How can we get stata version of CPS outgoing rotation group variable EARNWEEK for ASEC sample, years before 1989?
It seems there is no stata read (do) file that allows me to read the CPS ASEC data in dat file from NBER website.
I’ve checked other data sources such as CERP, Census but it’s either that they do not have early years of ASEC data (Census) or they did not combine the outgoing rotation group variables in the ASEC file.
Also I am curious about the source of reasoning that why the EARNWEEK variable is unreliable for ASEC sample before 1989.
(i.e. the exact reason that IPUMS did not contain EARNWEEK for ASEC sample 1982-1988.)
That way I might be able to figure out a way to use those information with some adjustment.
Thank you.
As noted in the EARNWEEK description, the Census Bureau reports that the results in the CPS public use files for this data series included errors for years prior to 1990, so only data from 1990 forward are part of IPUMS-CPS.
This NBER page of programming files has some .do files available for March pre-1989. The corresponding .dat files are available on the NBER CPS Supplement page. The CPS basic monthly files page (which has corresponding .do files available for data from 1989 on) also includes a brief explanation on how to properly use these files. However, given that there are issues with the data in these years, I’m not sure how helpful this will be in finding EARNWEEK in samples prior to 1989.
The Census Bureau does not elaborate on what these errors actually are. You may want to contact the Census Bureau or the Bureau of Labor Statistics as they would have more information on the subject.