=========================================================================
Date:         Wed, 5 Jul 2006 11:35:42 -0400
Reply-To:     "Luo, Peter" <pluo@draftnet.com>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         "Luo, Peter" <pluo@draftnet.com>

A researcher came to me with a stat question. She's trying to determine the
how a group of consumers perceived a brand. Here is what she did:



For each brand she deducted a 'net' gain, which is the balance of the % of
consumers who answered "more likely' and % who answered 'less likely.' It's
a 3 choice question: 1-less likely, 2-the same, 3-more likely.



Then she took the average of the net gains cross all 15 brands.



Her question is: to what extent one can determine if a particular brand was
significantly more (or less) favored by these consumers, say if brand A got
a net gain of +10%, but the average net gain is +15%, can I be confident to
say the brand A is less favored?



Anyone here on the list can give me some hints as to how to solve this stat
question?  There are two things that concern me: 1. she's trying to compare
the 'part' against the 'whole,' and 2. the number of consumers who answered
each brand varied from 35 to 74.