When matching data from IPUMS NHGIS tables to the GIS files, I noticed that for the Census 1990, what seems like uninhabited blocks (things like city parks, train stations, ponds, etc) just don’t appear at all in the table data, whereas these uninhabited blocks do show up in the Census 2000 and 2010 tables with 0 population. Can someone confirm that uninhabited blocks don’t seem to be included in the Census 1990 data tables (total population or race by age, specifically)? I just want to make sure that I can replace these missing census blocks as having 0 population and that I’m not missing some other issue.
Correct. The 1990 summary files omitted records for blocks that had 0 population and 0 housing units. (There are some blocks included in the 1990 tables that have 0 population, but that’s only because they have nonzero counts of housing units–all vacant.) You can safely assume that any 1990 block that appears in the GIS file and not in the tables has 0 population and 0 housing units.
Great! Thanks so much for your help.