BAYESIAN COMMUNICATION OF RESEARCH RESULTS OVER THE WORLD-WIDE-WEB

The World Wide Web provides a unique opportunity to reconsider how the results of scientific studies can, best be presented to clinicians. F or decades statisticians, philosophers, medical investigators, and oth ers interested in data analysis have assumed that the Bayesian paradig m is the proper approach for reporting the findings of scientific anal yses for use by client computers and readers. At the heart of that app roach is the inclusion of the reader's preexisting knowledge and belie fs. Yet, to date, the methods for inclusion have been too complicated for non-statisticians to use. We believe that the World Wide Web provi des an ideal environment for putting the Bayesian paradigm into practi ce: the author publishes the data from the server side, the reader use s the client to represent her or his prior belief, and a downloaded pr ogram (a Java applet) combines the two. This article describes a proto type implementation for two-arm clinical trials with normally distribu ted outcomes.