Few would disagree that grade inflation is a major problem at colleges and universities across the country. In this article, the author argues that a significant contributing factor-perhaps the single most important factor has been the institution of the now almost universal practice of administering anonymous, multiple-choice student evaluations of instruction at the end of most courses. He argues that the process has been examined incorrectly from a short-term perspective--whether there is a correlation between the grades an instructor gives and instructor evaluations in a particular course. The problem must be addressed from a much broader academic cultural perspective.