Search All of the Math Forum:
Views expressed in these public forums are not endorsed by
Drexel University or The Math Forum.



RE: Level of Significance
Posted:
Jan 6, 2000 6:28 AM


I think the pyschology here is "no tails" rather than "5% probability" For instance, I think the number of people who would doubt the run HHTHHHTHHH would be much smaller. We are more prone to believe in total fraud (a two headed coin) than in marginal fraud (a coin biased to 75% heads). Something about the black/white nature of our judgement systems.
Pat Ballew, Misawa, Jp
"Statistics means never having to say you're certain."
Math Words & Other Words http://www.geocities.com/paris/rue/1861/etyindex.html
The Mathboy's page http://www.geocities.com/paris/rue/1861
Original Message From: Chuck Tiberio <tiberio@tiac.net> at eduinternet Sent: Wednesday, January 05, 2000 1:39 AM To: apstatl@etc.bc.ca at EDUINTERNET Subject: Level of Significance
Maybe one of the guru's out there can provide more details about this story...... As I heard it, Fred Mosteller from Harvard used to bring a two headed coin to his prob/stat class. He would take it out, flip it and record the results on the board. After four consecutive heads, about half the class would raise their hands to question the fairness of the coin. After the fifth head in a row, the other half of the class would join them. Pr(4 heads in a row given a fair coin) = .0625 and Pr(5 heads in a row given a fair coin) = .03125. So apparently most people start "feeling uncomfortable" somewhere between .0625 and .03125. So .05 was chosen as a convenient compromise. I do, however, totally agree with everything Chris said. Statistics means never having to live or die by .05!
Chuck ======================================================================= The Advanced Placement Statistics List To UNSUBSCRIBE send a message to majordomo@etc.bc.ca containing: unsubscribe apstatl <email address used to subscribe> Discussion archives are at http://forum.swarthmore.edu/epigone/apstatl Problems with the list or your subscription? mailto://jswift@sd70.bc.ca =======================================================================
======================================================================= The Advanced Placement Statistics List To UNSUBSCRIBE send a message to majordomo@etc.bc.ca containing: unsubscribe apstatl <email address used to subscribe> Discussion archives are at http://forum.swarthmore.edu/epigone/apstatl Problems with the list or your subscription? mailto://jswift@sd70.bc.ca =======================================================================



