For evaluation and further development of the First-Year Communication Skills Program, a classroom observational study was conducted to determine the extent to which teachers demonstrated the desired instructional behaviors and to measure pupil response rate. A sample of 10 classes was selected for observation by two observers during a regularly scheduled lesson on blending (word attack) outcomes. Observers were trained in the use of two observation scales developed for the study: one for stimulus-response analysis of instructional transactions between teacher and pupils and one to measure the patterning and dispersion of pupil responses and the extent to which pupils are presented stimuli appropriate to the program outcomes. Results were used to revise training objectives and develop corresponding materials for the 1969-70 installation. The effectiveness of the new training program was then measured, and based on results and instruments, new procedures and revised classroom monitoring instruments will be developed for use by program supervisors in the 1970-71 tryout. Comparison of trained and untrained teachers shows the program effective for pupil response rate, relevance of instructional stimuli to lesson outcomes, pupil praise, and use of prompts. Behavioral analysis of instruction appears to be useful in identifying training requirements and evaluating program effectiveness. (The observation forms and description of instructional procedures taught to teachers are included.) (JS)