Looking for an income variable that I can use, using monthly CPS data. From 92-12 I have 9.9 million responses for age 50+, but earnweek is only in universe for 892,000 and hourwage is only in universe for 489,000. I feel like I should have more since labforce gives me 4.5 million responses in labor force for this group. Something like Faminc (combined family income, groups of 5,000) which is asked in the survey would be nice, but I can’t find the variable that matches that in the IPUMS data. Can anyone provide a good income variable or tell me how to get the data from Faminc?
Income information in non-ASEC samples is sparse. EARNWEEK and HOURWAGE are part of the Outgoing Rotation Group/Earner Study, which means only respondents in sample months 4 or 8 are eligible to respond. This is why you are seeing a large number of NIUs.
Since CPS respondents are in the sample for 4 months, leave for 8 months, and then re-enter for a final 4 months, some respondents will also appear in the ASEC. You can match respondents between these basic monthly samples and the ASEC by using HRHHID and HRHHID2 to link households along with LINENO, SEX, and AGE to link respondents. Once linked, you can use the income variables available in the March ASEC.
Finally, as you mention, there is a categorical family income variable (hufaminc) available in the basic monthly samples; however, it is not currently available via IPUMS-CPS. See this previous answer for information on linking your IPUMS-CPS dataset to the source NBER data. If you choose to use hufaminc, it will be available for all respondents in your basic monthly samples.
Hope this helps.
Thank you Tim, I appreciate your help! The number of responses makes much more sense now.