Hello! I am part of a research team that is exploring the use of CPS data as denominator data for industry, occupation, and some demographic variables for a single state (and looking at trends across 5 years; 2019-2023). Despite using the broadest industry and occupations variables available (A_MJIND and A_DTOCC from the March Supplement) we noticed that there are significant fluctuations in population totals from year to year. Is there any guidance on how to manage this issue? Is this a function of the survey design, changes in how industries and occupations are classified from year to year, the process of looking at small units (e.g., a specific industry within a single state), a combination of all of these factors, or something else? It looks like IPUMS provides data harmonization (to address issues of how variables change over time), but we’re really trying to get a handle on the best approach to obtaining denominator data (and whether the ACS might be preferable). Any thoughts or feedback is welcome. Thank you!