This pilot study was designed to appraise the effectiveness of two methods for facilitating learning in the affective domain in an elementary school setting. The two methods evaluated were: (1) teacher reinforcement of verbalization of feelings; and (2) feelings classes. Pre- and post-treatment measures on the two groups of randomly assigned third graders included: (1) observation of childrens' behavior in class; (2) childrens' self-social constructs; and (3) sociometric choices. Findings indicated that the two groups differed significantly in several categories of pupil behavior in the classroom. The feelings classes seemed to promote an open class atmosphere while the other method lent support to a teacher centered class. There was no significant difference between the two groups for the childrens' Self-Social Construct measures and sociometric choices. The importance of the study lies not only with the reported findings, but also with the feasibility of the two basically developmental approaches in classroom programs. (TL)