I’m hoping to download data for Ottawa County Michigan for 1920. Are the personal names coded out? I’m trying to answer these 2 research questions and wondering if IPUMS is a solution of if I need to continue looking:
A current and real case researcher would like to use the 1920 U.S. Census for Ottawa County and the City of Holland to research World War I veterans living at home due to injuries sustained during the war. Other researchers, like me, would like to be able to look at particular areas like Ottawa County or local cities to look for particular patterns, for instance boat company employees, factory workers, etc. Thank you!
There are a couple ways to do this. I think the easiest way is to use the household income variable HHINCOME and then reduce your dataset to only one person per household by keeping only PERNUM==1. There is also a (very rarely used) extract data structure option for “Household records only”. From the Extract Request page, you can change the “STRUCTURE” field to be “Household records only”. Either option should work fine.
I hope this is helpful.
A couple notes: First, in general personal names are not included in public use data. This is due to confidentiality purposes. IPUMS does have a limited number of licences available for restricted access data that does include personal names. For more information about this data and how to apply for access, see this page. An exception to this rule is with historical samples (like 1920). See the NAMEFRST and NAMELAST variables. Second, there is a WWI veteran status variable (VETWW1), but it is not available in the 1920 samples. Third, a variable relating to occupation (OCC) is available in the 1920 1% sample, but may not include the level of specification you are looking for (i.e. boat company employees). The codes used for OCC in 1920 are available here.
I hope these notes are helpful.