🚀 go-pugleaf

I'm posting this for a friend.  I'll try to forward any followups I
receive; in any case, please email replies to kem@pshrink.chi.il.us,
not to me.

Bill Higgins                            Internet: HIGGINS@FNAL.FNAL.GOV 
Fermi National Accelerator Laboratory   Bitnet:   HIGGINS@FNAL.BITNET

=========
Hi!

My name is Kelley Mathews.  I am in the process of doing background
research for my dissertation.  The current focus of my dissertation is
exploring Bayesian-influenced strategies for the construction and
validation of predictive tests.  I am currently working with two data
bases-- one database relates to students who did or did not drop a
college course; the other relates to individuals who were or were not
rearrested for DUI (Driving Under the Influence of alcohol).  I am
looking at the impact of various construction and validation
techniques on prediction and discrimination.

Given that brief intro-- I am interested in material that might be
relevant in general.

In addition I am seeking some specific information:

1.  Journal articles etc that contain commentary, analysis or criticism
on the current state of the art in predicting human behavior using
psychological tests.  I have reviews by Ghiselli (1973) and Boehm
(1982), but haven't been successful in finding more recent work.

2. Articles relating to Bayesian test construction and validation
techniques and the effectiveness of such approaches.

3.  Articles relation to other approaches to test construction and
validation.

4.  Information on Resampling.

5.  A good technical book/journal that provides formulas for item
analysis-- item discrimination indexes and inter item correlation.  The items I
am analyzing are dichotomous and discriminate two groups.

6.  What is good software for running this type of analysis?

My e-mail address is kem@pshrink.chi.il.us.

My snailmail address is Kelley Mathews
                        741 Whitesail
                        Schaumburg, Il. 60194

My home phone is: 708-884-6757.

Thanks 
Kelley Mathews