Question

Status:Closed    Asked:Oct 27, 2016 - 07:47 AM

earnwt missing in 1979-1981 in IPUMS CPS Basic Monthly

When I extracted the CPS data using a different interface, the earnings weight was available for all years, including 1979-1981.

When I extract the CPS data using the IPUMS CPS website, earnwt is available from 1982 onwards only.

Is it possible to obtain earnwt also for 1979-1981?

 
Do you have the same question? Follow this Question
 

Staff Answer

avatar

Jeff Bloem

Staff

The EARNWT variable is only available from 1982 onward on the IPUMS-CPS site. It should be noted, however, that the EARNWT variable is only recommended for the Earner Study variables, none of which are available prior to 1982 either. For any other analysis using ASEC data, researchers are recommended to use the WTSUPP variable and for analysis on non-ASEC data the WTFINL variable.

 

Oct 27, 2016 - 09:26 AM

0
0
Report it

Answers

Thanks for your helpful answer. I use "earnwt" when using the variable "empstat" from the "Basic Monthly" CPS samples. When comparing my own aggregations to the "official" statistics from the CPS Labor Force Statistics, it seems that "earnwt" is indeed the correct weight for this variable. But "empstat" *is* available between 1979-1981. Can you give advice which weights to use when aggregating "empstat"?

 

Oct 27, 2016 - 10:06 AM

0
0
Report it

If you are using basic monthly samples then the WTFINL variable is recommended. More detailes are provided on the variable description tab.

 

Oct 27, 2016 - 10:21 AM

0
0
Report it

Thanks again. Yet, I have doubts. I downloaded the "official" Labor Force Statistics from the Current Population Survey and compared this to the aggregated results one obtains using different weights.


When I use EARNWT, I can more or less perfectly replicate the "official" Labor Force Statistics (the two stars indicate two problematic months (2001:4 and 2001:6) (see attachment).


When I follow your suggestion and use WTFINFL instead, the series look less similar to the reference series (see attachment).


This may be coincidence, but I still wonder whether the official data are indeed aggregated using the WTFINFL weight. Do you have information on that?

 

Oct 28, 2016 - 07:47 AM

0
0
Report it

OTHER QUESTIONS NEEDING ANSWERS

USA IPUMS 1980 ancestry variable -- mapping to country names
Did the version of database change since 2011?
Are DATANUM, SERIAL, and PERNUM the same in single-year and 5-year ACS sample...
How can I accurately calculate the median family income by state for families...
Detailed Variables Dropping From Dataset
Are weights adjusted when using "select cases" in an extraction?
Login   |   Register


Share |