Bruce Trumbo and Joan Sieber

The principal investigators will create materials for a computer-based elementary statistics course, using data sets from recent research in the biological, social and other sciences. These materials will improve statistical literacy and encourage students to view statistics as an interesting and important means of answering scientific questions. Students will learn modern methods of exploratory data analysis (including those based upon computer graphics) and statistical inference.

The materials will address the practical and intellectual problems connected with the classroom use of computers and real data sets in the elementary statistics curriculum. Specifically, they will include:

(1) A collection of carefully documented data sets from recent and important scientific research ready for computer use,

(2) An instructor's manual, and

(3) Student workbooks.

Many teachers of elementary statistical courses have neither the time nor the specialized computer expertise to prepare such materials. Throughout the project, faculty and students from diverse institutions will test and critique these materials.

The following additional detail may be helpful:

The data sets will be presented in several forms, including (a) flat ASCII format readable by almost all statistical software, (b) Minitab worksheets with variable labels, (c) other formats selected to meet expressed needs of those who preview our materials, such as SPSS or SAS files.

The instructor's manual will contain several sections, including at least the following: First, a section assuming no previous instructional computer use, which will discuss some of the difficulties involved in teaching statistics as a lab science. Second, suggested statistical concepts, beyond those mentioned in the student workbooks, which may be illustrated with various of the data sets. Third, the fullest possible background will be given for each data set, including the original journal article or report that first used the data, any known additional studies that have used the data, any known previous uses of the data for pedagogical purposes, and any suggestions for obtaining future data sets of similar type.

For selected data sets, the student workbook will first lead the student through an exploration and analysis of the data using Minitab. Then a collection of guided explorations will encourage students to look at the data set more deeply, performing similar analyses on different variables, working with a different subsample of the data, using different statistical techniques, or branching off in a direction quite unrelated to the presentation in the workbook. (Parallel workbooks keyed to other statistical packages may also be prepared.) Other data sets will be presented more briefly, with an introduction to the purpose of the original researcher and suggestions for explorations and inferential procedures which imitate or augment what the original researcher had in mind.

Return to Cobb Paper | Return to Table of Contents | Return to the JSE Home Page