This study uses a sub-sample from the Youth Leisure Study (YLS; Yardley et al., 1996). The sample for this study contained 420 students randomly selected from secondary schools in the Niagara Region of Ontario, Canada. The sample was stratified by grade and contained 44% males and 56% females.

Sensation seeking refers to "the seeking of varied, novel, complex, and intense sensations and experiences and the willingness to take physical, social, legal, and financial risks for the sake of such experience" (p 10) and was measured using Cooper’s (1994) scale. Bandura (1997) defines self-efficacy as “people's judgments of their capabilities to organise and execute courses of action required to attain designated types of performances. It is concerned not with the skills one has but with judgments of what one can do with whatever skills one possesses” (p. 391). While Bandura’s self-efficacy scale was domain specific, in this study a general measure of self-efficacy was used (Scherer et al., 1982). Substance use was measured using the YLS substance use scale created by Yardley et al. in collaboration with professionals in the substance use field.

Factor analyses, reliablities analyses and examination of residuals tested the distributions of predictor and outcome variables. In all but one case, the outcome variables conformed to theoretical models. The residuals for the marijuana use outcome were unaffected by transformation procedures and remained problematic.

Moderated hierarchical multiple regression (MHMR) analyses were used to examine relationships among predictor, moderating, and outcome variables. In order to examine the moderating relationship, an interaction term (sensation seeking x general self efficacy) was computed. Prior to creation of the interaction terms, each variable underwent a centering transformation to reduce chances of multicollinearity (Aiken & West, 1991). In this study, the regression procedure contained 3 steps; step 1 - demographic covariates, step 2 - predictor variables, and step 3 interaction terms. Step 1 contains demographic covariates (age, gender, importance of religion) which were identified in the literature as being influential in substance use relationships. By entering the interaction terms in step 3, the unique predictive ability of these terms can be examined.

Results of the regression analyses are presented in Table 1 below. General self-efficacy was a significant, negative, predictor of all forms of substance use (alcohol B= -.086, p<.05; tobacco -.125, p<.01; and marijuana use -.154, p<.001; binge drinking -.109, p<.05; and number of times drunk -.113, p<.05) indicating that as general self efficacy increases, substance use decreases. In addition, sensation seeking was found to be a significant, positive predictor of all forms of substance use (alcohol B= .369, p<.001; tobacco .215, p<.001; and marijuana use .287, p<.001; binge drinking .345, p<.001; and number of times drunk .334, p<.001) indicating that as sensation seeking increases, substance use increases.

Table 1:
Sensation-seeking and general self-efficacy moderated hierarchical regression analyses predicting substance use outcomes

------------------------------------------------------------------------------
                 Alcohol     Binge     # of Times     Tobacco      Marijuana
                   Use      Drinking      Drunk         Use            Use
Predictors 
------------------------------------------------------------------------------
Step 1 Covariates
(total R sq.)      .113***   .129***     .075***       .082***      .069***

Step 2 Substantive Predictors
Sensation-seeking
(beta values)      .369***   .345***     .334***       .215***      .287***

General Self-efficacy
(beta values)     -.086*    -.109*      -.113*        -.125**      -.154***

Step 3 Interactions
(Total R sq.)      .002      .003        .002          .002         .012*

sensat x selfeff
(beta values)     -.044     -.052       -.040         -.048        -.112*

TOTAL R sq.        .233***   .259***     .198***       .145***      .191***
------------------------------------------------------------------------------
* = p<.05,  ** = p<.01,  *** = p<.001

However, in step 3 of the regression analyses, the interaction term was significant for the sensation seeking marijuana use relationship only. Under conditions of high sensation-seeking, marijuana use for adolescents with low general self-efficacy was higher than for adolescents with high general self-efficacy. This relationship is presented in Figure 1.

The results indicate that sensation-seeking is positively and significantly related to each of the substance use measures in this study. This finding extends previous research (e.g., Bates, Lobouvie, & White, 1986; Zuckerman, Ball, & Black, 1990) by including substance abuse (i.e., binge drinking and number of times drunk) which were also found to have a positive relationship with sensation seeking. This significant relationship among sensation-seeking and forms of substance use is consistent across each of the substance use outcomes (alcohol, tobacco, and marijuana use, binge drinking and number of times drunk). This positive relationship with sensation-seeking may indicate that adolescents use and abuse substances simply for the sensation they provide.

