Hi, I’m seeing a lot of “blank” earnings for employed spouses/partners:
count if spempnot==1 & spearnweek==99999.98
12,932
Why is this? I tried cross-tabulating with ‘spempstat’ but that doesn’t clarify.
Perhaps relatedly, is there an ATUS variable for spouse/partner employment equivalent to ‘CLWKR’ for respondents? If not, what is most comparable variable?
Thanks much, SG
Please ignore, I just realized blanks are for partners earnings not updated at ATUS. Please see my other question about getting earnings for those partners from CPS.
The variable SPEARNWEEK comes from ATUS respondents’ spouse’s earnings as recorded in the final CPS interview. The ATUS is fielded to a subset of outgoing CPS households, and it is possible to link ATUS respondents to their CPS interview responses. The ATUS offers a number of variables from the CPS, such as EARNWEEK_CPS8, which reports an individual’s weekly earnings as recorded in their eighth (final) CPS interview. For ATUS respondents who changed jobs or employers since the final CPS interview, EARNWEEK was updated at the time of the ATUS interview to reflect their new earnings level. For ATUS respondents who did not change jobs or employers since the final CPS interview, EARNWEEK is equal to EARNWEEK_CPS8. The variable EARNWEEK_CPS8 is available for all people who responded to the CPS (and who were employed and not self-employed during the final CPS interview), while EARNWEEK is only available for ATUS respondents.
The variable SPEARNWEEK is created by IPUMS and it simply reports, for ATUS respondents only, the value of EARNWEEK_CPS8 of their spouse or partner. Weekly earnings information is not updated for non-ATUS respondents at the time of the ATUS interview (note that there is only one respondent per household in the ATUS). SPEARNWEEK is missing when EARNWEEK_CPS8 is missing for an ATUS respondent’s spouse. EARNWEEK_CPS8 is missing when an individual in the CPS was not employed, was self-employed, or was working without pay at the time of the final CPS interview. Because SPEARNWEEK comes from the CPS, and because it is only updated at the time of the ATUS for ATUS respondents themselves, you cannot gain more information about ATUS respondents’ spouses’ earnings by using EARNWEEK_CPS8.
You may choose to link ATUS respondents’ spouses to earlier CPS interviews (those from months in sample 1-7) and gain information about their past earnings, if they had any. Whether and how to use that information is up to the individual researcher. To link an individual in the ATUS to their CPS responses, you can simply use the variable CPSIDP from IPUMS ATUS and IPUMS CPS to uniquely identify a person across all CPS interviews and their ATUS record.