RE: Final Exam

[John Kulig]

Expressed this way, these are odds, so .9/.015 translates into a 60 to 1 ratio (.9 = 60* .015) favoring the top half (innocence). To switch to a probability format, it’s 60/(60 + 1) or .98. This “odds” ratio version of Bayes I get directly from Abelson’s Statistics as Principled Argument. I’m not a Bayesian, either, but am intrigued with it because it gives p (hypotheses / data) – what we ultimately want, rather than the usual p (data / hypothesis), frequently misunderstood as the p of the null hypothesis .To get into the Bayes equations you need to make hunches as to probabilities. To purists, those hunches are not true probabilities. But as researchers, these hunches, as well as prior literature, theory predictions, and a host of other factors, SHOULD influence our decision about which hypotheses are true, rather than pinning the decision of an entire study on one little p value.   ============================================ John W. Kulig Professor of Psychology Plymouth State College Plymouth NH 03264 ============================================ -----Original Message----- From: Michael Scoles [mailto:[EMAIL PROTECTED] Sent: Thursday, December 02, 2004 12:24 PM To: Teaching in the Psychological Sciences Subject: RE: Final Exam   Bayesian methods are not my strong suit, but wouldn't "p innocence" be 60/121 = .50?   Michael T. Scoles, Ph.D. Interim Chair, Dept. Psychology & Counseling University of Central Arkansas Conway, AR 72035 >>> [EMAIL PROTECTED] 12/2/2004 11:05:45 AM >>> So, posterior odds = (1/1) * (.9/.015) = or 60 to 1 in favor of innocence, or p innocence = 60/61 = .98. --- You are currently subscribed to tips as: [EMAIL PROTECTED] To unsubscribe send a blank email to [EMAIL PROTECTED] --- You are currently subscribed to tips as: archive@jab.org To unsubscribe send a blank email to [EMAIL PROTECTED]