A study was undertaken to determine (1) if six student teachers who received instruction (in eight weekly individual or group training and feedback sessions) in the use of Bloom's taxonomy would operate within the classroom at a higher cognitive level than a control group of six who received equal time instruction using a placebo-type treatment; (2) if two groups of trainer observers (administrators who were experienced in the taxonomy and cooperating teachers who were introduced to the taxonomy in much the same manner as their student teachers) could agree between and within themselves as to the cognitive level of questioning occurring within the classroom during the teacher-pupil dialogue as measured on the Teacher-Pupil Question Inventory (TPQI); and (3) if there were a difference in the percentage of above-memory questions asked by teachers who had training in the use of the taxonomy and those who had no training. Each student teacher made a weekly tape recording of a social studies lesson; tapes of the third, fifth, and seventh weeks were rated by both groups of observers. Analysis of the data, using chi square and analyses of variance, showed a significant difference favoring the experimental group of student teachers, and test results found no significant difference in the rating scores of the observers. (Conclusions and recommendations are included. Appendixes contain the TPQI and a summary of the taxonomy, as modified by Sanders in 1966.) (JS)