Ancestry DNA told me my ethnicity. Then it kept changing its mind.

new estimate

This is really irritating.

It’s especially irritating because I can’t get back to the original ethnicity estimate that Ancestry gave me after I sent them a vial full of spit.

That one made sense, and I thought I understood it. It went something like:

  • 45 percent Western European
  • 20+ percent from England, Wales and Scotland
  • 20+ percent Irish
  • 5 or 6 percent Scandinavian
  • Something like less than 1 percent from Southern or Eastern Europe — or was it Iberia? (I don’t know because there have been several versions since then).

I thought that gave me a pretty good sense of how my ancestors were distributed on my tree, which now has more than 7,000 people on it.

As for that big number from Western Europe — it wasn’t that I had a lot of people come to this country from France, Belgium, Germany and the like. No, the vast majority of people I’d been able to trace came here from England, and the rest from Scotland and Ireland. But a lot of those from England had ancestors from Western Europe, if I was lucky enough to trace them back a few more centuries.

And yes, as crfazy as it may sound, I was able to follow quite a few lines well back into Medieval times, and some (only two or three) to before the Norman Conquest. And it turns out a lot were Normans, and some were Welsh, but I’ve yet to find anyone I can point to and say yes, that’s a Saxon. And that means a lot of my English ancestors had nonEnglish ancestors.

And I was able to get a few of those Normans back to Scandinavia, which is where Normans came from. Of course, those were so far back they were highly dubious but it seemed consistent with what the DNA said.

So that first “ethnicity estimate” made some sense.

But since then, Ancestry has redefined its ethnic groupings, and scrambled everything up. Two or three times. Above you see the latest version.

Since when is “England, Wales and Northwestern Europe” a single ethnicity? That’s so broad and general that yeah, I’m unsurprised that I’m 65 percent that. But what does it tell me, really? pretty much nothing.

And look, below, at a closeup of that distribution:


OK, that’s England, sure enough — all but a bit of Cornwall. And that’s most of Wales, all right. But as for the “Northwestern Europe” part — say what? The Cotentin peninsula, plus another little bitty slice of Northwestern France that runs a bit south of Calais? That’s IT? How is THAT “Northwestern Europe?” You should call it “England and Wales (and a few people who lived close enough across the Channel that they probably crossed back and forth a lot).”


And why did Scandinavia go in the other direction, getting more specific, narrowing down to Sweden?

And what on Earth is “Germanic Europe,” If it leaves out most of Austria?


This is maddening.

I suspect Ancestry just has too many employees, and they need something to do, so they are assigned to frequently scramble and rescramble the “ethnicities.”

But each time they do it, they get drunk first.

37 thoughts on “Ancestry DNA told me my ethnicity. Then it kept changing its mind.

  1. Doug T

    I envy you for being able to trace that far back. I can’t find my grandfather for goodness sake. I did find a string of my wife’s lineage going back to 1600’s Scotland.

    1. Brad Warthen Post author

      The shortest line I have is the one that ends at Henry Waller, a great-great grandfather — my mother’s father’s mother’s father. He died at Petersburg in 1864 when my great-grandmother was a baby, and I know nothing about his ancestors.

      Then there are a couple — the Warthen line and the Bradley line — that I can get back to England and Ireland, respectively, and then that’s it.

      But plenty of other lines I can get well back in the Old Country.

      It’s frustrating…

    2. Rrr

      I agree with you. Luckily, I was smart enough to never use their services, but a few family members have. I find the results laughable on how they change. I called attention to it as well. No way you can test for a category, that territory was never changed by company, but yet the 2nd test result completely removed Sweden Denmark portion of DNA results. Now, 1% of DNA goes back 7 generation, so at minimum you should have at least 1% of what your grandparents were, but yet on the test results I seen not a piece of our maternal grandma was in there. As I said, I find these companies to be a joke and a big scam to give up rights unknown to person using services.

  2. Mab

    Drunk minds may be redistricting, but criminal minds can frame you with your spittle.

    I can’t believe intelligent folks fall for this. How much do you people pay for this? SMH.

  3. Mr. Smith

    Off topic, but still about, ah-hem, “heritage”:

    From the WPost:

    “Former South Carolina governor Nikki Haley told conservative radio host Glenn Beck on Friday that the Confederate flag represented ‘service, sacrifice and heritage’ for people in her state before mass murderer Dylann Roof “hijacked” its meaning.”

