|
|
Article: Two statistical paradoxes in the interpretation of group differences: illustrated with medical school admission and licensing data.
- Article from:
- The American Statistician
- Article date:
- May 1, 2004
- Author:
CopyrightCOPYRIGHT 2004 American Statistical Association. This material is published under license from the publisher through the Gale Group, Farmington Hills, Michigan. All inquiries regarding rights should be directed to the Gale Group. (Hide copyright information)
|
To count is modern practice, the ancient method was to guess
--Samuel Johnson
Evidence may not buy happiness, but it sure does steady the nerves
--paraphrasing Satchel Paige's comment about money
1. INTRODUCTION
Modern policy decisions involving group differences are both based on, and evaluated by, empirical evidence. But the understanding and interpretation of the data that comprise such evidence must be done carefully, for many traps await the unwary. This essay explores two statistical paradoxes that can potentially mislead us and illustrates these paradoxes with data used in the admission of candidates to medical school, and one measure of the ...