The Tennessee Career Ladder Evaluation system, which uses a multiple data source approach to assessment, was developed to identify excellent teachers. It is used to determine whether or not the teacher should receive increasing career benefits. Evaluation data are generated by three evaluators, the teacher candidate, the school principal, three peer teachers, and twenty or more students. Extensive review was used in generating six categories of teacher competency: planning; instructional strategies; evaluation, classroom management; leadership; and effective communication. Items were field tested in different types of data collection: classroom observation; interviews with teachers and principals; questionnaires for students and peer teachers; portfolio rating; a written test; and evaluator judgement. Results from each instrument were analyzed separately and then reanalyzed as part of the set. Intercorrelations were low, confirming that each source measured different viewpoints or aspects of performance. The resulting scores were reduced to a smaller number of scores, and then weighted. It was concluded that the multiple data source system was difficult and expensive to build, relatively inflexible, and complicated to explain. However, it provided a thorough and equitable evaluation, was relatively stable, and was logical. (GDC)