Re: Re: st: randomly selecting subjects from clusters

[n j cox <n.j.cox@durham.ac.uk>]

Thanks for the very helpful extra clarification.
You are correct: the problems do differ slightly.

Nick
n.j.cox@durham.ac.uk

Jeph Herrin

The FAQ almost answers the question, but not quite: to get
one random individual per household (as opposed to a random
sample of households), try:

  set seed 12345678
  gen random=uniform()
  bys household (random): gen byte mysample=_n==1

Then -mysample- will be one randomly selected individual
per household.

n j cox

> The Statalist FAQ advises, inter alia, consideration of
> the existing FAQs. In this case,
>
> How can I sample clusters, not individuals?
> http://www.stata.com/support/faqs/data/sampleby.html

Mariano Sana

> I have a data set where each record is an individual in a household. 
I have
> all members of each household. How do I make a dataset that has only one
> randomly selected member from each household?

*
*   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/