Data gathered from 42 secondary school speech communication teachers and their students formed the foundation for a study that examined the relationship between: (1) differential use of Behavior Alteration Techniques (BATs) by teachers trained or untrained in communication in instruction and (2) learning of students of varying quality levels. Twenty-two of the teachers had recently completed master's degrees in speech communication with an emphasis in communication in instruction, and 20 teachers had no communication training. Each teacher was provided with instruments to be given to five of their best students, five average students, and five of their worst students. The students were asked to rate on a one-to-five scale how frequently their teacher used statements of varying types to attempt to change student behavior in the classroom. The teachers were asked to complete the instrument three times, indicating how frequently they used each message with good, average, and poor students. Results indicated that immediate reward from behavior, deferred reward from behavior, self-esteem, and teacher feedback contributed to positive affective learning. Punishment from teacher, legitimate-teacher authority, debt, responsibility to class, and peer modeling were negatively associated with affective learning. Results also indicated that appropriate training in communication in instruction may have led to more appropriate choices of BAT usage and increased student learning. (The research upon which this paper is based received the 1984-85 Distinguished Research Award for the Association of Teacher Educators.) (Author/HOD)