Assuming the average for white Hispanic is similar to the white population overall, aren't you then implying that the number for non-white Hispanic should be even higher since white Hispanic would be bringing down the average?
You are mostly wrong. The U.S. Census, which is the standard setter in this area, draws a clear distinction between race and ethnicity [1], where ethnicity is defined as Hispanic or Latino, or Not Hispanic or Latino. Any race may identify as either.
For example, the Wikipedia to which GP links sources the relevant data from the County Health Rankings (a project run by U of W and Robert Wood Johnson Foundation). The technical documents describe the distinction between Race and Ethnicity, which matches the Census [2].
The Census' approach has apparently not been ideal for a lot of people, and it looks like it may be changing to just have a combined question for race and ethnicity. The problem is that the Census insists that Latino or Hispanic can only be an ethnicity, which doesn't really match up with a lot of people's experiences.
> If approved, the changes would address longstanding difficulties many Latinos have had in answering a question about race that does not include a response option for Hispanic or Latino, which the federal government recognizes only as an ethnicity that can be of any race.
There are white Mexicans, black Mexicans, and Indian/mixed Mexicans. I’m not sure if you would consider any of them to be Hispanic/Latino, but the ethnic groups exist in Mexico (and other Latin American countries surely), so isn’t it weird that they are collapsed into one ethnicity in the United States? Maybe some weird aspect of immigration physics I’m not considering? See:
> 20%+ of Hispanic/Latino people in the U.S. identify as white.
When asked a question that doesn't have Hispanic as an option. As another comment mentioned, the Biden administration has proposed changing that question (by including a Hispanic option) to better reflect people's identities.
