Why Your Latest Ancestry.com Results Could Include More Scottish or Irish in Your Ethnicity Estimates

Five days ago I published a press release written by Ancestry.com entitled Ancestry® Now Delivers More Precise Ethnicity Estimates and I added a couple of brief comments of my own.

The article generated a lot of comments from newsletter readers, many of whom questioned if the new and “improved” results were really more accurate than earlier reports or if perhaps the new reports were even less accurate.

Now an article published in the Ancestry Blog explains the new results amongst Scottish and Irish ancestors and it may also help explain some of the new results amongst other countries as well.
If you are wondering about the recent changes to Ancestry.com’s reports, especially amongst Scottish and Irish ancestry, you will want to read the article at https://blogs.ancestry.com/ancestry/2020/09/11/why-your-latest-results-could-include-more-scotland-in-your-ethnicity-estimates/.

20 Comments

I am with your previous correspondents who find Ancestry’s recent iteration of their ethnicity estimates worse than ever and the excuses for the explosion of Scottish and Irish elements, unconvincing, to say the least.
I have a carefully researched paper trail for my ancestors, going back to the early 1700s. Having taken advantage of recent special offers from Living DNA and My Heritage DNA, I can report that their ethnicity estimates show no such skewing of Scottish and Irish results.

Liked by 1 person

My new estimate comes up 10% Swedish, but I have gone back many generations on all lines with no Swedes. Paper records all show German and British ancestry, Ancestry’s preceding estimate made much more sense.

Like

    Have you tried 23andMe? Their estimates may be better (certainly more detailed) and they’ll give you chromosome maps so you can see if your “Swedish”, if any, is 100 little pieces or some big chunks. They map specific locations for larger DNA percentages. (Honestly, I don’t work for them. Just obsessed with genealogy.)

    Like

I agree with you. I have no ancestors from Scotland. They came from Ireland. However, now it shows I’m more Scottish than Irish. This is nuts. Lumping in two separate regions makes no sense. But then what can I expect from a company that keeps sending me my own obituary but won’t let me see it unless I sign up again!

Like

Their original analyze was fairly good, the last one is a joke. My maternal grandparents immigrated from Ireland and that they still cover fairly good; however, my paternal grandfather immigrated from Switzerland and my grandmothers parents both immigrated from Germany. None of that shows, what I get shows a relationship to the for eastern coast of France. Garbage.

Like

    If you can’t trace your line back at least 10 generations then Ancestry.com knows more about your origins than you do. My scandinavian percentages went way high even though my mother’s line appears to have none, but her English side is heavily saturated with nobility, almost all lines leading to either William the Conqueror (perhaps a hundred times over) or to members of his entourage, who were Normans with roots in Norway, Sweden and Denmark. I also get a lot of east France, Ostfrankenreich, probably due to Grandpa Charlemagne and his prolific retinue.

    Like

    I don’t know about “10 generations” but like John Granacki says, going back to your 5th or 6th great grandparents will reveal a lot more diversity than you know, IF you can. DNA cousin matches can validate even 5th or 6th GGPs. (However my impression is that Ancestry.com is somewhat light on non-North-American DNA cousin matches and document hints.)

    Like

I was surprised to see so much Scottish on my latest estimate from Ancestry. It is now my highest percentage ethnicity with them. Even higher than the previously combined category. It’s not true, but it is interesting. I now have 3 different countries from 3 different tests as my highest percentage. Only one of them matches my family tree. It looks like there is some overlap between Ireland and Scotland on Ancestry. I’m wondering if that is where the confusion lies.

Like

What is the deal with the “England and Northwestern Europe” group? I can sort of see it, but only if you count going back to about 1200 and before, when the Angles, Saxons, etc., invaded. If they’re going back so far, they need to go back a bit farther and include the current Italy to pick up Roman ancestry!

Like

Just confirming that their results did become less accurate (significantly) with this update. I’ve validated many of my lines going back into the 19th century through DNA connections and the admixture is just wrong. Scottish did go up quite a bit. My last full blooded scottish ancestor was in the 17th century. 10%? Not a bloody chance.

