Date: Fri, 25 Feb 2005 15:54:53 +0100
Reply-To: Spousta Jan <JSpousta@csas.cz>
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: Spousta Jan <JSpousta@csas.cz>
Subject: Re: compute problem
Content-Type: text/plain; charset="iso-8859-2"
Hi Paul,
Aggregate the data by subjects computing the average sex in each combination. Then recode the average sex into 1-2-3 and merge the aggregated file back to the original file as a "keyed table" using subjects as keys.
Much more interesting will be probably to test whether the result was due to a chance or whether it is a true property of the gender, I think :-)
Wishing a nice weekend to you and to all
Jan
-----Original Message-----
From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU]On Behalf Of
Paul Mcgeoghan
Sent: Friday, February 25, 2005 12:54 PM
To: SPSSX-L@LISTSERV.UGA.EDU
Subject: compute problem
Hi,
I have the following problem:
There are 20 variables Subject1 to Subject20 with values 1 if they chose the
subject and 0 if they didn't.
I want to compute a new variable where if males and females have chosen the
same combination of subjects, the value is 1; if only males choose a particular
combination of subjects, it is 2, if only females choose a particular
combination of subjects, it is 3.
So if I have for example 3 variables: Subject1 to Subject3 with data:
Row Sub1 Sub2 Sub3 Gender NewV
1 1 1 1 1 1
2 1 1 1 2 1
3 1 0 0 1 1
4 1 0 0 2 1
5 1 0 1 1 2
6 0 0 1 2 3
In the last row of data (row 6), only females chose that particular combination
whereas in row 5, only males chose that particular subject combination.
What is the easiest way of doing this in SPSS?
Paul
==================
Paul McGeoghan,
Application support specialist (Statistics and Databases),
University Systems Group (USG),
Information Services,
Cardiff University.
Tel. 02920 (875035).
