The extent to which a teacher served as a model for the verbal style of his students was examined with the use of a modified form of the Observation Schedule and Record 4V (OScAR). Four students from each of 24 first grade classes were separated into groups on the basis of frequency of teacher and peer reinforcement. Verbal characteristics of these students were scored using six scales of imitative behavior developed from the OScAR protocols. The data were analyzed using a 2 X 2 factorial design (teacher reward X pupil reward). Students observing high rewarding teachers imitated significantly more than those observing low rewarding teachers on 4 of 6 verbal categories. In three of the analyses differences between frequently and infrequently peer-reinforced students were located, and for three of the analyses there were also reliable teacher reward X pupil reward interactions. The potential of the OScAR for measuring student behavior and constructs such as limitation was demonstrated. (Author/RT)