The long-term stability of students' evaluations of instructional effectiveness was investigated. The focus was on the extent and nature of realtionships between end-of-term and retrospective ratings. Ratings given by alumni one year after program completion were compared with ratings given by these individuals in 85 undergraduate and graduate business administration classes, subsequent to course completion. Large and statistically significant stability coefficients were obtained between end-of-term and retrospective ratings, based on either individual or class average assessments. Mean differences were statistically significant for all but three of the 11 comparison items, but only one of these differences represented a shift of more than half a rating category. The results provide strong support for the stability of students' evaluations. (Author/SW)