The findings for general self-efficacy with the substance use variables indicate that substance use decreases as general self-efficacy increases. This finding supports previous research that found relationships among self-concept variables and forms of substance use. In this study, the consistent pattern across all substance use variables (alcohol use, tobacco use, marijuana use, number of times drunk, and binge drinking) shows that general self-efficacy is an important variable in understanding adolescent substance use and abuse. Higher levels of general self-efficacy are associated with lower levels of substance use in adolescents. This may be due to adolescents using substance use as a means of coping with a low self-evaluation, such as low self-efficacy (Bandura, 1977; Cooper, 1994). General self-efficacy predicts the use of these substances but in this study it does not moderate the influence of sensation-seeking- alcohol use and abuse relationships, or the sensation-seeking-tobacco use relationship. Therefore, it appears that, with the exception of marijuana use, the prediction of substance use by sensation-seeking is not moderated by general self-efficacy. High sensation seekers with high levels of general self-efficacy were less likely to be marijuana users than high sensation seekers with low levels of general self-efficacy. The nature of marijuana use may be important in understanding this relationship. The other substance use variables in this study (alcohol use, tobacco use, number of times drunk, and binge drinking) may not hold the same sensational quality that marijuana use does. In marijuana use, the risk of legal repercussions are more serious and, therefore, one might argue that the activity is more sensational. While for many adolescents, the use of alcohol and tobacco is still illegal, they result in a less severe criminal penalty than the penalty for marijuana use. This supports the hypothesis that for adolescents, high levels of general self-efficacy may provide a buffer or insulate against sensation-seeking drives toward deviance.

Future research may examine other dimensions of self-concept to determine their role in the sensation-seeking-substance use relationship. In this study general self-efficacy, a more general measure of self-efficacy was used. It is possible that different dimensions of the self-concept may influence sensation-seeking-substance use relationships in different ways. Therefore, the moderating effects of other measures of self-concept (e.g., self esteem, self confidence, physical self-concept) should be examined to determine their influence on sensation-seeking-substance use relationships.

A greater understanding of the mechanisms which influence and contribute to substance use and abuse may lead to the development of programs designed to address the complex needs of adolescents. Increased research may lead to decreased adolescent substance use creating beneficial repercussions in many areas, including crime prevention, school performance, as well as psychological and physical health. However, a strong foundation of knowledge regarding the multi-faceted behaviour of adolescent substance use is required before these positive outcomes can occur.

1. Aiken, L. S. & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. London: Sage.
2. Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioural change. Psychological Review, 84, 191-215.
3. Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman and Company.
4. Bates, M. E., Lobouvie, E. W., & White, H. R. (1986). The effect of sensation-seeking needs on alcohol and marijuana use during adolescence. Bulletin of the Society of Psychologists in Addictive Behaviours, 5, 29-36.
5. Butler, J. T. (1982). Early adolescent alcohol consumption and self-concept, social class and knowledge of alcohol. Journal of Studies on Alcohol, 43, 603-607.
6. Cooper, S. (1994). Motivations for alcohol use among adolescents: Development and validation of a four-factor model. Psychological Assessment, 6, 117-128.
7. Kinnier, R. T., Metha, A. T., Okey, J. L., & Keim, J. (1994). Adolescent substance use and psychological health. Journal of Alcohol and Drug Education, 40, 51-56.
8. Newcomb, M. D. & McGee, L. (1991). Influence of sensation seeking on general deviance and specific problem behaviours from adolescence to young adulthood. Journal of Personality and Social Psychology, 61 (4), 614-628.
9. Sherer, M., Maddux, J. E., Mercandante, B., Prentice-Dunn, S., Jacobs, B., & Rogers, R. W. (1982). The self-efficacy scale: Construction and validation. Psychological Reports, 51, 663-671.
10. Yardley, J. K., McCaul, K., Baker, J., Hornibrook, T., & Christian, M.(1996). Leisure activity and substance use connections among youth. In Proceedings of the 8th Canadian Congress on Leisure Research, 317-321.
11. Zuckerman, M., Ball, S. A., & Black, J. (1990). Influences of sensation-seeking, gender, risk appraisal, and situational motivation on smoking. Addictive Behaviors, 15, 209-220.