Status:Closed    Asked:Jul 21, 2015 - 08:52 AM

Using variables with different weights

I want to do an analysis that involves some variables (WKSTAT, EMPSTAT) that take the WTSUPP weighting, and another (PAIDHOUR) that takes the EARNWT weighting. How can I do this? When I use WTSUPP I get a large NIU, and when I use EARNWT the totals get all wacky. Any suggestions? (My ultimate goal is to isolate the population of full-time, nonhourly-paid workers, and then analyse them by pay and occupation, for which I was going to use INCWAGE and OCC respectively. If I should be going about this some other way, could someone clue me in?)

Do you have the same question? Follow this Question

Staff Answer




The EARNWT variable is more restrictive than WTSUPP. So, if one of your variables requires EARNWT (e.g. PAIDHOUR), then you should use EARNWT. I also recommend comparing the Universe statements for each variable.

Hope this helps.


Jul 21, 2015 - 05:44 PM

Report it


Does anyone have a method for handling missing data in CPS "EDUC" variable? 2...
How do I find the data file:Ipumsi_00001.dat for the 1974
Variables from Full Year Consolidated Files currently not included in IPUMS-M...
Wrong HH weight Tanzania 2012 census?
Assign Individuals to PUMAS, append dataset
Regarding: Does CLASSWKR in ASEC before 1988 include incorporated and non-inc...
Login   |   Register

Recently Active Members

View More »

Share |