Date: Thu, 29 Dec 2005 12:35:43 +0300
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: Anton Balabanov <Anton.Balabanov@fup.unn.ru>
Subject: Re: creating a composite variable to count number of mentions
Content-Type: text/plain; charset="ISO-8859-1"
if I understood properly, this situation is typical to apply multiple
response option (under Analysis --> Multiple Response). It could be run also
Suppose, you have the following data:
DATA LIST LIST/id v1 to v4.
1 1 3 2
2 4 5 3 1
3 3 6 4 3
4 1 2
5 4 6
VALUE LABELS v1 to v4
1 'tought 1'
2 'tought 2'
3 'tought 3'
4 'tought 4'
5 'tought 5'
6 'tought 6'
7 'tought 7'
8 'tought 8'
9 'tought 9'
10 'tought 10'.
The following command will do this job:
GROUPS=$toughts (v1 v2 v3 v4 (1,10))
It gives you the percentage from the total number of opinions expressed as
well as the persentage from the total number of people expressed at least
Hope this helps,
From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU]On Behalf Of Nico
Sent: Thursday, December 29, 2005 4:59 AM
Subject: creating a composite variable to count number of mentions
I have 4 variables that are the result of coding an open-ended question.
There are four because some people's answers contained mroe than one thought
(a maximum of four thoughts).
I'd like to create a single variable (I think?) or compute something that
tells me simply what proportion of the time a certain code was mentioned.
Say I have 10 codes - I'd just like to be able to report what proportion of
people mentioned a code AT ALL.
Thanks in advance for any tips.
I'm on SPSS 13.
Nico Peruzzi, Ph.D.