Nicole M Bouchet <nbouchet@usgs.gov>
08/24/2006 10:09 AM

To: Richard E Hipkins/NBC/OS/DOI@DOI
cc:
Subject: Fw: drug test eligible people for the whole DOI

Richard,

I enjoyed speaking with you earlier. If you do decide to put the Q and A section up on your web site and want to quote me on the gambler's fallacy, this is what I would say:

"A commonly held belief, the gambler's fallacy, often gets in the way of understanding the likelihood of random drawings with replacement. The gambler's fallacy is the simple but common mistake of seeing statistically independent events as if they are dependent. If a drawing is done with replacement so that after each drawing all people are put back into the selection pool, each person has the same likelihood of being drawn as they did in the previous drawing.

A simple way to think about this is to take the example of a coin toss. The probability of heads is the same as the probability of tails, .50. If you get five heads in a row and then conclude beforethe sixth one is drawn, that it has to be a tail, you would be committing the gambler's fallacy. This is assuming that the sixth toss is dependent on the previous five, when in reality, each toss is actually independent of the others." Nicole Bouchet, Ph.D., Statistician, USGS

Nicole Bouchet
==========================================

Nicole M. Bouchet, Ph.D.
Statistician, Policy and Strategic Initiatives

Office of Human Resources
U.S. Geological Survey
12201 Sunrise Valley Drive, MS601A
Reston, Virginia 20192
Phone: (703) 648-7441
Fax: (703) 648-7451
nbouchet@usgs.gov