    As the descendant of at least one reb vet, that’s not what it means to me. I do not thank them for their service. It seems she’s reversed course on this to try to circle back to pandering to the die-hard Confederate apologists. Roof didn’t “hijack” anything. Haley again demonstrates her shallowness.

    1. Scout

      I am a descendant of multiple confederate vets. I don’t blame them for their service. They are a product of a different time and culture. I do think people today who can’t recognize and acknowledge the negative connotations of the flag are just tone deaf. I don’t need to honor that symbol to be able to respect my ancestors. I believe you may be right that she is trying to pander to that crowd again.

    2. Mark Stewart

      I took it as a sign she has pulled back from national aspirations. Seemed more calculated o her part than just an unintentional revelation. Is Joe Wilson retiring? Or is Haley eyeing primarying him? Or Lindsey?

    3. Brad Warthen Post author

      Here’s the thing about Nikki — she’s not a deep thinker.

      She makes a great impression, and seems to have pretty good emotional intelligence, in terms of reading a crowd and knowing what to say to make them like her.

      She even comes across as bright. I imagine she did fairly well at school — probably better than I did, since I was such a slacker.

      But you won’t see her diving deep into something as complicated as the Confederacy.

      And the Confederacy IS complicated, despite what the neo-Confederates AND their “woke” adversaries might like to believe.

      There are certain “simple” statements that are true. For instance, despite the obfuscation you get from the Confederacy’s defenders, the war — AND the Confederacy — were about slavery. Trying to make it out to be about something other than slavery — say, tariffs — requires a willful blindness. Slavery had been THE dividing line since the nation’s founding, and secession and the war were the moment when the conflict came to its inevitable bloody head.

      Does that mean each Confederate soldier was a conscious defender of slavery? Of course not. Wars don’t work like that. Personal motivations are complex. A huge percentage of soldiers go off to war in large part because that was what all their friends and neighbors were doing. Plenty saw themselves as defending their homes from Yankee invaders. People had a very narrow and archaic notion of what “their country” was.

      And when they got to the front lines, they fought for the same motivation as soldiers today do — they fought for the men next to them, knowing that the men next to them were fighting for THEM.

      For those reasons I agree with what Scout says: “I am a descendant of multiple confederate vets. I don’t blame them for their service.”

      I’m not sure how many direct ancestors fought in the war on the Southern side. I suppose I should try to count them up. But I know two direct ancestors died at Petersburg: My great-great grandfather Henry Waller and my great-great-GREAT grandfather Wesley Samuel Foxworth. And I lost a great-great uncle, Thomas Chiles Bradley, killed at Trevillian Station. (By the way, his grave marker is wrong. He was NOT killed at Balls Bluff in 1861. He died almost three years later. There are existing letters from him after 1861)

      I couldn’t begin to tell you what motivated them.

      All I know is that I, personally, utterly reject the Confederate clause, and any symbol that unmistakably invokes it.

      And I condemn Nikki’s nonsensical statement. What I can’t tell you for sure is whether it is calculated, or simply a case of her not having a clue what she’s talking about. I suspect both are involved…

      1. Mr. Smith

        That all sounds so hunky dory.

        But no one said anything about blame.

        The issue here is: Are we to thank our reb ancestors for their service to the cause?
        If so, then we join ranks with the so-called Confederate heritage defenders.
        If not, then we can have no truck with the memorial culture they defend.
        Because the cause is what the Confederate memorial culture is all about.
        There is no fudging, no tidying it up to make it palatable to all and sundry.

        1. Brad Warthen Post author

          Yes, and I would have thought that my comment was perfectly compatible with what you just said.

          “Are we to thank our reb ancestors for their service to the cause?” Of course not. I made that clear in my penultimate graf.

          And no matter what one believes, there is, as you say, “no tidying it up to make it palatable to all and sundry.”

          Nikki wanted to cozy up to someone, and she decided to cozy up to the people who believe (or say they believe) the flag was, as you say, hunky-dory until Dylann Roof came along. By necessity, when you do that, you poke other people in the eye.

            1. Brad Warthen Post author

              I have no idea where you’re coming from on that. This is a blog in which each of us expresses what we think about something. I responded to what Scout thinks by saying what I think. Presumably you are sharing your own thoughts as well.

