Date:         Thu, 10 Jul 1997 17:10:20 -0800
Reply-To:     Hank Becker <hjbecker@UCI.EDU>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
From:         Hank Becker <hjbecker@UCI.EDU>
Organization: University of California, Irvine
Subject:      Correl. for 1 var. recorded in pairs of records
Content-Type: text/plain; charset=us-ascii

(Sorry for re-post.  It disappeared from the list in just a couple of
days.)
 
I have a data file where the records are in pairs (much like one would
have with separate pre- and post-test files but merged into one).  I
want to compute simple correlations between the pairs of responses for
each variable.  I have hundreds of such variables.  Is there any SPSS
procedure or simple macro that would enable me to merge pairs of records
without having to rename all those variables?
 
I saw a previous suggestion which was to aggregrate by record type and
create new pairs of variables out of the old single variables.  But this
seems very time-consuming to do for hundreds of variables (already
named).  Any other suggestions?
 
Hank Becker
UC Irvine, Education