I understand prior to 2017 the best way to link the March basic file with ASEC data was to merge by MARBASECIDP. However, the variable is no longer available in the March basic file from 2018 forward (IPUMS CPS: descr: MARBASECIDP). I just tried to merge 2018 data using CPSIDP but there are many observations from the March basic file that did not merge to the ASEC (which surprises me, as I understood the ASEC to include all March respondents as well as some additional respondents).
What do you recommend? Thank you for the help!
You’re right to use CPSIDP for this linking, and your understanding is generally correct. However in 2016 and 2017 there were some people who received different forms of the health insurance questions. These “split path” people are not matchable between March Basic and ASEC. This might also be the case in 2018 but it is not totally clear at the moment. The IPUMS CPS team is looking into this issue right now. I’ll post an update when we know more.
Great – I appreciate the response. Thanks!
Any update on this? I’m currently having this issue with the link between march basic and the ASEC for 2016-2018. Is there any way to confirm which individual are failing as a result of “split path” rather than some other reason?
No update at this point. I’d assume all the unlinkable people are from the split path issue.