Date:         Tue, 28 Nov 2006 20:08:15 -0500
Reply-To:     Statisticsdoc <statisticsdoc@cox.net>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Statisticsdoc <statisticsdoc@cox.net>
Subject:      Re: Composite Chi square
In-Reply-To:  <00f501c712f2$eef97560$0400000a@userdyy5ufajt7>
Content-Type: text/plain; charset="windows-1256"

www.statisticsdoc.com
Stephen Brand

If I understand the question correctly, you want to know whether the
distribution of responses differs between questions within the same section.
I am not sure that this can be examined using the chi-squared goodness of
fit statistic, since the observations on two items are not independent.
However, you might gain some valuable information by running a repeated
measures analysis in which question is a within-subject factor.

Sincerely,

Stephen Brand, Ph.D.

For personalized and professional consultation in statistics and research
design, visit
www.statisticsdoc.com


-----Original Message-----
From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU]On Behalf Of
matara
Sent: Tuesday, November 28, 2006 8:42 AM
To: SPSSX-L@LISTSERV.UGA.EDU
Subject: Composite Chi square


Hi all

I have a questionnaire that consists of 4 sections. There are 8 questions in
each section. There are 5 responses to each question ranging from strongly
agree to strongly disagree. For each question, I performed the goodness of
fit chi square to find out if these are differences between the responses. I
want to combine the chi square results that I got for each question to get a
composite chi square for each section of my questionnaire. How can I do
that?

Regards

Matar