Hello,
I would like to combine CPS-ASEC, including 2020, into a multiple-year sample for a person-level analysis. Would that be an issue? If I can combine data across years including 2020, what weight should I use to analyze the data?
PS: I saw a similar post but was unsure about including the 2020 sample.
Thanks a lot!
It is possible to include the 2020 ASEC in a multi-year sample, if the proper weights are applied. Because of the very unusual conditions of March 2020, the Census Bureau released a special set of weights just for the 2020 ASEC. These can be found in the variable ASECWTCVD. The paper in the linked documentation describes the rationale for these weights and the procedure used to construct them. The proper way to incorporate these weights into a pooled sample would be to create a new weight variable, which takes the value of ASECWTCVD in 2020 and ASECWT in all other years.
Thanks so much!