This document from the Census caught my eye today:
CPS households are interviewed for 8 months. The household identification
(HRHHID) variable can be used to identify prior interviews from the same
household. In order to ensure that geographic information is suppressed as
outlined above, the household identifier variable will be renumbered starting with
data for January 2022. Thus, 2022 data for a household will not be able to be
matched to any 2021 data.
Am I correct in assuming that this will affect IPUMS’s variables CPSID & CPSIDP and make 2022 unlinkable with 2021?
Thanks for bringing this to our attention. Based on this documentation, the change in household identifiers that accompanies the change in available geographic detail will make linking CPS months between 2021 and 2022 impossible. CPSID and CPSIDP are based on the Census Bureau household identifiers, and when those are altered to intentionally prevent linking, we are unable to circumvent those changes with CPSID/P. We do not currently have any further information about whether there will be further disruptions to linkability in the 2022 data. We will document these changes and their impact on the panel component as data becomes available and will include any new information we learn about this issue. I encourage you to reach out to the Census Bureau at the email address (DSD.CPS@CENSUS.GOV) from the document you shared with further questions.