Date:         Fri, 18 Feb 2005 11:29:59 -0700
Reply-To:     Geoff Matthews <gmatthew@wgu.edu>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Geoff Matthews <gmatthew@wgu.edu>
Subject:      Re: Recoding for the earliest date
Comments: To: "Johnson, Wendy RDECOM (PKI)" <Wendy.Kullen.Johnson@us.army.mil>
Content-Type: text/plain; charset="us-ascii"

Worked wonderfully.  Thank-you.

Geoff Matthews

-----Original Message-----
From: Johnson, Wendy RDECOM (PKI)
[mailto:Wendy.Kullen.Johnson@us.army.mil]
Sent: Friday, February 18, 2005 11:21 AM
To: Geoff Matthews; SPSSX-L@LISTSERV.UGA.EDU
Subject: RE: Recoding for the earliest date

I think this should work:

Compute firstdat = min(date1 TO date68).

HTH,

~Wendy.


-----Original Message-----
From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU] On Behalf Of
Geoff Matthews
Sent: Friday, February 18, 2005 1:13 PM
To: SPSSX-L@LISTSERV.UGA.EDU
Subject: Recoding for the earliest date

I have a dataset which contains 68 date variables.  I need to find the
earliest one for each 5,000+ subjects.  While it would have made sense
to have date1 earlier than date2, this is not the case.



I had consigned myself to writing:



COMPUTE firstdat = date1.

DO IF (firstdat >= date2).

RECODE

Date2

(ELSE=COPY) INTO firstdat.

End if.



68 times, but there has got to be an easier way.  Any suggestions?





Geoff Matthews

Institutional Research Analyst

Western Governors University

801-327-8113