We welcome any query on Who When Where. If you have previously posted it on another forum (including the old WDYTYA forum), please state this in your opening post - this will save people redoing the research which has been done before: they can look at it and possibly go further with it.

Typical English DNA results on Ancestry

Post here for your queries about DNA, or help fellow researchers understand theirs.
Post Reply
User avatar
AdrianBruce
Posts: 358
Joined: 14 Jun 2020, 18:57
Location: South Cheshire

Typical English DNA results on Ancestry

Post by AdrianBruce »

Thanks to blogger Chris Paton, I found this article https://blogs.ancestry.com/ancestry/202 ... estimates/ about Ancestry's latest ethnicity estimate update.

One chart in there shows typical English results - these include something like 5% Germanic Europe and roughly half that for both Norway and Sweden. (This is in addition to Scotland, Wales and Ireland, which we would probably expect anyway). Mine has 5% Norway, 3% Germanic Europe and 2% Swedish (better get the Abba records out again ;) )

My interpretation of this is that if these are typical figures for English people with no recent immigrant ancestry, then the bulk of the Germanic and Scandinavian elements go right back into pre-Conquest times with Saxon, Viking - and even Norman? In other words, don't bother looking for a timber merchant from the Baltic in genealogical times if the more obvious answer is that most people in England look like that in their DNA.

Conversely, at least 6.25% of my DNA should be Irish according to my known ancestry - but if that 6.25% actually includes a good dollop of Viking settlement (in Ireland) then it might be interpreted as Scandi, which is why I've actually been estimated as only 2% Irish.

I think that it is useful to know typical percentages then we might get some feeling for whether the numbers say anything about us specifically, rather than the (long-term settled) English in general. That blog may help.
Adrian Bruce
Hardwork
Posts: 86
Joined: 14 Jun 2020, 14:15

Re: Typical English DNA results on Ancestry

Post by Hardwork »

I doubt there is much hard evidence for these ethnicity elements. What are they using as base data? Even people with all great-grandparents born in the same county doesn't prove anything as to their various origins. The recent update has given me a big increase in Scottish ancestry and an increase in Irish. Most of my ancestry is from East Anglia, mainly Suffolk and whilst I have a couple of known false paternities on my Nottinghamshire branch, I suspect Living DNA are nearer the mark in whipping up a smidgeon of Welsh ancestry for me. The rest is less than 20% solidly from the West Riding of Yorkshire and I have most of my lines back into the mid 1700s with some much further back. Of course, we can never be sure the paper record is correct but I have no known ancestry further west than Coventry and that a quite distant ancestral line. I think taking ancestry tests to find detailed evidence of geographic origins in the British Isles is doomed to failure, on the whole. Y DNA testing is much more accurate in terms of broader origins of the male line, but even there there can be a dependence on reference to current populations.
User avatar
AdrianBruce
Posts: 358
Joined: 14 Jun 2020, 18:57
Location: South Cheshire

Re: Typical English DNA results on Ancestry

Post by AdrianBruce »

There's plenty of hard evidence - it's the interpretation of it that's moot.

Since no-one in Ancestry or Living DNA or ... is going round digging up dead bodies, they can only use current populations in an area as their base. (People do carry out whole genome analysis of skeletal remains, but - unless anyone knows better - that data doesn't get into genealogy databases). So the basic assumption is that the current population in an area represents the old population. Of course, where there has been population churn, that's not true at a detailed level (i.e. below the level of a continent).

My French ancestry doesn't show up, for instance - there are two possible explanations for that - (1) it's just a fairy story that I have Huguenot ancestry and / or (2) there is virtually no French in the base data (because it's illegal to sell Customer DNA testing in France).

Great-grandma, as well as telling Dad that she had French ancestry, came from Bristol, so I ought to have 12.5% South Western ancestry. That doesn't show up at all, presumably because the SW ancestry can't be distinguished from the Midlands ancestry which it does acknowledge. Except - one of my Bristolian ancestors was actually born on the banks of the River Tyne - so at what generation are you going to start measuring ethnicity? That was the point that I took away from the blog - the only sensible interpretation of the ethnicity numbers is that we're seeing evidence from way back, outside our genealogy period.

The interesting thing is that, having said all that, as I recollect, there are certain DNA characteristics that are found only in certain areas of England so if you happen to have that characteristic, Occam's Razor would suggest that far back, you have ancestry from that area. (I think I saw the maps of DNA characteristics in a TV program that Professor Alice Roberts did about the Saxon migration into the British Isles and since she was also wielding a large Saxon sword I wasn't about to argue!)

