On Thu, 22 Jun 2000 16:16:55 -0700, sig
<[EMAIL PROTECTED]> wrote:
> I took a sample to determine if users preferred one of 3
> different designs (A, B, and C). Of the 740 people
> sampled, 300 replied "no preference", 160 preferred
> prototype A, 130 B, and 150 C.
>
> Is there a significant difference between the protoypes? If
> so, which ones? Which test do I use? How do I handle
> the "no preference" data? I need help.
What are your choices for handling "no preference"? Unlike the
instance of an equal amount of Missing Data - which would cast extra
doubt on the representativeness of the survey - you do have the
information here, that 40% of the sample don't care.
It seems unlikely to me that you expected there to be 25% of the
responses falling into each category or none, which is one easy thing
to test for. Maybe, equal across the 3?
A better survey might have asked for strength-of-preference.
Or you might have asked "Are any of these particularly BAD?"
Or, there is a style of Voting, in which you look at the second
preference for each of the 130 B's (since B was the low category).
This is used in some elections in Canada, if I remember right.
--
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
===========================================================================
This list is open to everyone. Occasionally, less thoughtful
people send inappropriate messages. Please DO NOT COMPLAIN TO
THE POSTMASTER about these messages because the postmaster has no
way of controlling them, and excessive complaints will result in
termination of the list.
For information about this list, including information about the
problem of inappropriate messages and information about how to
unsubscribe, please see the web page at
http://jse.stat.ncsu.edu/
===========================================================================
