Date: Tue, 5 Mar 2002 17:15:44 -0500
Reply-To: Chun Ling Lu <cllu@HSPH.HARVARD.EDU>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Chun Ling Lu <cllu@HSPH.HARVARD.EDU>
Content-Type: text/plain; charset="us-ascii"; format=flowed
I have a problem in creating a interaction variable.
id ratio working
1 0.12 1
2 0.67 0
3 0.89 1
where ratio is continuous var, wroking is dummy.
I wanted to create rawork=ratio*working. However, using my dataset, I got
something like this
id ratio working rawork
1 0.12 1 0.13
2 0.67 0 . (missing)
3 0.89 1 0.89.
where rawork for id=2 is missing, which does not make sense.
I then pull out these obs with unreasonable missing values such as id=2
case, and created a new interaction variable with the same calculation:
raworknew=ratio*work, this time, it worked, I got things like this:
id ratio working rawork raworknew
2 0.67 0 . (missing) 0
Can anybody tell me what's going on?