This study identified, refined, and tested three sets of student feedback items comprising separate questionnaires for use as one component in a total teacher performance evaluation system. Essentially, this study affirmed that student raters, using valid feedback items, can reliably discriminate between teacher performance when differences do, in fact, exist;The analysis of the data for this study supports the following observations and findings. First, all but four of the original items developed by Judkins (1987) and those subsequently refined by the Iowa State University's School Improvement Model (SIM) continued to demonstrate item discrimination power when used with a non-voluntary sample of teachers and student raters. Second, local teachers and administrators were capable of developing discriminating items for use in student feedback rating instruments for teacher performance. Third, grade level, subject area, core versus elective, career ladder status, and teacher gender were extraneous variables that did have an effect on students' mean score ratings of teachers. Finally, factor analysis can be used to identify factors which can be used to represent relationships among sets of discriminating items;One important element of this study was the non-voluntary sample, consisting of the total school system population of Cave Creek Unified School District No. 93. Cave Creek is located just north of Scottsdale, Arizona. In May of 1990, all grades 6-12 Cave Creek Public School teachers participated in the study by administering the questionnaires to their classes and were consequently rated by their students;All previous studies regarding students' ratings of teacher performance have relied upon volunteers as subjects. The Menne and Tolsma methodology was used to identify those items which possessed discriminating power. The findings of this study support the role of student raters, using valid feedback items, as one important source of feedback for assessing and improving teacher performance.