An evaluation of two approaches to drinking risk reduction with college students: cognitive-behavioral skills training and motivational feedback

Abstract

The current study examined the relative effectiveness of motivational feedback compared to cognitive-behavioral skills training delivered in bibliotherapy format to college students. The effectiveness of matching alcohol abuse interventions with participant's stage of change was also explored. The study's design utilized an assessment only control group and collateral informants to strengthen the validity of the conclusions.

Participation in the cognitive-behavioral skills training and motivational feedback interventions did not lead to greater reductions in participants' alcohol use and alcohol related problems than assignment to an assessment only control group. No significant differences were found between these three groups on any of the drinking measures from pre-test to l-month and from pre-test to the 2- month follow-up. Only time effects were observed on alcohol related problems and frequency of heavy alcohol use across the 3 experimental conditions.

Tentative explanations for the lack of support for the effectiveness of these interventions are discussed in the paper including sample size and statistical power, intervention compliance, and intervention fidelity. A number of changes which might enhance the effectiveness of these interventions are also discussed in the paper.

The study also found no significant Intervention X stage of Change X Time interactions to support the effectiveness of matching intervention to participants' stage of change. Tentative hypotheses for this finding are discussed in the paper including issues related to the measurement of stages of change. Competing hypothesis regarding the relative and incremental utility of outcome and self-efficacy expectancies were also examined in the study. Pre-test efficacy expectancies consistently predicted a significant proportion in participants' frequency of alcohol use and moderate use at 1 and 2-month follow-ups beyond concurrent use outcome expectancies were unable to add to the prediction of drinking behavior beyond self-efficacy and concurrent drinking behavior.

This pattern of results support Bandura's (1986) view that efficacy expectancies subsume most of the predictive power of outcome expectancies. Further, as predicted by Bandura, prior drinking behavior alone did not fully explain future use, rather students perceptions of efficacy also guided future use.