Svyset for regression using Colombia 2005 Census

INTERNATIONAL
1

Disclaimer: I am a stats newbie, so apologies for asking perhaps an obvious question…

I am running a regression to see how household water and sanitation facilities affect years of schooling for children. I am uncertain as to whether I have weighted the data correctly using svyset. I have had a look at most of the literature on weighting etc but still am unsure!

So far I have:

svyset serial [pweight=perwt], strata(strata)

Is this correct?

Thank you in advance and hope everyone is safe and healthy!

2

Yes, it seems that you have specified this correctly for individual-level regression analysis. Note that there are numerous ways to specify sampling weights in Stata and so this is only one approach that you can use. Here is a nice resource for using svy in Stata.

3

Thank you Jeff!