Date:         Mon, 29 May 2006 14:30:04 -0400
Reply-To:     Richard Ristow <wrristow@mindspring.com>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Richard Ristow <wrristow@mindspring.com>
Subject:      Re: Select cases before adding variables from another file
Comments: To: Bob Schacht <schacht@hawaii.edu>
In-Reply-To:  <6.2.1.2.2.20060525105419.0289ae60@hawaii.edu>
Content-Type: text/plain; charset=us-ascii; format=flowed;
              x-avg-checked=avg-ok-5BF379AB

At 05:02 PM 5/25/2006, Bob Schacht wrote:

>Suppose I select cases in my working file, and then do a "Merge
>Files/Add variables." For the merge, will only selected cases be
>examined, or will
>the selection be ignored?

Menu entry "Merge Files/Add variables" generates a MATCH FILES command.
If you include the working file in MATCH FILES, it'll be the working
file as it exists when the MATCH FILES is run. All earlier
transformations, including selection, will have been applied. I'm
pretty sure that filtering will be ignored, though.

>I'm asking because in my working file, I've got active cases as well
>as closed cases. I want to add variables from another file that
>contains only closed cases, so I don't want SPSS to get confused
>trying to match closed cases with active ones. So, if I select only
>the closed cases, will the "add variables" menu affect only the
>selected cases?

To make sure we both understand what you want: Your working file has
some closed cases and some open ones. Another file has other
information on some closed cases. And you want a working file with an
entry for every closed case (but not the open ones) that occurs in
either file. That working file will have information from whichever of
the two files the case occurs in; from both, if it occurs in both.

By the way, you write,

>I don't want SPSS to get confused trying to match closed cases with
>active ones.

If there's some case identifier that's unique over all cases, closed
and open, SPSS wouldn't get confused. It would simply keep the open
cases in the combined file. Presumably their entries would not have any
information from the other file, since that file contains only closed
cases.

You still probably want to do the selection, but it's because of what
you want, not because of what SPSS needs.