March 17, 2006

More on genealogical DNA testing

The point I wanted to make about paying for DNA testing to understand the racial admixture in your genealogy is that at present it's mostly useful for, say, African-American intellectuals like Harvard's Henry Louis Gates, who discovered he was about half white. Most white Americans, on the other hand, are over 95% white, and thus their nonwhite proportions are down in the margin of error for the test. For example, if the test reports that you are 98% Caucasian and 2% American Indian then you can be assured of, well, not much. You might have had a few Indian ancestors, or that might be an error caused by Siberian genes migrating into Europe, or who knows what else. So, the reason that racial admixture tests tend to be not very exciting for white Americans is because, in contrast to what we're constantly being lectured, white Americans aren't very admixed at all.

However, the tests are quite good at looking at large sample sizes of peoples. So, if they say that white Americans are x% American Indian on the whole, then that's pretty reliable.

My published articles are archived at -- Steve Sailer

No comments: