Re: st: calculating group means

[David Kantor <dkantor@jhu.edu>]

At 04:40 PM 7/25/2002 -0400, Gayatri Koolwal. wrote:
> I have a data set of individuals who belong to different households, each 
> with a value of a variable x2.  For each household, I would like to create 
> a new variable that is the mean of x2 for the individuals in that 
> household.  Would anyone be able to help me with the procedure for 
> this?  Do I need to collapse the data set first somehow?
As Marcela Perticara indicated, you can use egen ... mean ... by(householdid)

But note that there are two possibilities.  If you want to take a household 
mean and spread it onto your set of individuals (by household), then use 
egen ... mean ... by(householdid).  But if you want to reduce to a set of 
households (i.e., one observation per household), then -collapse- is more 
convenient.  (You could still use egen ... mean, but you would then want to 
reduce it to one representative per household.)

I hope this helps.
-- David

David Kantor
Institute for Policy Studies
Johns Hopkins University
dkantor@jhu.edu
410-516-5404

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/