Question

Status:Closed    Asked:Sep 05, 2013 - 01:26 PM

Which weights to use with October/Education supplement?

I was wondering which weights should I use if I am focusing on the data from October/Education supplement but also include other socio-demographic variables such as age, sex, total income, etc. Would WTSUPP work?

In addition, how do I use these weights in STATA? Is this the only command that I need:

" svyset _n [pweight=wtsupp], vce(linearized) singleunit(missing)"

or do I need to also declare sampling units, strata, and finite population correction?

Thank you.

 
Do you have the same question? Follow this Question
 

Staff Answer

avatar

Joe_Grover

Staff

You would want to use WTSUPP for an analysis focused on the Education Supplement. For a more complete discussion of when different weights should be used for IPUMS-CPS, see the response to this question. Your svyset command should produce results comparable to those published by the Census Bureau, however, due to the complex sampling structure of CPS you may also want to look into recent published research for best practices in your specific research area.

 

Feb 11, 2014 - 02:59 PM

0
0
Report it

OTHER QUESTIONS NEEDING ANSWERS

CPS October Education Supplement Availability after 2014
Are there replicate weights or is there a strata variable for the October CPS...
Unable to reproduce weights from CPS Voting supplement
Login   |   Register

Recently Active Members

View More »

Share |