Date:         Wed, 26 Jan 2005 09:03:46 -0500
Reply-To:     Philip_Moore@CARMAX.COM
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Philip_Moore@CARMAX.COM
Subject:      Re: Comparing Similar Surveys (methods question)
Comments: To: michael healy <healym@earthlink.net>
In-Reply-To:  <18883256.1106693016162.JavaMail.root@misspiggy.psp.pas.earthlink.net>
Content-type: text/plain; charset=US-ASCII

Are the categorical answer codes on the surveys randomly rotated?  Are the
ordinal scale answer codes flipped?  If not, you could see if the order
bias on both surveys is similar/identical.  Are there questions on the mall
intercept version that collected unaided responses, versus the web version
where the respondent can see all the answers?

Philip Moore
Market Research Manager
(804) 747-0422 x4831
(804) 935-4549 FAX

The information in this email is extremely confidential.  It is intended
solely for the addressee.  Access to this email by anyone else is
unauthorized.  Please do not copy or disseminate any portion of this email.



             michael healy
             <healym@earthlink
             .net>                                                      To
             Sent by:                  SPSSX-L@LISTSERV.UGA.EDU
             "SPSSX(r)                                                  cc
             Discussion"
             <SPSSX-L@LISTSERV                                     Subject
             .UGA.EDU>                 Comparing Similar Surveys (methods
                                       question)

             01/25/2005 05:43
             PM


             Please respond to
               michael healy
             <healym@earthlink
                   .net>






Hello,  This is a methodology question.  I have two forms of a survey that
are administered either online or via mall intercept.  My task is to
compare the surveys to determine whether they are comparable in terms of
the information being collected.  I was planning on doing t-tests, anovas,
and general qualitiative observation of the means.  The problem is that the
although the questions are quite similar (but not totally identical) the
rating scales differ--one survey may have a 4-point rating scale while the
other one might have a 5-point scale with a "don't know" or "no opinion"
option.  So what i'm finding is that even though the questions are close
enough to compare, the number of intervals, the   rating labels, and the
overall available options are differing.  Can anyone suggest methods to
approach this problem of comparing survey performance when the instruments
are not identical?
Thanks,
 Mike