For those of you who teach high school math or elementary statistics, here's the best example ever of the importance of distinguishing "mean" from "median."
Last November, EPA proposed to make more stringent air pollution standards for particulate matter (PM). EPA claimed its proposal would save 20,000 lives annually.
On April 2, 1997, EPA issued a press release lowering its estimate of reduced PM mortality from 20,000 to 15,000 annually. It turns out EPA erroneously relied on a chart in one study that was labeled as an arithmetic "median," when it actually represented a "mean."
Although EPA did not identify the erroneous study, EPA's PM mortality estimate is based on an epidemiologic study published in 1995 by C. Arden Pope (Brigham Young Uuiversity) and others in the American Lung Association's rag Journal of Respiratory and Critical Care Medicine. This study reported a 17 percent increase in mortality for those exposed to airborne particulates in the most polluted city as compared to the least polluted city. (Click here to read my Wall Street Journal op-ed on this study.)
For purposes of EPA's proposal, EPA valued each human life saved at $5 million. That means for the extra 5,000 annual deaths erroneously calculated by EPA, benefits of the proposed PM rule were overestimated by $25 BILLION ANNUALLY! (5,000 deaths per year times $5 million per life).
Now how much confidence can we have in the rest of the "science" behind EPA's air pollution proposal?
Material presented on this home page constitutes opinion of the author.
Copyright © 1997 Steven J. Milloy. All rights reserved. Site developed and hosted by WestLake Solutions, Inc.