The last five to ten years has seen a renewed interest in the stability of teacher behavior and effectiveness. Data on teacher performance and teacher effectiveness are being used increasingly as the basis for decisions about continued employment, tenure and promotion, and financial bonuses. The purpose of this study is to explore the stability of both teacher performance and effectiveness by determining the extent to which performances and effectiveness of individual teachers fluctuate over time. The sample consisted of 132 teachers for whom both observational and state standardized test data were available for five consecutive years. Neither teacher performance nor effectiveness were highly stable over multiple years of the study. The observed relationship between teacher performance and teacher effectiveness was reasonably stable over time, but the magnitude of the relationship was quite small. Teacher performance was also likely to be inflated in low performing schools. We also discuss when different observed patterns may be acceptable based on the purpose for which the data are used.