Use of locus of control and self-efficacy to predict outcome in alcohol treatment



Journal Title

Journal ISSN

Volume Title


Texas Tech University


Alcohol misuse is undoubtedly one of the most significant drug problems confronting this nation today. Unfortunately, the attrition rates from various alcohol treatment programs range from approximately 13% to 39% with an average of 28% (Baekeland & Lundwall, 1975). Thus, it is important to identify those factors that are predictive of treatment outcome in an alcohol and drug abuse program (ADAP) which could ultimately lead to a reduction in dropout rates. The present study examined two social-learning constructs, namely locus of control (Rotter, 1966) and self-efficacy (Bandura, 1977), in terms of their ability (both independently and together) to predict successful completion of an ADAP. A number of other variables including self-deception/impression management, depression, and demographic factors were analyzed. It was hypothesized that individuals with relatively higher self-efficacy and a more internal locus of control would be more likely to complete treatment successfully. In addition, it was hypothesized that locus of control would correlate negatively with self-efficacy. Finally, self-efficacy was predicted to be a stronger predictor of treatment success than locus of control. A sample of 150 patients admitted to a 30-day ADAP in a west Texas state hospital were administered measures of self-efficacy, locus of control, depression, and self—deception/impression management plus a demographic questionnaire. Although locus of control was correlated negatively with self-efficacy, it was the unsuccessful treatment group which demonstrated relatively higher self-efficacy as well as a slightly more internal locus of control (opposite of what was predicted). While some of the self-efficacy factors were stronger predictors of treatment success than locus of control, other self-efficacy factors were not as strong. Interestingly, the unsuccessful group also had higher scores in self-deception and impression management. This latter finding was given as one possible explanation for the unexpected results. Another possibility for these findings might lie in the fact that the self—efficacy instrument employed in this study is fairly new and has yet to undergo rigorous validity testing. Finally, the overall utility of these constructs for making predictions regarding successfully completing complex tasks was examined.