Forty-five test-anxious students were randomly assigned to one of four treatment conditions: (a) relaxation/cognitive therapy, (b) study-skills training, (c) a combination of relaxation/ cognitive therapy and study-skills training, or (d) no treatment. Pretreatment and posttreatment measures were collected on self-reported state anxiety and classroom examination performance. The relaxation/cognitive therapy was found to be effective in reducing anxiety but failed to improve classroom test scores. Study-skills training had no significant effect on either measure. The combined therapy both reduced anxiety and improved performance relative to the no-treatment control condition and was significantly more effective than was either treatment alone.This study was based on a master's thesis completed by the first author (Dendato, 1985) under the supervision of the second.We would like to thank Russell Hurlburt, Frederick Kirschner, and Diane Turnbough, who served on the thesis committee, and Lori Kolbet for her helpful comments and suggestions.