Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Nick Cox <njcoxstata@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: Missing value problem with new variable |
Date | Tue, 11 Sep 2012 10:39:03 +0100 |
I guess that the problem is not whether -occupation- is ever missing. It is whether every household has someone for whom -relationshiptohhh == 1-. You can check that by e.g. egen nohhh = total(relationshiptohhh == 1), by(SRLNO sites) and looking for zeros. Also your code is contradictory. If -Q1_1_7- is the right variable, why use -occupation-? Nick On Tue, Sep 11, 2012 at 10:20 AM, Wameq Raza <wameq.r@gmail.com> wrote: > For my data set that deals with health seeking behavior for a acute > illnesses, I've been trying to include the occupational details of the > household head for each of the patients who seek care. More > specifically, I want to create a variable where the occupation of the > household head is stored for every individual in the household. > > To do this I've followed these steps: > 1. gen emp_hhh=occupation if relationshiptohhh==1 /where > Q1_1_7 is the occupation variable (categorical variable) and > relationshiptohhh equals to 1 if it's the head of the household > 2. bysort SRLNO site: egen emp_hh = max(emp_hhh) /where "SRLNO > site" are the ID variables for each household. > > My understanding is that the new var "emp_hh" should have what I want, > but there seems to be a number of missing values. I've checked the > original occupation var for missing values from which the new var was > constructed, but there were none. Might there be something wrong with > my syntax? > > I'm having a similar problem with the education variable as well. * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/