Date: Wed, 30 Jun 2004 09:02:22 -0400
Reply-To: Dr Julius Kitutu <jmm+@pitt.edu>
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: Dr Julius Kitutu <jmm+@pitt.edu>
Subject: Re: date computation (6- decimal places needed)
In-Reply-To: <40E2B214.4@verizon.net>
Content-Type: text/plain; charset="us-ascii"
I had send the appended message to Paul...but hope it is an alternative to
what Kendall is suggesting.
Julius
Hi Paul,
I do not know the format of your date variables, but if you have specific
dates like mm/dd/yyyy, for both variables, then you can try the following
syntax:
COMPUTE agerece = (XDate.Tday(intervie) - XDate.Tday(birth_da))/365.24.
With this kind of computation, you can have as many decimal places as you
wish. Good luck.
Julius
-----Original Message-----
From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU] On Behalf Of Art
Kendall
Sent: Wednesday, June 30, 2004 8:29 AM
To: SPSSX-L@LISTSERV.UGA.EDU
Subject: Re: date computation (6- decimal places needed)
I have modified a small age calculation demo I have. I added the age
in days and the age as fractional.
note that the age as fractional is inaccurate because of leap days.
Would age in days serve you as close to what you want? Wouldn't that be
more than enough accuracy even if you were talking about neonates?
Cut-and-paste then run the syntax at the end of this message.
Hope this helps.
Art
Art@DrKendall.org
Social Research Consultants
University Park, MD USA
(301) 864-5570
new file.
*make up some data.
data list list / caseid (f2) birthdat(adate10) refdate(adate10) wanted
(f3).
begin data.
01 4/10/1931 4/10/2002 71
02 4/10/1933 4/10/2002 69
03 4/10/1932 4/7/2002 69
04 4/10/1932 4/8/2002 69
05 4/10/1932 4/9/2002 69
06 4/10/1932 4/10/2002 70
07 4/10/1932 4/11/2002 70
08 4/10/1932 4/12/2002 70
09 4/10/1932 4/13/2002 70
10 1/1/2000 1/1/2003 3
11 1/1/2001 1/1/2003 2
12 1/1/2002 1/1/2003 1
13 12/31/2002 1/1/2003 0
14 1/2/1933 1/1/2003 69
16 1/1/1933 1/1/2003 70
16 1/1/1899 1/1/2000 101
end data.
numeric age (f3) agedays (f10)agefract (f17.6).
compute agedays = ctime.days (refdate-birthdat).
compute agefract= agedays/365.25.
* method2 see if birthday in current year has occurred yet.
compute bmo = xdate.month(birthdat).
compute bdom = xdate.mday(birthdat).
compute byr = xdate.year(birthdat).
compute amo = xdate.month(refdate).
compute adom = xdate.mday(refdate).
compute ayr = xdate.year(refdate).
compute nearbirt = date.mdy(bmo,bdom,ayr).
do if nearbirt le refdate.
compute age = ayr-byr.
else if nearbirt gt refdate.
compute age = ayr-byr-1.
end if.
list /variables = caseid birthdat refdate wanted age agedays agefract.
Paul Dickson wrote:
>Hi servers
>
>We are using the following data computation syntax. The problem is that it
is
>rounding ages up and down to '0' decimal places. We were hoping to get the
>syntax to create new ages based on 6 decimal places to facilitate optimal
coding
>
>Thanks in advance
>
>Paul
>
>COMPUTE agerece = xdate.year(intervie) - xdate.year(birth_da).
>EXECUTE .
>
>
>