Like

Make sense do to the king of England promising land to the Scottish in northern Ireland and when they got there the wasn’t any land so they came to Amrican after some many decades.

Like

Iceland??? Really? What happened to Sardinia and why go back only 5 generations for new reports???? I can’t agree with this because I have traced ancestors back 20 generations 😦

Like

Unless you have worked your family tree back to the beginning of time you can not be sure where your ancestors and their and now your DNA has been.

I was DNA tested about four years ago and have been researching my family tree in every spare moment since then. I was horrified to learn that the native Irish, a small, dark, wild and pagan people with possible Spanish/Portuguese ancestry where slaughtered by the English king who then moved Scots who were loyal to the English crown into the newly unoccupied Irish lands.

This is only one of many HISTORICAL EVENTS that explain why our DNA test results do not match our paper pedigrees.

Liked by 1 person

I had Ancestry DNA done a few years ago. It showed all of Great Britain and Northern Germany which lined up to my paper research. Recently I received another DNA test as a gift. It showed a large segment of Scottish heritage and NO GERMAN heritage. There no doubt that my fathers side came from Germany. My last name is German. And there are records stating where they came from. I now question all of Ancestys records both paper and DNA. What a waste of money and time. I’m ticked off.

Like

Seems like ethnicity estimates are a bit of art mixed with the science. Regardless, there’s never a “final answer” for genetic genealogy. I’ve tested on Ancestry, MyHeritage, and FamilyTreeDNA and can each give different ethnicity results. My brother and sister each tested on Ancestry and their ethnicity estimates are different than mine (yes, the DNA proves we have the same parents).
FTDNA hedges its bets by lumping the British Isles into one place. MyHeritage and Ancestry are comparable now, but were very different before the update.
While I take it all with a fairly large grain of salt, it’s been my experience that Irish or Scottish heritage doesn’t show up in my tree until I’ve reached back into the early 1700s or mid to late 1600s. That’s not easy to do for all lines on any tree.
As I understand Ancestry’s white paper on the topic, they’re basically trying to replicate your ethnic origins from several hundred to a thousand years ago or more. If your ancestry is dominated by early immigrants there’s a pretty good chance you’d never be able to replicate on paper what Ancestry’s done. Consider non-paternal events. Consider records lost in fires and whatnot.
I’m not trying to say Ancestry is correct – I expect additional changes in the future. On the other hand, paper trails are rife with errors as well. Folks will try to add Native American ancestry where none exists. They try to attach ancestors to nobility and royalty when none exists. They fail to account for the historical aspects of immigration.
Religious persecution, war, poverty, famine, etc., drove migration then as it does today. Some poor serf is more likely your ancestor than is William the Conquerer.
If you think your ancestors came from Ireland, but Ancestry shows Scottish heritage, remember that quite a number of Scots migrated to the Ulster area of Ireland during the 1600s to escape poor living conditions, and eventually on to the Americas for the same reason. Scots fled to the Americas during and after the Jacobite revolutions of 1715 and 1745, and many Scots who were on the “wrong” side of the Jacobite revolution were deported to the Americas.
So enjoy having a bit of the Luck ‘o the Irish while you can and embrace the heritage you never knew. Next update you’ll be from someplace boring.

Liked by 1 person

Would those with the 10 or ’20’ generations please acknowledge that they don’t have any blanks in their research. I have 4 end of lines from my 4th great grandparents and then in the 5th upwards there are 10-12 lines with no parents and some with only father’s as is often found in the early Scottish Old Parish Records (OPR’s), and this is just on my father’s side! On my mother’s side there are two lines that only go as far as 3rd great grandfathers – one from England and one from Scotland so who knows what their DNA ancestry would tell me. It burns me that the one from England is given a father in some trees that if the researchers did even a little comparison would realize they have the wrong father for him. Same with my 3rd great grandfather from Ireland researchers have given him a father from the USA which by dates they give is plain impossible, yet his wife who was of the Quaker faith has well documented lineage back to the mid 1600’s in England. Some lines can easily go way back in time and others simply, for many reasons, just stop in their tracks, these are called “brickwalls” and we all have them, AS WELL, we all make mistakes. Just found I had the two wives in the wrong order so the children were attached to the wrong mother – all the right people – but it makes a difference to which line I should follow for DNA purposes.
People need to ‘get a grip’ and enjoy their research and even do some “Do Over” work of their files.

