During the 1984-85 school year, a metropolitan school system investigated the research on effective schools to determine which program, once put into place, would serve the system in the remediation of low academic achievement. The program chosen for implementation was the Teacher Expectations and Student Achievement (TESA) Program. This study examined the relationships between the frequency of use of TESA-prescribed behaviors in the classroom and (a) time lapse since TESA training, (b) who taught the TESA classes, (c) training type, (d) training quality, and (e) principal support. A teacher survey was developed to collect data on the five predictor variables. One hundred percent of the teachers responded. The TESA observation technique was employed to measure the criterion variable--the frequency of use of the TESA-prescribed behaviors in the classroom. Data were collected during the months of May and June in the 1988-89 school year. Correlations, t-tests, and stepwise regression analysis were employed to analyze the data. Who taught the TESA classes, time lapse since taking TESA training, and training type were found not to significantly predict the frequency of use of TESA behaviors in the classroom. However, statistically significant relationships were found between specific TESA behaviors and certain factors within the multi-item predictor variables of training quality and principal support. / Ed. D.