Date:         Wed, 4 Jun 2003 09:59:07 +0100
Reply-To:     Peter Crawford <Peter@CRAWFORDSOFTWARE.DEMON.CO.UK>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Peter Crawford <Peter@CRAWFORDSOFTWARE.DEMON.CO.UK>
Organization: Crawford Software Consultancy Limited
Subject:      Re: Create groups

VJ <vjoshi73@YAHOO.COM> writes
>Hi,
>I have a variable age on my dataset.  I am trying to create a new variable
>AGE_CATEGORY that creates groups such as 25-30, 31-35 based on value of age.
>Currently I have done this with help of Proc Univariate by getting the
>mean, median of age etc.  I feel that my way may not be the most efficient
>way since this process of creating age groups is something I might have to
>repeat on multiple large datasets.   I also don't want to end up with small
>cell sizes in any of the age group.
>
>I have tried to see if there is a SAS proc that will automatically create
>groups based on distribution of a variable and I have not been successful.
>Does anyone have any ideas or other solutions?
>
>Thanks in advance.
>Vaishali


Sounds like a good use for percentiles of the range
I think these are available to you on v8 proc means, but certainly
through proc univariate.


--
Peter Crawford

available for SAS consultancy contracts