Dear SPSS-L List members,

I am currently working on a Visual Basic (VB)/SPSS Application in which I am creating an Item Analysis package. Some of the things this package will do is create a frequency table of responses for each items, number right and wrong for each item, and an item-total correlation.

Is there any way to pull "cells" from SPSS output and placed them in a SPSS data file (.sav) so that a Report procedure could be performed. The ultimate goal is to condense the output. For example SPSS frequency and correlation tables look something like this:

Frequency Percent Valid Percent Cumulative Percent Valid 0 112 80.57 1 8 etc. 2 8 3 3 4 3 Total 134

Missing System 5 Total 139 ----------------------------------------------------------------------- Frequency Percent Valid Percent Cumulative Percent Valid Right 112 80.57 Wrong 29 etc.. Total 134

Missing System 5 Total 139 ----------------------------------------------------------------------- Correlations CI1 SCORE CI1 Pearson Correlation 1 .044 Sig. (2-tailed) . .619 N 131 131

I would like to condense the output to look like this:

ITEM RIGHT WRONG BLANK DIFF PT.BIS 0 1 2 3 4 1 112 29 5 .8057 0.044 112 0 8 8 3 2 etc 3 4 etc., up to potentially 200 items

Where the PT.BIS is the Pearson Correlation, 0 1 2 3 4 is the frequency from the first frequency table, DIFF is the percent in the frequency tables.

I'm thinking that if these values could be "pulled out" and into a data set one could use REPORT or TABLES to create this type of format. (Or perhaps there could be another way??).

Any ideas or help on how to achieve this would be greatly appreciated.

Thanks,

Craig