Thanks Richard, it's work fine.

-----Original Message----- From: Richard Ristow [mailto:wrristow@mindspring.com] Sent: September 2, 2004 3:35 PM To: Mahbub Khandoker; SPSSX-L@LISTSERV.UGA.EDU Subject: Re: Date again

At 02:06 PM 9/2/2004, Mahbub Khandoker wrote:

>I would like to convert the string date to normal date format. >The dates (string) are now as follows, >2004 Jan,11 >2003 Oct,18 >2004 Jun,10 >2003 Aug,15 >2004 May,03

Many date input problems in SPSS require some string manipulation and the NUMBER function. Fortunately, that's usually all they take. Here's one solution for this format. Tested code; this is SPSS draft output:

NEW FILE. DATA LIST FIXED /CHAR_DT 01-11 (A).

Data List will read 1 records from the command file Variable Rec Start End Format CHAR_DT 1 1 11 A11

BEGIN DATA. 2004 Jan,11 2003 Oct,18 2004 Jun,10 2003 Aug,15 2004 May,03 END DATA. /*-- LIST /*-*/.

STRING INTNL_DT (A11). COMPUTE INTNL_DT = CONCAT(SUBSTR(CHAR_DT,10,02),'-', /* Day */ SUBSTR(CHAR_DT,06,03),'-', /* Month */ SUBSTR(CHAR_DT,01,04)) /* Year */. /*-- LIST /*-*/.

- NUMERIC SPSS_DT (DATE11). - VARIABLE WIDTH SPSS_DT (11). - VARIABLE ALIGNMENT SPSS_DT (CENTER).

COMPUTE SPSS_DT = NUMBER(INTNL_DT,DATE11). /**/ LIST /*-*/.

List CHAR_DT INTNL_DT SPSS_DT

2004 Jan,11 11-Jan-2004 11-JAN-2004 2003 Oct,18 18-Oct-2003 18-OCT-2003 2004 Jun,10 10-Jun-2004 10-JUN-2004 2003 Aug,15 15-Aug-2003 15-AUG-2003 2004 May,03 03-May-2004 03-MAY-2004

Number of cases read: 5 Number of cases listed: 5