tract: 55117010501; my extract is #95 if you can ref that.
so i’m looking at wisc, my extract has a number of variables associated w/ social determinants of health. it’s 5 year ACS data; and a portion of my tracts are ‘empty’ (i think it’s 2017-2021). i just listed one tract for ref - it’s a fairly decent % of the tracts that are empty.
i thought that you get data for 5 year samples for most geographies. thoughts?
i also tried this for diff 5 year samples to see if something had changed.
Tract 55117010501 is a 2010 census tract. 2010 census tracts are used in 2010 census data and in ACS data release years 2010 through 2019. The 2017-2021 5-year ACS data use 2020 census tracts.
It sounds like you might have joined tract data from the 2017-2021 5-year ACS to a shapefile representing 2010 census tracts. This would result in ‘empty’ data for any 2010 census tract whose ID doesn’t match a 2020 census tract.
To map 2017-2021 ACS tract data, you could use a 2021 tract shapefile. If you’d like to do direct comparisons of tract data from before and after the 2010-2020 definition changes in tracts, you could use NHGIS crosswalks to produce geographically standardized estimates for either 2010 tracts or 2020 tracts.
ok, i figured out i think - although the data is for these later years, it appears that they are mapped to the previous census; so 55117010501 didn’t exist in earlier years; there was just 55117010500; but what i don’t understand is that there is data for 55117010502 but no 55117010501? thoughts?
I just checked an extract of tract data from 2015_2019_ACS5b and I see a record for tract 55117010501. The Census Bureau structured the tract ID codes differently in that release, so the GEOID for that record is “14000US55117010501” and the NHGIS GISJOIN is “G5501170010501”.