You can do this by using the BEST. format -- also, the LENGTH statement has influence on granularity, as where the FORMAT statement influences the output format being displayed. As well, the higher the date, the larger integer portion of the SAS variable occurs. Also, the SAS OS platform has some influence. Suggested links for reading provided below. Scott Barry SBBWorks, Inc. SAS 9.2 Language Reference: Concepts - Numeric Precision in SAS Software: http://support.sas.com/documentation/cdl/en/lrcon/61722/HTML/default/a000695157.htm SAS 9.2 Language Reference: Concepts - About SAS Date, Time, and Datetime Values http://support.sas.com/documentation/cdl/en/lrcon/61722/HTML/default/a002200738.htm Step-by-Step Programming with Base SAS(R) Software - Working with Numeric Variables: http://support.sas.com/documentation/cdl/en/basess/58133/HTML/default/a001304311.htm ... View more

Hello ieva and Cynthia. Thanks for your answers. I was confused about when the where statement is about the input or the output. But hey, it's only my first week of SAS ;-). I also found this in the help which clears up things a lot. Example 1: Selecting Observations from an Input Data Set This example uses the WHERE= data set option to subset the SALES data set as it is read into another data set: data whizmo; set sales(where=(product='whizmo')); run; Example 2: Selecting Observations from an Output Data Set This example uses the WHERE= data set option to subset the SALES output data set: data whizmo(where=(product='whizmo')); set sales; run; I think it is safe to assume according to the help that the WHERE statement is always about the input set. Furthermore it is kind of tricky to know when drop= and where= are applied. From the help I cite: "Do not confuse the purpose of the WHERE= data set option. The DROP= and KEEP= data set options select variables for processing, while the WHERE= data set option selects observations." Message was edited by: Michiel Borkent ... View more