Re: Beginners please

[Kromkowski@aol.com]

In a message dated 11/01/00 10:13:47 AM Pacific Standard Time, 
den.list-d-request@deming.ces.clemson.edu writes:

>   Also, why is there never any discussion of chi-squares?  For example, if 
I 
> am concerned about probabilities of success in certain career fields based 
on 
> levels of education and math/reading test scores, aptitude test scores, 
> interest inventories, etc. that a client achieves on a standardized test, 
> wouldn't a chi-square distribution give me the best information to help 
> determine the best training programs for my clients?  What about adding 
socio-
> economic status, race, gender, religion?  If I were to study six-sigma, 
would 
> it help me with these questions, i.e. provide me with a better indicatior 
of 
> success probabilites than a chi-square would?  If so, what additional data, 
> if any, would I need besides current level of education, test score  and 
> employment follow-up data?  

1.  Well, we could start with Deming's remark at 272 in OTC:  "Incidentally, 
chi-square and tests of significance, taught in some statistical courses, 
have no application hear or anywhere."

2.  You might want to talk with a Bayesian with respect to the other points 
mentioned.  We have a resident-Bayesian who, nonetheless, is actually a very 
bright person. Here is where the Bayesians have to come to grips with the 
come to grips with the Pygmalion factor.

John David Kromkowski
Kromkowski@aol.comTelephone:  (410) 377-6248Facsimile:  (410) 372-0624