Case File Structure

[Zaid Ansari]

I have data from a survey with 27 respondents and 101 questions and 5
response options (values). These data are arranged that each case has 101
responses(then it repeats for case 2 with 101...so on) with a total n=2727
(101X27). I need to develop a frequency distribution, and compare some
pre/post frequencies. What's the best way to deal with this?

Much thanks in advance.

Zaid A. Ansari, MS ABD
Director of Assessment and Institutional Research
Morehouse College
[phone]-Office
[phone]-Fax
[hidden email]

[Max Jasper]

For your 5-level cat variables you might need to do:

*       Profile Analysis: set up custom linear hypotheses
*       Testing Parallelism: are the population mean profiles of....and
....parallel to each other
*       Testing for Equality of Profiles: assuming parallelism, are the
profiles of....and...equal?
*       Testing for Equality of Answer Means: assuming parallelism, are the
answer means equal?

by using:
/LMATRIX=....
/MMATRIX=...

for a GLM Multivariate procedure.



|-----Original Message-----
|From: SPSSX(r) Discussion [mailto:[hidden email]] On
|Behalf Of Zaid Ansari
|Sent: Monday, May 07, 2007 3:40 PM
|To: [hidden email]
|Subject: Case File Structure
|
|
|I have data from a survey with 27 respondents and 101
|questions and 5 response options (values). These data are
|arranged that each case has 101 responses(then it repeats for
|case 2 with 101...so on) with a total n=2727 (101X27). I need
|to develop a frequency distribution, and compare some pre/post
|frequencies. What's the best way to deal with this?
|
|Much thanks in advance.
|
|Zaid A. Ansari, MS ABD
|Director of Assessment and Institutional Research
|Morehouse College
|[phone]-Office
|[phone]-Fax
|[hidden email]

[Maguin, Eugene]

Zaid,

I may be going off in the wrong direction but it seems to me that you can't
figure out how to get frequencies on your 101 variables. I'm assuming that
your spss datafile consists of 27*101 rows and maybe two columns (id and
response) and not 27 rows and about 102 columns (id and 101 responses). If
my assumption is true, then you need to convert your long format datafile
into a wide format datafile via the Casestovars command before doing your
frequencies.

By the way, you mention pre-post comparisons. Where does the posttest
assessment data fit into what you described?

Gene Maguin