[BostonCard]

80% of the UK is white, so if the group of people who didn’t state their race in the study were more or less randomly distributed (it of course is not, but you can expect the majority of that group to have been white), then the group that didn’t state its ethnicity to look pretty similar to the white group.  The estimate was probably pretty close, and depending on the number of people who declined to state a race, the difference may not have been statistically significant.  I have no insight behind the model, but it wouldn’t surprise me if they collapsed terms that weren’t statistically significan so that internally “no race” is treated the same as “white”.

BC

[BostonCard]

80% of the UK is white, so if the group of people who didn’t state their race in the study were more or less randomly distributed (it of course is not, but you can expect the majority of that group to have been white), then the group that didn’t state its ethnicity to look pretty similar to the white group.  The estimate was probably pretty close, and depending on the number of people who declined to state a race, the difference may not have been statistically significant.  I have no insight behind the model, but it wouldn’t surprise me if they collapsed terms that weren’t statistically significan so that internally “no race” is treated the same as “white”.

BC