Date:         Thu, 22 Apr 2004 21:23:28 -0700
Reply-To:     Bruce Bradbury <BruceBrad@INAME.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Bruce Bradbury <BruceBrad@INAME.COM>
Organization: http://groups.google.com
Subject:      Re: phreg when individuals are clustered
Content-Type: text/plain; charset=ISO-8859-1

> I am conducting a study on chick survival in sage-grouse and want to
> develop a proportional hazard model.  I radio-marked 2-3 individual
> chicks within a family unit.  I am trying to figure out the best way to
> analyze the data.  Can anyone give me some help.

You can do a fixed-effects partial likelihood estimation using the
/strata option in phreg. This is like adding in a dummy variable for
each family.

See chapter 8 in Alison, Paul (1995), Survival Analysis Using the SAS
System; A Practical Guide (He discusses this in the context of
repeated spells, but your problem appears to be similar).