              How does that make everything about me? (Aside from the fact, of course, that this is MY blog. It’s got my name on it and everything.)

              And again I wonder, what is it I did at some point in the past that causes you to always have this slight edge of antagonism toward the things I say? I seem to irritate you easily.

              Or, in keeping with your theme that it’s not about me, does EVERYONE irritate you that easily?

              1. Mr. Smith

                You may have meant to respond to Scout, but if you’ll follow how the comments flow you’ll see that you actually responded to my comment. The rest follows from that.

                The Confederacy isn’t really that complicated. And there’s no reason to let the individual Confederate soldier off the hook by saying he my not have been “consciously defending slavery.” Wars just don’t work that way. Personal motivations don’t count for squat. Once you lend yourself to “the cause,” you are part of it, whether you like it or not, whether you consciously embrace it or you don’t.

                1. Bill

                  It’s not complicated.In the end ,no on is right or wrong but all are forgiven:
                  Down the street the dogs are barkin’
                  And the day is a-gettin’ dark
                  As the night comes in a-fallin’
                  The dogs’ll lose their bark
                  An’ the silent night will shatter
                  From the sounds inside my mind
                  Yes, I’m one too many mornings
                  And a thousand miles behind
                  From the crossroads of my doorstep
                  My eyes they start to fade
                  And I turn my head back to the room
                  Where my love and I have laid
                  An’ I gaze back to the street
                  The sidewalk and the sign
                  And I’m one too many mornings
                  An’ a thousand miles behind
                  It’s a restless hungry feeling
                  That don’t mean no one no good
                  When ev’rything I’m a-sayin’
                  You can say it just as good
                  You’re right from your side
                  I’m right from mine
                  We’re both just one too many mornings
                  An’ a thousand miles behind

                2. Brad Warthen Post author

                  Everything about human beings is complicated.

                  As I said, some things are pretty straightforward — such as the fact that the Confederacy was about slavery.

                  But people, and their motivations, are complicated. That’s why I can never feel comfortable on the left or the right, because the members of those tribes try to oversimplify things…

                3. Mr. Smith

                  Then you must be opposed to all public memorials — because they, too, simplify things since they can never adequately cover all the varied motivations and actions of the individuals involved in whatever is being memorialized.

                  1. Brad Warthen Post author

                    If they simplified things all that much, we wouldn’t have all these terrific arguments over taking them down.

                    Visual art is much vaguer than the written word. It gives everyone license to say it means what THEY want it to mean.

                    A bumper sticker is simple. A statue is open to interpretation…

  4. Scout

    Mine changed too. Now I’m 100% England, Wales, and Northwestern Europe. I used to have some Ireland and Scotland and a tiny bit of Scandinavia. They say that they refine their estimates as they get more people in their database and that makes it more accurate. But I’d like to know more about this process. I mean even with more datapoints, if those new people also don’t know where they came from – what is the control or standard that they are using to revise and say, oh this group of similar people came from here, not there. Are they comparing to the DNA of the current populations of those places? Or do they have some people with ironclad proven family trees that they are assuming are a standard for a certain region. It intrigues me. But this is not my area. I guess maybe computer modeling can magically sense patterns that figure this out.

    But my parents have also done Ancestry DNA – they each have 15-20% Ireland and Scotland and we know of some of those lines in our tree. So I’m a bit confused how I got none, but I guess it’s possible.

    What are you using to prove connections that far back – do you just accept what other people have in their trees? Or are you finding actual records that far back? I’ve not paid for access to European records yet, so I don’t know what’s available. I’ll be busy for awhile just proving my lines back to the Immigrant within the US. So far my most recent Immigrant ancestor was in the 1830s, so my people have been here awhile.

    The most interesting thing we’ve found is something a cousin figured out. He has done Y DNA on my grandmother’s paternal line – Flanders. He is a Flanders and found that his Y DNA does not go back to other Flanders, as it should, but is still related to that Flanders line, just not paternally. He has figured out the generation where the oldest daughter had a child (my third great grandfather) out of wedlock at 15 and he was raised by his grandparents as a sibling of his mother. The actual paternal line appears to be Ludwig from Germany. So there is my Northwestern Europe – along with some other German and French lines we also know about.

    I’ve also done 23 and me – they also revise their estimates as they get more data. They currently say I’m 98% northwestern European, of which they can specify within that: 57% British and Irish, 27% French and German. And also 1% Southern Europe and 1% unspecified. This is believable to me.

