Author

Linda Flaga

Access Type

Open Access Dissertation

Date of Award

1-1-1999

Degree Type

Dissertation

Degree Name

Ph.D.

Department

Educational Psychology

First Advisor

Stephen Hillman, Ph.D.

Abstract

Alcohol use among adolescents continues to be a widespread problem with as many as four million alcoholics under the age of eighteen. The cognitive variables of self-efficacy and outcome expectancies are important in determining behavior and can be viewed as mediational components whereas predisposition variables are translated into actual alcohol use or abuse. The arena where these predispositions may find the greatest expression, via cognitive variables, is within the social/peer influence realm. This study examined the influence of each of these variables, the combination of the two, and the social/peer influence factors within each model on adolescent alcohol use. The study used a population of 11th and 12th graders from a suburban school system. The Situational Confidence Questionnaire (SCQ) was used as a measurement of self-efficacy. The Alcohol Expectancy Questionnaire-Adolescent Form (AEQ-A) was used as a measurement of outcome expectancies and the Adolescent Drinking Index (ADI) was used as a measurement of alcohol use. The data were analyzed using stepwise multiple regression analyses to compare the utility of the AEQ-A alone in predicting alcohol use, the SCQ alone in predicting alcohol use and the two instruments combined. In addition, the subtests reflecting "social/peer'' influences from both the AEQ-A and the SCQ were analyzed to assess their predictability related to alcohol use. The results of this study firmly support the premise that the combination of self-efficacy beliefs and outcome expectancies, in comparison to each alone, provides more accuracy when attempting to predict alcohol use in adolescents. Based upon the results of this study, the combination of self-efficacy belief measurement and outcome expectancy measurement combined is a better predictor of alcohol use in adolescents than either of these two models individually. It also appears that the model of “social/peer influence" within the realm of both self-efficacy and outcome expectancies related to alcohol use in adolescents may have some merit in prediction, although further refinement of this measurement needs to be completed.

Share

COinS