Y-DNA, by the way, is clearer on the "who" but suffers from the same problem about "where" - it has to use current populations - it's just that it's easier to think it through because there's only the one line!
Adrian Bruce
Hardwork
Posts: 86
Joined: 14 Jun 2020, 14:15

Re: Typical English DNA results on Ancestry

Post by Hardwork »

Surely, if it is open to interpretation it isn’t evidence? Evidence has to support a contention that something is provably true. If it can support more than one contention it is just data. As you admit, where there is population churn the results cannot be at a detailed level and I think you will find for many, it’s a detailed level they are expecting, rightly or wrongly. I’m sure ancient skeletal remains are used for comparison but whether any appear in commercial databases I cannot say. Until more ancient remains are tested accuracy will remain questionable.

There is a third reason your French ancestry may not show. It may have just become so diluted it cannot be clearly identified.

As for ethnicity, it is precisely that a definition of ethnicity is only discernible in a very broad sense using autosomal DNA that makes regional results usefulness debatable. And I agree, where do you start measuring that from?

Y DNA isn’t easier “to think through” as you put it, but it is from one line, and one line that mutates slowly and is not diluted in the same way as autosomal DNA. Of course, you are correct in that it still requires reference to modern populations but for Y DNA, that is generally on a continental scale.

Autosomal “ethnicity” accuracy may be possible if one has plenty of ancestry from a relatively small area, like a region, and if one has many ancestors from there over a long period of time, as it is more likely that a wide section of the population share genetic signatures but even that has to be tempered with other conditions that may have caused widespread population movement and replacement, not always recorded in history.

In conclusion, personally I feel trying to micro-analyse autosomal DNA for anything but to get a general flavour of ethnicity, in a small area such as the British Isles, isn’t likely to work for most people. DNA testing kits are often advertised with an eye to discovering ethnicity but personally I feel that the variation in interpretation between companies’ analyses shows that it is not yet “scientific” enough to be sufficiently reliable other than for taking a test for a bit of fun. At for testers with roots in the UK.
User avatar
AdrianBruce
Posts: 358
Joined: 14 Jun 2020, 18:57
Location: South Cheshire

Re: Typical English DNA results on Ancestry

Post by AdrianBruce »

Hardwork wrote: 16 Sep 2020, 01:00 ... I feel trying to micro-analyse autosomal DNA for anything but to get a general flavour of ethnicity, in a small area such as the British Isles, isn’t likely to work for most people. ...
I'd agree with that - Judy Russell, "The Legal Genealogist" says:
... these estimates are pretty darned good at the continental level, distinguishing between Europe, Asia and Africa, just to name three continents, in their estimates ...
(See https://www.legalgenealogist.com/2020/0 ... -not-soup/ )
Adrian Bruce
meekhcs
Posts: 468
Joined: 02 Jun 2020, 18:19
Location: Lincolnshire, but Hampshire born and bred!

Re: Typical English DNA results on Ancestry

Post by meekhcs »

Edited To simplify Edited, I have just looked at mine and my husband's and quite honestly, as with thru lines, I think they just base it on the DNA linked Trees they have.

I have research trees attached to our DNA results as I am trying to hook matches from certain lines, which differ from our full tree. The ethnicity results for us reflect the records we have in our Reserch trees.

I did wonder at the 4% swedish but then given some of the DNA trees I have looked at with records supposedly back to 1066 and even beyond I shouldn't be surprised
Sally
User avatar
AdrianBruce
Posts: 358
Joined: 14 Jun 2020, 18:57
Location: South Cheshire

Re: Typical English DNA results on Ancestry

Post by AdrianBruce »

No, ethnicity is definitely based on DNA - you can get ethnicity results without having a linked tree so Ancestry can't tell who you're possibly related to.

For most people, at least in the British Isles, my impression is that the groupings in the ethnicity estimates go back more like 1,000 years so the genealogy for those people is, frankly, irrelevant, and there is no point saying, "But I haven't got any Scottish ancestry and I've got back to 1750!"
Adrian Bruce
meekhcs
Posts: 468
Joined: 02 Jun 2020, 18:19
Location: Lincolnshire, but Hampshire born and bred!

Re: Typical English DNA results on Ancestry

Post by meekhcs »

I know I have a lot of things on my mind this week but I am making some very elementary Mistakes!!

Thanks for putting me straight Adrian.
Sally
Post Reply