The Carroll County (Georgia) Competency-Based Teacher Certification Project developed a measure of teacher effectiveness, the 341-item Georgia Assessment of Teacher Effectiveness, based upon five previously published classroom observation systems. Three concerns were drawn from the development of this instrument: (1) the reduction of observational data to meaningful units for later analysis; (2) the scoring of these data to produce reliable indices of teacher behavior, and the provision for meaningful feedback in the form of behavioral profiles; and (3) the use of discriminant analysis to contrast a priori groups of teachers on the behavioral indices. The data were recorded by observers and keypunched to be transcribed to machine-readable form. Flexibility, ease of transcription, and compactness were useful characteristics of this process. The observers coded teacher behavior according to low-inference rating processes. The computer program then provided scores for behaviors which were already determined to be indicators of effective teaching. It provided functions for data aggregation, scoring of keys, extimation of reliability, and provision of profiles of each classroom. Discriminant analysis was used to compare the results for three groups of teachers (metropolitan, small town, and rural) and two groups of student teachers (from urban and rural universities). The two groups of student teachers were most extreme in their differences. The data are appended. (GDC)