Teacher performance data was used in this investigation to determine whether the stability of variance components and generalizability coefficients could be obtained by increasing sample sizes. A section of the Teacher Performance Assessment Instruments (TPAI) was utilized to assess 2 competencies for each of 750 teachers Each competency, defined by five performance indicators rated on a five point scale, was evaluated twice. Three sets of analyses were performed for each competency: the number of teachers was increased in increments of 50-700; 5 replications based on data from 50 teachers and; 5 replications based on data from 150 teachers. Variance components and frequency of negative variance components were estimated for each analysis and generalizability coefficients calculated for each teaching competency. Results indicated that stability increased as the sample size increased. A minimum of 150 teachers appears to be required in order to obtain stable variance components and generalizability coefficients and to maintain the number of negative variance components. (AEF)