    I’m rambling but it is interesting.

  5. Brad Warthen Post author

    Yes, it is — to me, anyway — and it can cause one to ramble.

    I’m embarrassed to answer this question: “What are you using to prove connections that far back – do you just accept what other people have in their trees? Or are you finding actual records that far back?”

    I’m sloppy, but not as sloppy as I could be.

    Up to a point, I accept not only what I find in other trees, but in databases such as, wikitree and findagrave.

    If they seem reasonable — if time and place and dates relative to parents or children are a good match — I go with them until I find evidence to the contrary, or a contradiction I can’t resolve.

    I’m fatalistic about it. I know that absent physical documents, no one on the planet can tell me about generations that precede the memories of people now living.

    So, knowing a given link can’t be proven WRONG (or right), I just enjoy the game, and see where I end up by following a streak on the databases, step by step: X was the son of Y who was the daughter of Z and so on.

    I’ll even hang on to a link that I’m almost sure is wrong — such as Betty Crowley — just because I learned so much history following it and hate to ditch all that info. But I keep in mind that I’m almost certainly not descended from Strongbow on THAT line. Although I seem to be descended from him one or two other ways that don’t depend on Betty.

    By the way, cousin, I’m descended from a bunch of people named Flanders, too. The most recent is Matilda Flanders, and I’m REASONABLY confident of it. It goes like this:

    Matilda Flanders 1475-1532
    11th great-grandmother

    Sir Ralph Edward John Mills 1499-1579
    Son of Matilda Flanders

    John Mylles 1520-1612
    Son of Sir Ralph Edward John Mills

    Nicholas Miles 1599-1658
    Son of John Mylles

    John Miles 1653-1697
    Son of Nicholas Miles

    John Miles 1685-1731
    Son of John Miles

    John Miles III 1708-1761
    Son of John Miles

    Eleanor Miles 1756-1811
    Daughter of John Miles III

    John Thomas Rabbitt Jr. 1779-1863
    Son of Eleanor Miles

    Thomas Henry Rabbitt 1820-1890
    Son of John Thomas Rabbitt Jr.

    Rebecca Jane Rabbitt 1863-1898
    Daughter of Thomas Henry Rabbitt

    Gerald Harvey Warthen 1888-1958
    Son of Rebecca Jane Rabbitt

    Gerald Harvey Warthen was my grandfather…

    1. Brad Warthen Post author

      Oh, by the way — I don’t care WHAT kinds of documents you have about your family tree, I place perfect trust in no document.

      Yes, if you have a document that’s more solid than finding the connection on some stranger’s family tree. But it’s still not necessarily the last word. Family connections from three or four hundred years ago are going to be fuzzy whatever you’re looking at. So I accept that, and just have fun with the tree.

      For instance, on another comment I noted where a fundamental fact that is literally CARVED IN STONE was wrong. His gravestone says a great-great uncle died in battle in 1861, when he lived until another battle in 1864.

      How did that happen? Did his family and other contemporaries have nothing to do with that stone? I don’t know.

      But it keeps me from completely trusting ANY form of evidence…

    2. Brad Warthen Post author

      Hmmm, I just noted a couple of fishy things on that Flanders line.

      Nicholas Miles was born when his father was at least 78 years old.

      And Nicholas’ son was born when Nicholas was at least 53.

      And while this is less jarring a couple of others in that line were born when their fathers were in their 40s.

      All of which is possible, but still unusual…

            1. Brad Warthen Post author

              Which is something ELSE that happens. But talk about hard to prove… the legitimate lines are tough enough.

              I thought I’d made a major find — in terms of connecting to people who were easy to trace — when I found myself descended from one Helen Tudor, daughter of Jasper Tudor. This, of course connected me to Henry VIII, et al. (I was already a cousin of Anne Boleyn, through a different set of connections).

              Thing is, though, I wasn’t descended from Jasper’s wife, Catherine Woodville (another cousin, just to keep it complicated), but from a Welsh woman named Myvanwy verch Dafydd. Apparently, she bore two illegitimate daughters to Jasper — Helen (or Ellen) and Joan.

              But still, hey, DNA is DNA. I was still a Tudor by blood. And also directly descended from Jasper’s mother Catherine of Valois — who, in addition to having been wife of Owen Tudor, was earlier married to Henry V (and of course directly descended from a long line of kings of France). She has a couple of good scenes in Shakespeare’s play.

