I am trying to determine what share of persons in the labor force were in different industries at the state level from the year 1979 to 2013. The IND1990 variable appears to provide a comparable classification over time, but it appears to only be available for the March samples from 2011-2014 and 1979-1988. Since I am looking at the state level, I’m not sure that the March sample sizes alone will be sufficient for those years depending on the level of industry classification I am looking at.
Is this one of the variables that just hasn’t been added yet to the other samples? And if that is the case, is there a way to calculate the IND1990 values for other samples myself or might you be aware of another source that might have a consistent industry classification over time? So far, IPUMS has been the only one that I have found.
Thank you so much for your help. It is greatly appreciated.
The IND1990 variable is available for all of the IPUMS-CPS samples. IND1990 is only available in the March samples for 2011-2014 and 1962-1988 because IPUMS-CPS does not currently offer any other samples from those years. IPUMS-CPS is working to harmonize and provide all of the monthly samples from these years, and has already dramatically increased the number of available samples during this past year.
The unharmonized versions of the CPS files can downloaded from NBER or from the Census Bureau’s FTP website. Using the codebooks from the samples you are using, you should be able to determine which Census Industry Coding Scheme was used in that sample. You can then find an IPUMS-CPS sample that uses the same Census Industry Coding Scheme and download an extract containing both the IND and IND1990 variables to make a crosswalk.