Forty-eight speech-anxious students (25 males and 23 females) ranging in age from 17 to 48 were offered a 4-week speech anxiety group. The research participants were randomly assigned to one of three conditions: (a) graduated behavior rehearsal/feedback (GBR), (b) graduated behavior rehearsal/feedback plus systematic desensitization (GBR/SD), or (c) the no-treatment control. Each subject completed a series of tests at pre-and posttreatment, including the short form of the Personal Report of Confidence as a Speaker and the State-Trait Anxiety Inventory. A behavioral measure, the short form of the Behavioral Assessment of Speech Anxiety, was also obtained at pre-and postassessment. Except for the behavioral instrument, the battery of tests was repeated at a 6-week follow-up. The GBR and GBR/SD treatment groups demonstrated significantly greater speech anxiety reduction than the no-treatment controls at postassessment and follow-up assessment. The exception discovered was that the GBR group did not score significantly lower than the controls on the state anxiety measure of the State-Trait Anxiety Inventory. No significant differences were found between the GBR and GBR/SD treatment conditions on any dependent measure.