Tract Crosswalk Error for iPUMS and Census?

Hi, in your 2010-2020 block group to tract crosswalk here: https://www.nhgis.org/geographic-crosswalks, there appears to be an error. I found the same error here: https://www2.census.gov/geo/docs/maps-data/data/rel2020/tract/tab20_tract20_tract10_natl.txt. Namely there are some CA, OR, and NV tracts that get crosswalked to each other. See for instance tracts 06049000200 and 41035970500. This should be impossible, correct? Thanks.

Such cases of “state boundary change” are not errors in the crosswalk. To the best of my understanding, they reflect improvements that the Census Bureau made in its representations of these boundaries between censuses. To the extent there is an error here, it is in how the Bureau represented the state boundary line in 2010, which it improved in 2020. But the 2010 representation reflects accurately the boundaries that the Bureau used for 2010 tabulations. If a single housing unit was located on one side in 2010 and the other in 2020, it would have been counted in different states in each year. Therefore the crosswalk accurately reflects how 2010 data should be adjusted to correspond with the 2020 data.

The NHGIS crosswalks page has a note about this here:

Each crosswalk file is complete for the entire U.S. or for an entire state. State-level files include all target zones for the state as well as any source zones that intersect any of those target zones, including some source zones from neighboring states in cases where the Census Bureau adjusted state boundary lines between censuses.

Jonathan - thanks very much for this reply - it’s very, very helpful.