Hello david,

If I see it right, you must simply use the following:

if datepart(Srvy_DT) > '04APR2001'd;

greetings from schwaebisch hall, germany ronald hoellwarth

---------------------- Weitergeleitet von Ronald Höllwarth/PK/Kreditwerk on 09.05.2001 09:22 ---------------------------

"David L. Alderton Ph.D." <David.Alderton@ATT.NET>@LISTSERV.UGA.EDU> on 09.05.2001 05:19:58

Bitte antworten an David.Alderton@PERSNET.Navy.mil

Gesendet von: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>

An: SAS-L@LISTSERV.UGA.EDU Kopie: (Blindkopie: Ronald Höllwarth/PK/Kreditwerk) Thema: [SAS-L] Help with DateTime22.3 format

I'm pulling data off a SQL Server 7 database. One of the variables is the date a survey was filled out. It comes across as a SAS numeric variable with a datetime22.3 format. For example '07FEB2001:09:30:20.177' is one value with a crazy numeric translation of 1297157420.177 which I guess is the number of minutes since some date. Anyway, what I am trying to do is subset the data as I bring it across from the server based on the last time I pulled the data. Say I pulled the data from the beginning of the survey until 4 April 01. This time I want to get just the surveys that have been filled out since 4 April 01 (to today). I cannot for the life of me figure out how to translate the datetime22.3 into just a traditional date (e.g., '04APR01'd) nor can I figure out how to create a variable on the same metric as the datetime22.3 value to use for comparison.

Ideally, I simply want to create a constant for use in a subsetting "if" statement... if Srvy_DT > XXX; Where XXX would be directly comparable number to Srvy_DT (datetime22.3 format). Or, I want to translate the Srvy_DT into a standard date '04APR01'd and then compare it to another standard date.

Can anyone help? I have never used dates in this format and I am stumped.

Thanks, David. David L. Alderton, Ph.D. Institute for Selection and Classification Navy Personnel Research, Studies, and Technology Navy Personnel Command