Liked by 1 person

    DNA cousin matches can validate the ancestors your paper search has turned up even back to 7th great grandparents in some cases. It really builds confidence in your tree. It can help you find errors and even turn up “non-parental events”. Although it hasn’t happened for me yet, it might even help you break through 1 or 2 of those brick walls. Unfortunately you will have lost all DNA from many such distant ancestors so lack of DNA matches is not conclusive and for that reason DNA may not tell you about many lines, and DNA cousins can’t take you back much farther than 8-9 generations in any case.
    (Pet peeve – the guy who claims to have traced his ancestry back to some 3rd century BC Greek general !)

    Like

Estimates of ancestral percentages from autosomal DNA testing are neither an art nor a science. They are still at the stage of knowledge of a pseudoscience, astrology instead of astronomy. The concept is based on looking at specific mutations found at fairly high frequencies in certain populations and then assuming that if they are found in an individual test, it is valid to extend those frequencies into personal genetic history. This of course fails to recognize that no modern ethnic population is monolithic or representative of all its members. It also largely discounts the effects of in-breeding, a very common situation for most of humanity over most of its history. Let me give you a personal example. My grandmother was born in the Hallingdal region of Norway, in a village along a narrow river valley surrounded on both sides by mountains. Most of the interactions and intermarriages among the people living there was up and down the river for a village or two in either direction, over at least the past 500 years. Church records and probate records are quite detailed and a researcher can take most lines there back to the mid-1600s if not earlier, some going to the 1300s if reasonable assumptions are made from the succession of landownership and the names in charters. Because of the inbreeding, my grandmother descends from many of her identified ancestors in the 1600s five or six or even more times by different lines, as do most people there. Of my autosomal DNA matches, the vast majority appear in my match list because they have some ancestor or ancestors from this area – about 70 percent of my matches are related by descent from a Hallingdal ancestor to my Norwegian grandmother, although you would statistically expect only 25 percent of my overall matches to show this connection since my other three grandparents were of completely different ethnicities. With the current revision to Ancestry’s estimates, they have me as 70 percent Norwegian from this region. Of course I have only 25 percent ancestry from there, but their estimate simply counts the number of times I show certain mutations they have identified as associated with this region. That count is of course magnified by inbreeding. So their numbers are wildly inaccurate in my individual case. I should also point out that they have entirely failed to spot the 25 percent of my ancestry from the current Czech Republic, or the 25 percent of my ancestors from Denmark. For a long time, they did give me a huge British ancestry I don’t have, based presumably on considering older mutations from the continent as associated with the Danelaw in England.
It seems transparently clear that they simply have insufficient data in their overall database of results to identify the local mutations that would properly allocate that ancestry. As with so many things, their estimates work ok for the regions where a great many people have tested and for those individuals whose ancestors are primarily from such places. For those who have different regions in their ancestral lineages, Ancestry just ignores such areas or misidentifies them, and fills in the estimate with what is essentially gibberish. I’m not saying that such estimates may not eventually become reasonably accurate, as further data is identified and properly dealt with, but we are obviously a long way from the point where anyone should take some of these numbers so seriously as to doubt well-documented paper trail ancestry solely on that basis.

Like

I found that the new estimates were much more accurate. Ancestry is the only major DNA testing site to identify Polynesian ancestry, FTDNA and 23&me list those with Polynesian ancestry as South East Asian/ Asian/ Paupa New Guinea. My paternal grandmother was full blood Hawaiian and paternal grandfather was Portuguese from the Azores. My estimate went from 26% Polynesian to 25% Hawaiian/New Zealand Pacific Islander and 1% Samoan and Portuguese from 14% Portuguese and 9% Spanish to 24% Portuguese and 1% Spanish which is just what I would expect.

Like

Leave a Reply

Name and email address are required. Your email address will not be published.

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

You may use these HTML tags and attributes:

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <pre> <q cite=""> <s> <strike> <strong> 

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: