Today citizens are exposed to numerous and complex issues in public policy. We try to develop in our students the ability to decipher these important substantive issues, to improve their ability to distinguish between cause and effect on the one hand and coincidental relationships on the other hand. I propose to create 1-2 pilot modules that will simultaneously enhance the study of public policy problems and improve students' quantitative reasoning skills. These modules will be based upon the quantitative analysis of substantive public policy problems. In addition to creating data sets in specific policy areas, I will also create an instructional supplement with guided computer exercises. Therefore, the modules are geared towards deepening students' substantive understanding of complex policy issues and improving their understanding and use of quantitative data analytic techniques.

A rich source of data are available to members of the ACS through our membership in the Inter-University Consortium for Political and Social Research (ICPSR) at the University of Michigan. These studies permit in-depth investigations of public policy problems such as crime, welfare, poverty, racism and intolerance, violence, and many others. For instance, the consortium gives us access to data files from the United States Department of Justice regarding crime and the effects of sentencing sanctions. There are several studies conducted by the United States Bureau of the Census investigating the sources and intractable nature of poverty. And from the Project on Race and Violence in American Society there are several data files regarding race, (in)tolerance, and violence.

I propose to use these data sets to create instructional packages that can be used in public policy or analysis classes as well as classes in American politics. I will create new data sets that are subsets containing variables appropriate for undergraduate instruction. Next, I will create an instructional guide with data exercises designed to facilitate the interactive exploration of the data by students. Policy problems such as crime, poverty, and racial discrimination are substantively important and complex. By requiring students to conduct their own investigations of questions they find interesting, they will in turn make a new and vital contribution to the traditional classroom. Indeed, the classroom will be converted into a collaborative research environment between the professor and the students.

This project will also rely upon exploratory data analytic methods. Exploratory data analysis techniques have added a new dimension to the way people approach data. Students can easily uncover features concealed among masses of numbers. Computers are valuable aids to data analysis. Indeed, four major ingredients of exploratory data analysis stand out and will be emphasized in these modules:

• Displays visually reveal the behavior of the data and the structure of the analyses;
• Residuals focus attention on what remains of the data after some analysis;
• Re-expressions, by means of simple mathematical functions such as the logarithm and the square root, help to simplify behavior and clarify analyses, and;
• Resistance ensures that a few extraordinary data values do not unduly influence the results of an analysis (John Tukey, Exploratory Data Analysis).
  

There are numerous potential benefits for the instructional experience. By integrating data analysis with substantive policy investigations the classroom environment is changed. These benefits can be summarized by the following goals.

Development of the public policy modules will benefit several courses. It can be integrated into my courses Public Policy Analysis and American Public Policy-both are upper-level seminars. The modules also could be used in similar courses offered at other ACS institutions. I plan to introduce the first pilot module in Public Policy Analysis during the spring, 2000, semester.

The modules can be easily shared among other interested ACS members. The data modules and instructional supplements can be distributed via disk. Alternatively, the materials can be browsed and/or downloaded from a web page that I will create.

The quantitative aspects of the course will be assessed by requiring students to take a quantitative skills achievement test developed by Rollins' faculty. The course also will be evaluated by developing a survey to be completed by students at the end of the semester.