Teaching

We intended this Handbook to help instructors who teach statistics and research methods, either as separate courses or in others, such as introductory psychology and advanced content courses. We organized the 90 articles into two main sections, Statistics and Research Methods, and each major section contains subgroups of papers on common themes. Collectively, the articles include a stunning amount of information. Among other topics, articles in the first section cover (a) how to reduce students' anxiety about statistics, (b) general and specific strategies for teaching statistics, (c) how to illustrate some statistical concepts and techniques, and (d) several ways to generate data sets for student use. Among other topics, articles in the second section cover (a) ethical issues, (b) proposals for designing and conducting a research methods course, (c) techniques for enlivening and improving students' literature reviews, (d) general and specific strategies for teaching a variety of methodological concepts and procedures, (e) use of computers, (f) suggestions for successfully involving students in substantive research and encouraging formal presentations of their results, and (g) recommendations for making theses and dissertations more productive and pleasant. Equally important, many of the articles in both sections are rich sources of ideas for further research.

Are SAT scores useful predictors of success in college? I led a group of mathematics majors at Oberlin College in an exploration of this question as the core of a one-credit course, entitled Data Analysis, last year. This course, MATH 337, is an adjunct to the standard, junior-level, two semester sequence in probability and mathematical statistics that we offer each year at Oberlin. Unlike most statistics courses for mathematics majors, the Data Analysis course allows - indeed, it forces - students to "get their hands dirty" exploring real data and trying to answer real questions. Each year I select a set of data, such as the SAT data, to serve as the central focus of the course. I believe that it is imperative that student learn something of how statistical theory is applied in practice and I try to show this side of statistics in the courses I teach at all levels. However, it is particularly difficult to cover much material on applied statistics while at the same time covering the many important topics in the mathematical statistics course - probability, random variables, functions of random variables, expectation, the central limit theorem, estimators, confidence intervals, hypothesis testing, and others - that are fundamental to the discipline and are an essential foundation for advanced (graduate) training in statistics. The Data Analysis course provides a workable solution to this problem.

Brief descriptions of several model courses which have been developed by participants in a series of Statistics in the Liberal Arts Workshops (SLAW).

In South Australia mathematics students typically have had considerable exposure to calculus but little to probability on entering university. The amount of probability in school syllabuses has decreased, and our first year university students often associate the subject with fatuous but intricate examples. Some undertake a first year statistics subject, but many will come to a first course in applied probability/stochastic processes in second year applied mathematics without that background. They are ill at lease with any second year applied mathematics courses not based on calculus (forgetting the effort with which the comfort with calculus was won!) and will often demand evidence of meaningful applicability at an early stage. This creates some challenge and necessitates an examination of one's teaching philosophy.