Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-l] A challenge.

ItemAnalysis - yes - that was the programs name.
I may still have a copy of the code around somewhere even tho it was FOUO.

Pearson Product Coefficient
Stratified Alpha (different Subject Matter Areas within each MOS test)

Large Data Set application.

We would have had to get a dispensation from the Pentagon to use Fortran but COBOL was (is?) WAY more efficient when the application is I/O bound.
Fancy statistical formulas really aren't that hard to write in COBOL anyway.
We had to use Assembler on the older machine for lack of copious core memory (REALLY core memory - one magnetic torus of each bit of data ! ! ! )

At 1:34 PM -0500 11/28/09, Joseph Bellina wrote:
As I recall the process for assessing the value of questions and whether there is a positive correlation between performance on the overall test and on any given question is called items analysls. See, for example, http://www.statsoft.com/textbook/streliab.html, that I found by googling items analysis.

joe