Surname Tests - Realistic Expectations

[DevonCruwys]

I am well aware of mis-spellings; having my name mis-spelled almost every day!
(Your post today being typical! )

But, my paternal line have been very consistent in spelling the surname, or having the surname spelled, as Garrad throughout the census era and before!

However, that's irrelevant - the fact remains that, for me at least, DNA testing would be of no significance whatsoever.

I can well understand you having problems with your surname. It's very easy to jump ahead and assume that there is an R in the name as that is the most well known spelling. I'm glad I'm not the only one to make the mistake!

Do you not have any desire to verify your family history research? As a family historian I like to check every available record for my surname. Each new record brings new information. I wouldn't reconstruct a family tree without checking the censuses just as I wouldn't accept that a reconstructed tree was correct without the additional evidence from the DNA testing. I've found it very satisfying that I've been able to use DNA testing to add to the existing evidence that I've acquired from the documentary research. I have two well documented lineages that share a common ancestor in the mid-1400s, and by testing descendants from each line we've shown that our assumptions were correct and we don't have any "go-getting" mothers in these direct lines.

[giblet]

Ding ding ding  What's this round 3 or maybe 4 

Personally i think DevonCruwys and KGarrad are loving every minute of this     


When trying to find who my grandfather was i did come across a fella who i have no doubt is him but without DNA from him and my father we will never know. My father is still living but this other fella died years ago. His cousin did offer DNA if my father wanted to be tested. What was the use? It would either eliminate this fella or just say dad was "one of their lot" but it wouldnt pin point to the pacific person who fathered him. It still wouldnt prove 100% this fella was my grandfather so to me it was of very little use.

[DevonCruwys]

When trying to find who my grandfather was i did come across a fella who i have no doubt is him but without DNA from him and my father we will never know. My father is still living but this other fella died years ago. His cousin did offer DNA if my father wanted to be tested. What was the use? It would either eliminate this fella or just say dad was "one of their lot" but it wouldnt pin point to the pacific person who fathered him. It still wouldnt prove 100% this fella was my grandfather so to me it was of very little use.

This scenario is in fact something that you could prove with the new autosomal DNA tests. You would need to test yourself or your father and someone who is descended from the person you believe to be your grandfather. This could be a male or female descendant. The test would reveal how much DNA you share in common which can be used to predict the relationship. Two first cousins would normally share 12.5% of their DNA. There are other relationships where the same amount of DNA is shared (eg, half-aunts/uncles, half-nephews/nieces) but these can usually be ruled out by the ages of the people testing and the documented paper trails. See:

http://www.isogg.org/wiki/Autosomal_DNA_statistics

[DudleyWinchurch]

Interesting discussion. 

I had been quite sceptical about the value of these DNA resources for the various reasons mentioned above but, after going to a talk at a family history fair last weekend and hearing how the autosomal test works I have decided to go ahead and try this.

I have a specific problem (brick-wall) in my family tree which I have been trying for several years to overcome but seem to have hit a block with the lack of baptism records in Mayo for the period about 20 years before the famine.

At about the same time as, or shortly after, my great-grandfather and his newly-married wife arrived in England in 1851, there were two other persons, one male, one female, bearing the same family name who appeared in the same place too.  All three lived very close to each other and seem to have been godparents to each others' children. 

The other two married in England so I know the name and occupation of their father.  Irish records of that time do not supply these details so I cannot find that of my own G2grandfather.  However, both the other couples moved to the US at different times, initially to the same area.

For some years now I have been in touch with a descendant of the female of the same name.  After hearing how the Family Finder (autosomal) test works, we have decided to both apply for the test, listing family names but not specifying that we want a direct comparison.

If we are correct in concluding that, as they behaved like siblings, our great grandparents were in fact brother and sister, then we should both be sent information of the match, as the relationship should be well within the parameters of the test.  If not, then it seems unlikely that out of all the database, they would pick this particular match rather than any other ones (based just on surname matches etc.).  I have a fairly detailed tree that should be able to identify the direct links with anyone else on my other lines, within about 5 generations of me, that might happen to be in the database.

I look forward to getting the test done and will keep you posted on the outcome.

DW

[DevonCruwys]

Dudley, Thanks for sharing your experiences. This is exactly how DNA testing works best. You need to have a specific scenario in mind and you need to test the right people to prove or disprove the hypothesis. You should get a definitive answer from your autosomal DNA tests. Good luck!

