This study assesses the effectiveness of a group counseling treatment on the classroom behaviors and on the manifest anxiety levels of elementary school student teachers. The subjects were 44 volunteer student teachers, randomly assigned to three groups: 1) an experimental counseling group, 2) a Hawthorne seminar control group, and 3) a control group. The counseling group received eight weekly group counseling treatments with the Problem Identification Model, which employs psychodramatic techniques. The seminar group met for equivalent amounts of time. Classroom behavior was measured by a posttest only design: during the last week of student teaching, two 20-minute samples of each subject's teaching behavior were obtained using the Flanders Interaction Analysis Categories with Nonverbal Categories (FIAC/NVC). A repeated measures design assessed manifest state anxiety over time: the State-Trait Anxiety Inventory (STAI) was administered to all subjects prior to, in the middle of, and at the end of student teaching. The Confidential Evaluation Form (CEF) was filled out by cooperating teachers. Results showed no significant differences between groups in percentages of restricting classroom behavior or in decrements of state anxiety over time. On the CEF, however, the counseling group was rated significantly higher than the other groups on "innovativeness" and "potential as a teacher." (RT)