involvement, some high-risk involvement, predominantly high-risk involvement) based on self-reported involvement in nine risk behaviors (alcohol, smoking, marijuana, hard drugs, sexual activity, minor and major delinquency, direct and indirect aggression). Groups were compared across intrapersonal (risk behavior attitudes, temperament, well-being, religiosity, academic orientation), interpersonal (parental relations, parental monitoring, friendship quality, victimization, unstructured activities), and environmental (school climate, neighborhood conditions) domains. Despite some similarities between the complete non-involvement and some involvement groups, the complete non-involvement group had the most positive self-reports compared to each of the other groups in each developmental domain. At the same time, higher levels of positive development were not exclusive to the complete non-involvement group. Implications for research and theory related to connections between adolescent non-involvement in risk behaviors and successful development are discussed.

This research was supported by grants from the Social Sciences and Humanities Research Council of Canada and Human Resources Development Canada. Order of authorship is alphabetical for the last seven authors.

Notes

Note. N = 7290.

¹Given the relatively low internal consistency of the composite scales of academic orientation and religiosity, we reran the analyses with only one item from the academic orientation (i.e., grades) and one item from the religiosity (i.e., importance of religion) composites. The pattern of results did not change.

Note. Ns = 428, 1621, 4696, and 545 for NON, SOM, SHR, and PHR groups, respectively. Means and standard deviations are adjusted for age, gender, and parental education. Within a row, means are significantly different in pair-wise contrasts at p < .001 except those sharing the same subscript. ES = partial η² effect sizes.

Note. Ns = 428, 1621, 4696, and 545 for NON, SOM, SHR, and PHR groups, respectively. Cell entries are Cohen's d effect size estimates showing the standardized differences between groups based on adjusted means and standard deviations.

Note. Ns = 428, 1621, 4696, and 545 for NON, SOM, SHR, and PHR groups, respectively. Results should be read by row. Cell entries show the percentage of respondents in a given group (column variable) classified at the ‘asset-level’ for a given developmental index (row variable). Percentages are rounded to nearest whole number.

²The mean number of assets per respondent in the NON group was significantly lower for males than females in the present study (M = 6.58 and 7.83 for males and females, respectively, p < .001). In addition, the percentage of males in the NON group classified at the asset level was significantly lower than females for the risk attitudes, academic orientation, religiosity, parental monitoring, friendship quality, and peer victimization domains (all ps < .001). At the same time, however, with the exception of friendship quality, the percentage of males in the NON group classified at the asset level for these domains was higher than the percentage of their peers classified at the asset level in the SOM, SHR, PHR groups.