A “Reverse Order” Elementary Statistics Course

Abstract

In this article we describe a postcalculus statistics course that we have developed at Dalhousie in which the order of the topics differs substantially from that of standard courses. The course is a first introduction to statistics for students with one year of calculus. It serves students with a primary interest in mathematical sciences as well as science students who need statistics in their discipline. The course attempts to provide a broad-based and comprehensive introduction to both the principles and the techniques of statistical reasoning. The student should finish the course able to set up models, test hypotheses or estimate parameters, and interpret the results in the classical statistical setting. Throughout the course, students use the computer to fit the more complicated models and to carry out routine but tedious computations. To reinforce the practical side of statistics, each student is required to design and carry out a statistical experiment. Details of this aspect of the course can be found in Field (1981). The uniqueness of the course stems not from its goals but from the order in which the material is presented. It provides an early entry into statistics within a unified framework. The usual probabilistic models are not introduced until much later. The unity is achieved by an emphasis on formulating a model, fitting the model, and then assessing the goodness of fit of the data to the model. The next section provides a more detailed description of the course material, and the subsequent sections discuss the rationale in choosing the reverse order of topics and some of our experiences with the course.