This study was designed to discover which merit pay evaluation and implementation factors were considered important by members of the Florida Teaching Profession - National Education Association (FTP-NEA). Comparisons were made of opinions within various subgroups (district size, position within organization, sex, race, age, years teaching, highest degree earned, assignment, secondary assignment). The sample included the State Board of Directors, the Board of Directors of the United Faculty of Florida CUFF-university personnel), and FTP-NEA members in two-thirds of the local affiliates. All districts with support personnel groups were included. The questionnaire was comprised of merit pay factors which were cited most often in the literature, and factors being considered for Florida's State Master Teacher Program. The instrument included 15 evaluation and 25 implementation factors. Questionnaires were distributed by local affiliate presidents. The UFF board and state FTP-NEA board members were surveyed by mail. The results of 662 surveys were analyzed, using a Chi-square test for each factor for each subgroup within the sample. Respondents felt most strongly that "teaching experience/number of years teaching" and "administrator observations/evaluations" should be used as evaluation criteria in a merit pay plan. They were most opposed to "standardized teacher test scores" and "standardized student test scores," both "by school" and "by teacher." They agreed strongly with several of the implementation factors, including "each teacher should ahve access to his/her own records" and "an evaluation instrument should stress performance on the job in the assigned area." They were strongly against the use of a quota, extra pay for tecahers in shortage or alternative areas, and the involvement of business persons and legislators in planning a merit pay system. There were several Chi-square tests which produced valid, significant differences among various subgroups. Significant differences were evidenct in 26 cases for evaluation factors, and 14 cases for implementation factors. The largest numbers of significantly different opinions were in the categories of "sex" (10 factors) and "assignment" (8 factors). There were no significant differences among respondents with various seconday assignments and one significant difference between black and white responses.