Working papers are in draft form. This working paper is distributed for purposes of comment and discussion only. It may not be reproduced without permission of the copyright holder. Copies of working papers are available from the author.

Statistical design and analysis of experiments has revolutionized science and industry, from agriculture to industrial quality improvement. But the principles of experimental design go far beyond these traditional applications. Researchers in areas as diverse as direct marketing, computer simulation, and Web site design are also using statistical experiments to study and optimize their endeavors. For example, you can optimize Web site traffic by conducting an experiment on different choices for headline texts, product images, or button text. Or you can study customer behavior by designing a direct marketing campaign with different offers and prices. In this presentation we demonstrate specialized features of SAS/QC software that enable you to perform these kinds of experiments. Our aim is not only to show how SAS ® software can enable advances in the particular applications we cover, but also to inspire you to think outside of your own box by looking for new applications of experimental design. Experiments Come Naturally! Experimentation is the most natural of human responses to the task of acquiring information. In nearly every new endeavor, an initial discovery and pioneering phase is followed by a phase of experimentation—trying out various changes in the methodology or technique to see which ones make

www.jtaer.com