The purpose of this study was to compare the scoring of Teacher Performance Assessment Instruments (TPAI) indicators using discrete descriptors when some are considered "essential" with the scoring of these same indicators, and when no descriptors are considered essential. The two questions addressed in this study were: (1) To what extent does the use of essential descriptors affect the overall "pass-fail rate" for each competency? and (2) To what extent does the use of essential descriptors affect the dependability of the certification decision? Data was used from twenty-six teachers who volunteered to prepare a lesson plan portfolio and who allowed observers to come into their classes. Results of observations were scored using two different methods: (1) using essential descriptors according to the criteria anticipated when the instruments are used in certification; and (2) treating descriptors equally, with no essential designation. Analyses were conducted on the transformed data obtained using the essential and non-essential scoring systems. Results showed that the essential descriptors did not detract from the reliability of the measures; in fact, it was enhanced by them. Although the study requires replication with a larger number of teachers and more realistic conditions, the results were viewed as supportive of the essential descriptor scoring method. (LMO)