Re: Estimating Group Effects and Differences

Posted by Bruce Weaver on May 17, 2019; 12:48pm
URL: http://spssx-discussion.165.s1.nabble.com/Estimating-Group-Effects-and-Differences-tp5737907p5737910.html

If you cross-tabulate nativity, site and sex using your entire data set, what
do you see?  Try this:

CROSSTABS nativity BY site BY SEX.

In the output, right-click the cross-tabulation table and Copy As Plain
Text, and paste the plain text table into a reply.  It may not line up
perfectly in Nabble or in people's e-mail, but readers can copy it and paste
it into a text editor that uses a fixed font and insert tabs where needed.
E.g., here's the cross-tabulation I get using the 4 records you provided:

nativity * site * SEX Crosstabulation
Count
SEX site Total
                        1 2
f nativity 1 1 0 1
                2 0 1 1
        Total 1 1 2
m nativity 1 1 0 1
                2 0 1 1
        Total 1 1 2
Total nativity 1 2 0 2
                2 0 2 2
        Total 2 2 4


If nativity and site are indeed completely confounded in your entire data
set, you can't include both of them in your model.  And you will be unable
to determine how much of any difference between the two cells is due to
nativity, how much is due to site, or how much is due to the interaction
between them.  





Chao yawo-2 wrote

> Bruce et al:   Thanks for your response
>
> Here is an example of the data requested.   As you can see, foreign-born
> women are only recruited in the community and not clinics, and
> native/american born women are recruited and exposed in the clinic and not
> the community.
>
> DATA LIST /ID 1-3 SEX 5 (A) AGE 7-8 nativity(10) site (11) prep(12).
> BEGIN DATA
> 001 m 28 110
> 002 f 29 221
> 003 f 45 111
>  ...
> 128 m 17 220
> END DATA.
>
> Where:
>
>
>    - nativity(1--American; 2=foreignborn)
>    - Site: (1=clinic; 2-community)
>    - prep exposure: (0=no; 1=yes)
>
>
> Hope this helps - thanks - Yy

-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD