A Buffered Introductory Statistics Course

Abstract

Abstract An introductory statistics course that emphasized consistency of student requirements and diversity of instructors was taught at the University of New Hampshire. Over a four-year period, 24 instructors from 12 departments taught this course. Student requirements remained essentially the same from term to term or instructor to instructor because exams were prepared by sampling question frames that were largely independent of instructors. Student performance appeared to be buffered against instructor variation through a combination of visible question frames, abundant tutorial help, and exams that could be retaken.