Date:         Tue, 21 Nov 2000 09:43:51 -0600
Reply-To:     Carol Albright <syzygy@GARNET.TC.UMN.EDU>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Carol Albright <syzygy@GARNET.TC.UMN.EDU>
Subject:      Re: Grouped Means
Comments: To: Keith Wood <Kwood87@MEDIAONE.NET>
In-Reply-To:  <3A19CEDB.2EEE0A0B@mediaone.net>
Content-Type: text/plain; charset="us-ascii"

At 07:24 PM 11/20/00 -0600, Keith Wood wrote:
>I have a survey that has 10 questions. Questions 1 and 2 relate to the
>same characteristic. What do I have to do in SPSS to group those two
>responses together to get a shared mean?

Hi, Keith

        Is this what you mean by a "shared mean" -- adding the two responses
together and then dividing by two to get the mean for each respondent?
Then you can figure out the mean (or median or ...) of that new variable?

        Using syntax, you could do the following:

compute NEWVAR = (VAR1 + VAR2)/2.
variable labels NEWVAR "Mean of Whatever This Is".
execute.

        Substitute the name of the original variables for VAR1 and VAR2, and a
meaningful name for NEWVAR, the mean of the original items.  If either VAR1
or VAR2 is missing, NEWVAR will also be missing.

        On another note, with so few questions in your survey, I'm not sure it's
worth it to collapse items.  Also, unless you have at least 3 -4 items that
you're folding together, the new variable may not be very reliable (?) (get
a statistician's opinion on this as I'm wandering outside my domain).

Carol
-------------------------------------------------------------------------
Carol L. Albright, MS             | E-Mail   : syzygy@tc.umn.edu
Albright Consulting               | Phone   : 651/699-7218
2031 Goodrich Avenue         | http://www.tc.umn.edu/~syzygy
St. Paul, MN 55105 USA       |  Research data services
-------------------------------------------------------------------------