This study examined the concurrent validity of observed teacher behavior as a predictor of teacher philosophy, teacher professional attitude, and teacher ability to forecast first-grade reading achievement. The population consisted of ten first-grade teachers and 260 students from five Farmington, Utah, schools. In early September, before any reading tests had been analyzed, each teacher predicted the expected May reading level for each student. Also in September the Education Scale VII, measuring the degree of traditionalism or progressivism, and the Purdue Teacher Opinionnaire, measuring teacher morale or attitude toward the teaching profession, were administered to each teacher. Three 40-minute observations of each teacher were made in December using the Classroom Behavior Scale, a measure of both verbal and nonverbal interaction. Alternate forms of the MacGinitie Reading Tests were administered to the students in February and late April. No significant correlations were found between the negativeness or positiveness of teacher behavior as measured by the Classroom Behavior Scale and by each of the other three measures of teacher behavior. Several recommendations on methodology and instruments used were made. (Author/TO)