I am interested in replicating the racesing variable for the 2018 IPUMS ACS data. The IPUMS Stata program produces results for 2014 that are very close to the original racesing variable, so the program seems to be working.
My question is if there is a reason that IPUMS has not updated the racesing variable, if it’s just a matter of running the Stata program? In other words, is there a reason that I should be worry about applying the program to the 2018 ACS data?
The core issue with making the RACESING variable available in samples beyond 2014 relates to two general concerns. First, the Stata program was designed using the 2000 Census data, so with broad demographic changes in the past 20 years, it is difficult to assess how valid these single race predictions are in more recent samples. Second, changes to restricted use identifiable geographic areas prevented the IPUMS USA Team from officially releasing this variable and servicing it in years beyond 2014. With that said, if you are willing to take the assumptions within Stata program as they are, the program should run on the 2018 ACS public use data.
That is helpful context. Thank you for your response.