The purpose of this study was to compare the effectiveness of trad~tional group therapy treatment for alcoholic out-patient clients with group therapy plus approaches that incorporated behavioral self-control techniques. Reducing anxiety in social situations was the focus of the behavior therapy treatments. Only alcoholics who were highly anxious (eightieth percentile), according to their scores on the Institute for Personality and Ability Testing (!PAT) Anxiety Scale, were used as subjects. The subjects were 24 clients from the San Joaquin County Alcoholism Rehabilitation Center who were all undergoing group therapy at the beginning of the study. Trios of subjects were matched on their !PAT scores and then randomly assigned to one of three treatment groups: control (group therapy only), desensitization (group therapy plus systematic desensitization), and combined behavioral (group therapy plus systematic desensitization plus covert sensitization and covert self-reward). Treatment effectiveness was measured by administering the Michigan Alcoholism Screening Test {MAST) and the San Joaquin County Alcoholism Screening Test (SJCAST) to the subjects at the beginning of the study (pre-testing) and at the completion of the behavioral treatment three months later (post-testing). Each subject's spouse or significant other also responded to the MAST and SJCAST at pre- and post-testing. Eighteen of the original 24 subjects, 17 males .and one female, completed the study. Each of the four measures of treatment effectiveness was analyzed using a split-plot factorial 3.2 analysis of variance, with type of treatment as the between subjects variable and pre- and post-testing as the within subjects variable. There was no significant difference between groups at pre-testing (except for the MAST taken by alcoholics where the desensitization group scored worse than the other two groups). At post-testing, all four measures showed significant differences between the three types of treatment (F's = 6.5, 6.3, 4.7, 11.6; 2/15 df; p < .05), with the two behavioral treatment groups showing significantly more improvement than the control group in seven of .the eight comparisons. The two behavioral treatment groups did not differ from each other. The main effect -for pre- and post-testing showed significant improvement for all groups, again on all four measures (F's - 354.9, 120.5, 87.8, 72.0; 1/15 df; p < .01). Correlation coefficients between alcoholic and significant other's responses ranged from +.61 to +.81 (p < .01 in call cases) on both the MAST and SJCAST at pre- and post-testing. The results demonstrate that all three types of therapy were effective in achieving improvement from pre- to post-testing. Adding behavioral self~control treatment to group therapy, however, resulted in even greater improvement than group therapy alone.