On 27 Dec 2000 08:18:11 -0800, [EMAIL PROTECTED] (Jeff Rasmussen) wrote: < ... > Jeff > ' When I'm on such committees I do a rank ordering based on whatever actuarial data is available and know that doing otherwise is just mucking around with error. Most other faculty haruspicate via predictors such as the "number of full professors who wrote letters of reference", "impression of the quality of their undergraduate school" or other voodoo. They are usually indignant when I point out that such variables add nothing above and beyond the actuarial data. Is there a difference between a psychologist and a psychic, I often wonder. ' ---- Perhaps you were being polite. You forgot to mention what they are *known* to add, above and beyond the actuarial data -- biases. By the big ones: sex, race, social class, age, ethnicity. By more subtle ones: wealth, body language, speech accents, shopping habits. And if you don't meet the fellow "first-hand," how will you know if he might not give you the secret hand-shake? -- Rich Ulrich, [EMAIL PROTECTED] http://www.pitt.edu/~wpilib/index.html ================================================================= Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =================================================================
