Nominal & Ordinal Trend Analysis

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Nominal & Ordinal Trend Analysis

Anton-24
Dear Listers,

I have a data set that contains the responses to questions that have been
asked each year, over a period of seven years, of the *same* respondents.
Some questions are of the Yes/No variety, while others are ordinal-level
(e.g., Always, Sometimes, Never).  The researcher is less interested in
specific comparisons (for example, whether time 3 is statistically different
than time 6), than knowing what the overall trend might be.

So, I guess my question is this: In addition to graphing the data, is there
a statistical test(s) or method(s) in SPSS that I can use to describe the
overall trend in a series of Yes/No and ordinal-level data?

Any help is much appreciated.

Anton
Reply | Threaded
Open this post in threaded view
|

Re: Nominal & Ordinal Trend Analysis

Kooij, A.J. van der
...is there a statistical test(s) or method(s) in SPSS that I can use to describe the
overall trend in a series of Yes/No and ordinal-level data?

Yes, if you have SPSS Categories, this can be done with CATPCA. See the Categories manual (can be found on the net, a.o. at
http://www.sscnet.ucla.edu/labs/SPSS13/SPSSCategories13.0.pdf)
chapter 9, the "Symptomatology of Eating Disorders" example; measurements at 4 time points, the trend can be plotted as is illustrated and explainded in the "Examining the Structure of the Course of Illness" section (page 198).
 
Regards,
Anita van der Kooij
Data Theory Group
Leiden University

________________________________

From: SPSSX(r) Discussion on behalf of Anton
Sent: Tue 14/11/2006 01:50
To: [hidden email]
Subject: Nominal & Ordinal Trend Analysis



Dear Listers,

I have a data set that contains the responses to questions that have been
asked each year, over a period of seven years, of the *same* respondents.
Some questions are of the Yes/No variety, while others are ordinal-level
(e.g., Always, Sometimes, Never).  The researcher is less interested in
specific comparisons (for example, whether time 3 is statistically different
than time 6), than knowing what the overall trend might be.

So, I guess my question is this: In addition to graphing the data, is there
a statistical test(s) or method(s) in SPSS that I can use to describe the
overall trend in a series of Yes/No and ordinal-level data?

Any help is much appreciated.

Anton



**********************************************************************
This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify
the system manager.
**********************************************************************