THE PURPOSE OF THE RESEARCH WAS TO CONDUCT A PRELIMINARY ANALYSIS OF THE DENNY, RUSCH, IVES CLASSROOM OBSERVATION SCHEDULE DESIGNED TO IDENTIFY THE COMPLEX OF TEACHER AND PUPIL BEHAVIORS WHICH CONTRIBUTE TO PUPIL GAIN IN CREATIVITY. INTERRELATIONS OF PUPIL GAINS IN CREATIVITY WITH TEACHER PUPIL BEHAVIOR VARIABLES WERE ANALYZED. THE SAMPLE WAS LIMITED TO 30 SIXTH-GRADE CLASSROOMS WITHIN A 90-MILE RADIUS IN A MIDWESTERN STATE. MEAN IQ RANGED FROM 92 TO 117. MEAN SOCIOECONOMIC RATINGS RANGED FROM 3.67 TO 5.57 ON A SEVEN-POINT SCALE. MOST TEACHERS HELD B.S. DEGREES AND HAD TAUGHT SIXTH GRADE AN AVERAGE OF 7.33 YEARS. CREATIVITY WAS MEASURED USING A BATTERY OF TESTS PREPARED BY J.P. GUILFORD. PRETESTS AND POST-TESTS WERE ADMINISTERED. TEST VALIDITY WAS CHECKED BY CORRELATION OF EACH SUBTEST WITH A PEER NOMINATION AND INTEREST INVENTORY. OBSERVATIONS WERE MADE BY A TEAM OF THREE OBSERVERS WHO WERE TRAINED FOR THE OBSERVATION VISITS. THREE RANDOMLY SCHEDULED VISITS WERE MADE TO EACH OF 30 CLASSROOMS. ANALYSIS OF OBSERVATIONS AND ANALYSIS OF VARIANCE WERE USED. CONCLUSIONS INDICATED--(1) SCHEDULES SEEMED TO DIFFERENTIATE BETWEEN CLASSROOMS, (2) OBSERVATION SCHEDULE ITEMS SEEMED TO MEASURE DIFFERENT ASPECTS OF TEACHER-PUPIL BEHAVIOR, (3) BEHAVIORS SEEMED TO DIFFER FROM SITUATION TO SITUATION IN A SAMPLE OF THREE VISITS, (4) THE SCHEDULE SEEMED TO BE OBJECTIVE, (5) OTHER SOURCES OF VARIANCE WERE NOT ACCOUNTED FOR IN THE ANALYSIS OF THE SCHEDULE, (6) IDENTIFIED CHANGES IN THE OBSERVATION SCHEDULE MAY IMPROVE THE RELIABILITY, (7) RELIABILITY AND OBJECTIVITY COEFFICIENTS WERE EQUAL TO OR GREATER THAN PREVIOUSLY REPORTED SCHEDULES, AND (8) THE SCHEDULE WAS VALID FOR DIFFERENTIATING BETWEEN HIGH OR LOW MEAN-GAIN CLASSROOMS ON FLEXIBILITY AND REDEFINITION. (HB)