| cluster average of women's working status [message #13800] |
Thu, 28 December 2017 07:52  |
ahmed89o
Messages: 26
Registered: August 2013
Location: Germany
|
Member |
|
|
Dear DHS team and users,
I would like to develop a variable that the percentage of working women at the cluster level excluding the woman's own working status or I shall call it the cluster average of women's working status. I would be very grateful if you can share STATA codes.
|
|
|
|
| Re: cluster average of women's working status [message #13803 is a reply to message #13800] |
Thu, 28 December 2017 15:55  |
Reduced-For(u)m
Messages: 292
Registered: March 2013
|
Senior Member |
|
|
I think this is a question for the Stata forum - but one suggestion would be to use "egen , by()" commands. I'm assuming you have 0/1 variables for "working". In that case, first, compute the cluster-level denominator, which is N for that cluster (I think). Then generate the cluster-level numerator by adding up the 0's and 1's in the cluster for everyone. Then, for each person, use as the numerator "total - own" and for the denominator "total - 1".
Something like that should work...but again, the Statalist may be a better resource for you.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
