Why they’re not better… What they’re good for
In case you’ve lived on the Dark Side of the moon for the past few years, here’s a “news” flash: the Legal Genealogist is not a fan of ethnicity estimates.
Those percentages reported by the DNA testing companies — x percent Irish and y percent Scandinavian — are not scientifically valid representations of our exact ethnic heritage.
I’ve said this before.
Others have, too.
Now, read this post, by a an expert in population genetics, if you want to understand more about why these percentages aren’t better … and what they really might help tell us, if we understand their limits.
Dr. Joe Pickrell is a Junior Group Leader and Core Member at the New York Genome Center, and an Adjunct Assistant Professor in the Department of Biological Sciences at Columbia University. He has a PhD in human genetics from the University of Chicago and a BS in biology from the University of North Carolina-Chapel Hill. He’s also a co-founder of the new research website DNA.Land, and he’s the author of the blog post I want you to read.
And if you read nothing else, read this part:
…what you would ideally like to have is a detailed list of your ancestors at different time depths, each labeled with their geographic location and any ethnic self-identifiers. You could then say, for example, that 100 years ago 25% of your ancestors lived in Illinois and identified as Jewish, while 500 years ago 5% of your ancestors lived in present-day Andalucia and identified as Muslim.
Unfortunately genetic tests are about as useful as Ouija boards for obtaining much of this information, so we’re going to have to compromise with some dramatic approximations. Specifically, the approach taken by all of the commercial companies (and that we take as well) is to try to estimate the general geographic regions where your ancestors lived (and in a select small number of cases their ethnic identifiers) some indeterminate time in that past, probably something like a few hundred years ago.
Does this all sound a bit vague? It should because it is. The precision suggested by these reports is an illusion–there’s plenty of wiggle room in the definition of “general geographic regions” and “some indeterminate time in the past” to allow for very different interpretations.
Compromise. Estimate. Wiggle room.
NOT precise scientific analysis.
Estimates, not real percentages.
Not useless, since there are some things you can do even with these generalized data points. But not, not, NOT precise numbers.
Thanks for this, Judy. What gets me as I think about this issue is that many of us are handing the companies actionable data every day. I have, on Family Tree DNA, a test for my mother, a test for my father, and a test for me. These are all original tests done directly with FTDNA. So if you think about it, no one knows better than FTDNA who my parents are – the tests proved conclusively that they are indeed my parents. So WHY do they require so little of themselves as to give us the following percentages:
Mom – 45% British Isles, 39% Scandinavia, 16% Southern Europe
Dad – 100% British Isles
Me – 84% British Isles, 16% Western & Central Europe
If this doesn’t prove your case, I don’t know what does. You are so right, people need a grain of salt, and possibly a whole shaker.
I personally vote for an entire salt lick!
Don’t you think that maybe this Pickrell caution applies to more than just ancestral percentages? In particular, these ‘cousin finder’ tests that produce long lists of potential ‘cousins’ based on a statistical pattern matching algorithm? I was following two DNA forums for a while, and a distressing number of people seem convinced that the presence of tiny sequences of similar DNA sequences for two people ‘proves’ common ancestry. I for one would be interested in hearing what Dr. Pickrell thinks about these DNA tests. I know just enough statistics and probability to be dangerous, but it seems to me that the genealogy DNA companies might be promising- and getting paid for- way more than they can reliably deliver.
But then again, I did warn you- I know just enough to be dangerous 😉
There’s no doubt that some matches that are reported are false positives, Jim. Those very small segment matches reported as moderate confidence are — well — doubtful to some degree. But the matches with significant segment lengths and overall DNA in common (except in the worst of endogamous populations) should stand the test of time pretty well.
At least for my families, endogamy seems much more common than most people seemed to think it would be. As I slowly untangle all the paper trails, on my mother’s side many people are related two to three different ways. My father’s line is less intense, but still it is not unusual to discover distant cousin relationships. I wonder if we have properly factored endogamy into the equation.
Rural South Carolina was- until the 1940’s- peppered with more or less isolated communities. Bad roads and agricultural work limited access to potential mates. That was not the case for towns and cities, of course, but in my families, even when they did have access to wider populations in urban centers, they often seemed to prefer selecting partners “back home” well into recent times.
All of us with colonial roots are going to find those roots are a lot more twisted than we might have thought, Jim: cousins marrying cousins because they were the only ones within courting distance!!
As I keep saying, it’s a marketing tool for the testing companies.
Yep. You got it.
These are merely clues as are most genealogy finds. It is up to the individual to follow through and do the real detective work. Does anyone truly expect verifiable accuracy in these percentages?
Yeah, actually, some people really do believe these numbers — or want to believe them. Sigh…
Yes, unfortunately they do and are one of the main reasons many people take the tests. That is why I am so glad you continue to point out this issue in your blog. It seems it can’t be said enough.
And yet some of those who test are interested only in the ethnicity. I keep running into those individuals who share some level of DNA with me who don’t have trees and don’t really care to explore possible connections. They just wanted to know “what” they are. They certainly spent a lot of money for an estimate, but I don’t think they understand that.
No… sigh… they don’t understand it at all.