The purpose of this article is to provide some evidence on the effectiveness of graduate-student teachers in undergraduate economics courses by reporting our findings in a study of the introductory course at Princeton, which involved both graduate-student and regular faculty instruction. Our introductory course appears particularly well-suited for such a study because the subject matter (including reading assignments in Samuelson and assorted supplementary sources) is the same for all class sections. Graduate students teach some sections and regular faculty are instructors in the others. A primary purpose of the class sections, each of which meets twice a week for 50 minutes, is the mastery of analytical tools; this is supplemented weekly by a single mass lecture which typically is devoted to a problem of public policy directly related to the analytic material for the week. Near the end of the term, all the instructors (graduate students and faculty) jointly prepare a three-hour final examination, consisting of essays and problems, which is administered at one sitting to all the students in the course. Following a meeting with the class instructors, a group of persons not involved in teaching the course then grades the examination with each grader taking one or two questions and marking them across all the examination papers. The organization of the course thus suggests that if we compare the examination scores of those students taught by faculty instructors with those of students in class sections taught by graduate students, we can gain some information concerning the relative teaching effective-