Abstract
The purpose of the present study was to examine adolescent alcohol use within the context of youth involvement and reinforcement from alternative leisure activities and also to determine whether perceived reinforcement from alcohol (outcome expectancy liking [OEL]) impacts this relationship. Participants were 956 students in grades 7 through 9 who participated in the Project on Adolescent Trajectories and Health (PATH), a 3-year longitudinal study of adolescent risk behaviors and health outcomes. A path model included Time 1 and Time 3 youth self-reported alcohol use, Time 2 youth ratings of reinforcement potential from alternative activities and OEL scores, and Time 3 perceived access to alcohol. The final model provided a good fit for the data and revealed several significant paths. In particular, alcohol use was positively associated with reinforcement potential from party attendance and involvement in social activities, and negatively associated with reinforcement potential from religious involvement and involvement in home/family activities. In addition, OEL was a partial mediator for party attendance and religious activities. These findings highlight the need for prevention programs that focus on reducing adolescent alcohol use through increasing access to substitutable leisure activities that can compete with behaviors maintained by alcohol's positive reinforcement value for youth.
Acknowledgments
This research was supported by a grant from the Social Sciences and Humanities Research Council and Canadian Institutes of Health Research (PI: Marvin Krank). The authors would like to thank Tricia Johnson, Peter Molloy, Daniel Lai, and the Central Okanagan School District #23 for their invaluable contributions to the Project on Adolescent Trajectories and Health.
This article is dedicated to the memory of Dr. Anne-Marie Wall.
Notes
Note. OEL = outcome expectancy liking; scores range from −4 to +4.
***p < .001.
Note. Bolded values reflect factor loadings >0.30.
a These activities were not included in the principal components analysis.
Due to concerns regarding the reliability of the measurement metric for activity involvement, we transformed the reinforcement potential items into standardized scores and conducted a principal components analysis using the transformed scores. The findings from this PCA paralleled those for the analysis involving the non-transformed scores. Four components were identified, accounting for 45.9% of the variance in reinforcement potential scores. Items comprising each of the components were identical to those in the non-transformed PCA.