An experimental, randomized, control-group, pretest-postest design was employed in this study to examine the efficacy of the use of two short-term group counseling interventions employing Rational Emotive Behavior Therapy with at-risk adolescents in changing their beliefs, altering their attitude toward school, and moderating behaviors that are problematic in an educational setting. Forty-eight subjects were included in this study from an alternative evening high school setting and were assessed in the initial stage of treatment using The Idea Inventory and The Majoribanks Attitude-To-School Inventory. The School Social Behavior Scale was employed at this beginning point to observe these students by teachers and administrators with regard to social competence and anti-social behaviors. Subjects were selected from a pool of regularly attending students in an alternative high school program and randomly assigned to three groups, one employing REBT concepts and techniques alone, one using REBT concepts and techniques in conjunction with the therapeutic board game, Let's Get Rational, and a control group. The treatment and control group designations were also randomly assigned to the counseling groups. Ten weekly 50 minute group counseling sessions were conducted by two master's level counselors employing the tenets of REBT with adolescents. Group participants had the opportunity to learn new interpersonal skills and behaviors that may be helpful to them in staying in school. At the conclusion of the 10 week treatment group sessions, the subjects in the three groups were re-assessed employing the same instruments mentioned before to determine if the activities in the group sessions made significant differences in the rational thinking, attitude to school, and school social behaviors of these at-risk students in an alternative setting. Data were analyzed using descritpive statistics in addition to a factorial analysis of variance (ANOVA) to measure the pretest and posttest performance and understanding of the subjects. Results revealed that the differences in the scores of the treatment groups were significant the REBT group employing the game, Let's Get Rational, contributing to students learning and understanding of the tenets of REBT at the .05 alpha level on the four dependent variables. There were no significant differences between genders regarding the four dependent variables, and there was a lack of a significant interaction between the intervention approach taken and the gender of the group member. There was no significant interaction between the gender of the subject and the treatment group placement. The results from both REBT groups support the efficacy of using the tenets of REBT with school aged populations who are at risk for failure. Appropriate conclusions and recommendations based upon the study findings were made. Implications for secondary school counselors and those counselors working with at-risk students were delineated. / Ed. D.