              Then along come the spoilers, saying “no reliable sources” support Jasper having had these natural daughters.

              So… where does that leave me?

    3. Scout

      I do not know my Flanders that far back. My furthest back Flanders is Stephen, born 1620, probably in England, who emigrated to Massachusetts around 1646. My Flanders went from Massachusetts to Maine to Ohio to Kansas to Louisiana. Perhaps we are connected some way.

      I agree about sources. I know even recent presumably solid sources can have errors. I am just really enjoying the Census data, where you can see who was living with whom and what they did for a living and how they moved around, and infer the stories of their lives and then you get back beyond 1850 and it’s like a desert, comparatively. I’m working my way back slowly. I have lots of my geneology on paper from work relatives have done, but I’m entering it slowly and trying to verify it independently if possible. When I get back that far, I’ll see what records are there. I expect I’ll be doing the same as you when I get back that far.

      1. Brad Warthen Post author

        See, you’re serious. You’re like Lynn Teague. She is so conscientious that she keeps her tree private to protect other people from repeating any of her mistakes.

        I just love riding a wave of connections taking me as far back into history as I can go. If a connection turns out to be wrong, I can always fix it later.

        It’s just fun for me. And I hesitate to take it more seriously than I do, because I’m very conscious of the fact that there is usual NO way of confirming connections several centuries and more old. We just never know.

        And suppose I find out that one of my paths back to Charlemagne has a bad link. So what? I know I’m descended from him anyway. All people of European extraction are.

        Actually, it’s a bit inaccurate to say I don’t take it seriously. I do, particular the recent generations, where there is some chance of getting things right and confirming them, up to a point.

        I just don’t let fear of inaccuracies stop me from having fun, particularly when I get back into the Middle Ages. I just love feeling the connection to history, however tenuous. In fact, I’ve ended up learning a good bit of history I didn’t know before, just trying to track people from days of yore. It’s particularly educational when I encounter someone who has a page on Wikipedia.

        For instance, we had dinner at a castle one night in Ireland back in March — it was a thing for tourists, with medieval food and mead served out. And I got to wondering about something, started looking at the Ancesty app on my phone, and sure enough, I was descended from the guy who built and occupied this castle. Actually, not THIS castle, but the one it replaced.

        That made it more fun for me. Part of the shtick was that the people who ran the place picked one of the tourists to be lord of the manor and sit at the head table. The guy they picked not only wasn’t me, but not even a member of our particular tour group. So, I sat there through the meal grumbling into my mead about that usurper sitting up there.

        And for me, that’s fun…

        1. Scout

          Well I’ve done it both ways. There have been times, it’s like a game of suduko gone wrong. I think it’s going great until something absolutely doesn’t fit and have to backtrack and take things out, which I find very annoying. I just try to avoid that.

          I’m not aware that I’m related to a Danny Flanders around here but it’s not impossible – it would be distant though. My Flanders relations that I know of from the most recent generations are all in Kansas, Louisiana, and Texas. My Dad moved us back here in 1973 when he got a job at USC. I was born in Louisiana.

    4. Rhoderic Mills

      Howdy Cousin! Apparently we share the same grandfather Nicholas Miles I. 1599-1658. Now I’m wondering when he emigrated to America? Taking into consideration the landing of pilgrims in 1620, that has to narrow it down a bit. Have you any knowledge? thanks. Rhoderic Mills

      1. Brad Warthen Post author

        I think you win Best Question of the Day. (Unfortunately, there is no prize.)

        And I think I can answer it.

        The Findagrave database says, “Nicholas Mills, Immigrated to Maryland, in March 1651 with his sons John and Peter from Holland. Nicholas Mills, was deceased by 1658.” So he was about 52 when he got here, and didn’t last all that long. Maryland would have been a brand-new colony then, only having been settled in the mid-1630s, if I recall correctly…

        By the way, I think I have some ancestors who arrived in Virginia (but not Maryland) before the Pilgrims’ 1620 date. Remember, Jamestown was founded in 1607, although few of those ill-prepared colonists survived…

  6. Doug Ross

    If you think you’re DNA information isn’t going to end up in the hands of insurance companies and the government, you’re blissfully naive.


Leave a Reply

Your email address will not be published. Required fields are marked *