In 1987, a school-university collaborative project, Positive Attitudes in Tennessee Schools (PATS), was established to improve school-learning environments. This paper presents findings of a study that investigated the effect of school participation in PATS on school climate. A secondary focus was to determine which school-climate variables could predict differences between experienced and beginning PATS schools. Data were derived from the administration of the Tennessee School Climate Inventory (TSCI) to teachers at 92 Tennessee schools--51 elementary, 21 middle, and 20 senior high schools. No significant differences were found to exist between experienced and beginning schools. Elementary schools demonstrated more instructional improvement than did middle and senior high schools, regardless of project experience. TSCI dimensions of school climate were poor predictors of program experience level. Despite nonsignificant differences between experienced and beginner schools, overall net gains were observed for all dimensions. Data suggest that 61.5 percent of the schools could be expected to improve school-climate variables within 3 to 4 years and that only 38.5 percent could be expected to improve such variables within one year. A possible explanation for the apparent lack of significant improvement is the stability of the school-climate construct. (LMI)