Although there have been numerous studies of the reliability, validity, and uses of student ratings of instruction, relatively few efforts have focused on the association of the combination of various student, instructor, and course characteristics in relation to such ratings. The purpose of this paper is to examine the relationship between student demographic and academic characteristics, including expected course grade; faculty characteristics such as rank, teaching load, and experience; and student ratings of these teachers. Using the Student Instructional Report (SIR), data were collected from undergraduates in 3600 liberal arts and professional education courses toward the end of each of four semesters in six colleges of a large urban university. The findings seem to suggest that a student's expected grade in a course does account for the largest amount of variance in student ratings when compared with the other variables. However, even when combined, these variables were found to account for comparatively little variability in student ratings. (BW)