In the span of just a few days and just weeks before official census redistricting data is scheduled to be released, the Census Bureau reveals two rather startling revelations about the data.
“Fuzzy Blocks”
First, the Acting Census Director Ron Jarmin warned in the Director’s Blog that the information in individual census blocks – the smallest and most detailed unit of data – will appear “fuzzy” and at times nonsensical due to the Bureau’s application of privacy measures, and thus should not be used for building redistricting maps. Blocks are often used as the basic building blocks of redistricting maps because they allow for the most precision in equalizing the population between districts, but drawing congressional districts and minority districts will likely be more difficult if not impossible to draw in some cases if map-drawers heed Jarmin’s warning. An excerpt from Jarmin’s blog entry is below:
“census blocks may look “fuzzy,” meaning that the data for a particular block may not seem correct. Importantly, our approach yields high-quality data as users combine these “fuzzy” blocks to form more significant geographic units like census tracts, cities, voting districts, counties, and American Indian/Alaska Native tribal areas. Our calibration was designed to achieve acceptable quality thresholds for these levels of geography.
So, if you’re looking at block-level data, you may notice situations like the following:
- Occupancy status doesn’t match population counts. Some blocks may show that the housing units are all occupied, but the population count is zero. Other blocks may show the reverse: the housing units are vacant, but the population count is greater than zero.
- Children appear to live alone. Some blocks may show a population count for people under age 18 but show no people age 18 and older.
- Households appear unusually large. For example, you may find blocks with 45 people, but only three housing units.
Though unusual, situations like these in the data help confirm that confidentiality is being protected. Noise in the block-level data will require a shift in how some data users typically approach using these census data.
Instead of looking for precision in an individual block, we strongly encourage data users to aggregate, or group, blocks together. As blocks are grouped together, the fuzziness disappears. And when you step back with more blocks in view, the details add together and make a sharp picture. “
Dramatic Group Quarters Undercount
Over the weekend, as part of ongoing litigation, the Bureau released a document dump of over 2,000 pages relating to its handling of significant irregularities in the count of group quarters facilities such as college, prisons and nursing homes. These documents reveal that at least 20% of these facilities reported zero population when in fact there were people living in them on census day. Court documents confirm that the Bureau” decided to use some sort of “count imputation” procedures on unresolved group quarters, including for missing characteristics such as age or race, that it “had never before conducted” for group quarters previously.” In short, the Bureau faced unprecedented difficulty in the Spring of 2020 and its attempts to rectify it were apparently unprecedented as well. The documents include internal reports, meeting minutes and presentations all aimed at developing a strategy for dealing with the group quarters count.
According to the Bureau, “The item nonresponse rates, along with imputation rates, are types of response quality measures.” The graph below compares the relationship item nonresponse rates from the 2010 census to the 2020 census and shows the stark severity of the problem. Many census stakeholders have voiced concern over the possibility of a significant census undercount even before these documents were released.
Additional slides (below) from Bureau internal discussions shed some light on the gravity of the problem the Bureau faced in the Fall of 2020. The entirety of the Bureau’s FOIA release can be viewed in the motion for a preliminary injunction document.
Internal Census Documents Regarding Group Quarters Count
