Greetings Peter,

I would expect the following code to work:

proc sql; select a.idvar as a, b.idvar as b, c.idvar as c from a left join b on a.idvar = b.idvar left join c on a.idvar = c.idvar; quit;

Francis R. Harvey III WB303, x3952 harveyf1@westat.com

VB programmers know the wisdom of Nothing

> -----Original Message----- > From: Peter Baade [mailto:Peter_Baade@HEALTH.QLD.GOV.AU] > Sent: Tuesday, March 05, 2002 1:15 AM > To: SAS-L@LISTSERV.VT.EDU > Subject: SQL logic > > > Hi > > I have three datasets, each with a unique ID variable (IDVAR). > > The first dataset (A) has one record for each person. The > second and third datasets (B and C) have one record per > person, but not necessarily for all people. > > I was using SQL to join these, but if I use the where statement > IF A.IDVAR=B.IDVAR AND A.IDVAR=C.IDVAR > then only those records with data in each dataset are pulled out. > > So, pulling out a paper by Christina Williams (Proc SQL for > Data step die hards), I find that using a LEFT JOIN would be > better logic. However, it seems as if this only works for two > datasets, not three. > > Is there some trival SQL procedure that does what I want? Its > more the logic that I'm trying to find, not necessarily the > specific code. > > Thanks for any advice, and my apologies if I've missed some > important piece of information in the above question :-) > > Peter. <snip>