||Thirty undergraduate students reporting high levels of test anxiety were randomly assigned to one of three treatment conditions. Treatments included rational-emotive therapy (RET), which emphasized making S's aware of both their own anxiety-producing self-verbalizations and ways they might counter such verbalizations, relaxation placebo (RP), which emphasized the mastery of skills in progressive deep muscle relaxation, and a no-treatment (NT) condition. Seven dependent variables including performance and self-report measures were administered to all S's before and after treatment. A multivariate analysis of variance showed no significant differences among groups at pretreatment. Posttreatment measures, however, supported the conclusion that RET was significantly more effective in reducing test anxiety than either NT or the RP treatment used.