[acorngen]

Dudley would that talk have been at Doncaster by any chance?  If so I was at that talk and as much as I respected what was said the talk was very vague in reality and had so much in it that was wrong and not well researched.  I also felt it was not a talk about the uses of DNA (which I still say are non existent and why i asked the question I did and got the reply in the end that I wanted) but more about selling more tests.  If it was at Doncaster I dont suppose you got the speakers email address that you could share with me in private please.

[DudleyWinchurch]

Dudley would that talk have been at Doncaster by any chance?  If so I was at that talk and as much as I respected what was said the talk was very vague in reality and had so much in it that was wrong and not well researched.  I also felt it was not a talk about the uses of DNA (which I still say are non existent and why i asked the question I did and got the reply in the end that I wanted) but more about selling more tests.  If it was at Doncaster I dont suppose you got the speakers email address that you could share with me in private please.

Hi acorngen,

It was the talk at Doncaster and I also could not see how the tests in the talk would help with my particular brick-wall but waited after the talk and asked a question framing the problem and how could the tests answer it, and then got what seemed like a very straightforward explanation of the autosomal test that seemed worth following up.

Unfortunately, we ran out of time and had to rush off so I did not get the speakers email address but when I got home I went to the company website and found more information on how the process works.  Not sure that I understand it all, but discussed it (by email) with my potential third cousin and he found a comparison website that seemed to show the company had advantages over some other companies offering similar tests, so we decided to go for it.

This is definitely being done in the spirit of an experiment but after looking on the website, I noted that there were projects involving several of the names on my family tree.   

I have supplied names and dates of my most distant ancestors on the male and female lines and family names only, with country of origin, for my great-grandparents. 

As far as I am aware, there is only one family name in common between myself and the potential third cousin and there are clearly other projects involving both that name and others from my list, already in their system.

It is to be hoped that the test will find the relationship that we think probably exists between us (although of course, we may be mistaken there) and that any other relationships that it might find within other lines of my family, I probably have enough documentary evidence to confirm or deny.  It is on this basis that I hope to be able to establish my own personal confidence level in the test, though I appreciate that I may have to wait some time to get sufficient feed-back.

It's still quite a lot of money to perhaps get around one brick-wall but I've already made several trips to Dublin looking for further evidence on that as well as other parts of my tree, and even spent a week in Mayo without any further success, so not as much as my trips cost.

[acorngen]

DW I was at that talk and asked the question shouldnt Genetic genealogy and genealogy be kept seperate.  I still think they should be tbh.  What he probably didnt tell you with him selling the idea for money is that A-DNA is only accurate enough to be of any real use to 4 generations after that the chance of two people being a match is around the 10 percent mark.

I deliberately didnt stop back to ask questions after because I would have kept him talking for hours on the subject.  Whilst I dont profess to know the ins and outs of how DNA is formed etc I have done quite a bit of research on the subject.

I was pleased though when he brought up the fact that the Neanderthal genome had been mapped and that significant tests have been done to show there was interbreeding between Neanderthal and Homo Sapiens, something which I have been arguing was the fact for a long time now.  I was also pleased that he showed that africa had no neanderthal DNA again something I had argued.  The only part of my argument not yet proved is that aborigine man (AUS and NZ) are the closest modern people to Neanderthal. 

Back to this subject though.  If the DNA does show a match how will you actually prove where the connection lies?

[DudleyWinchurch]

Hi again acorngen,

The speaker on Saturday did point out the restrictions on the A-DNA matching and it is clearly stated on the website with the percentage match levels that would indicate each level of relationship and the corresponding confidence levels. 

However, the problem we want to solve is that we have documentary evidence that indicate quite a high likelihood that the other researcher and I are third cousins.  Our great-grandparents had the same family name, arrived in the same place in England, from Ireland at about the same time, and on two different censuses were found living as close neighbours and their names appear as godparents for each other's children.  The only thing we don't have is any document saying that they were actually related.

I have traced all 8 great grandparents, all with distinct family names.  So any third cousin should appear on my list as such.  According to what I understand of the test methodology, if any closer relationship was indicated, it should only be because one of them has a double-dose of genes from the same family further back (but not much further or too diluted to make much difference, so perhaps one or two generations further back).  If a relationship, but more distant, was indicated then, the two potential siblings would be cousins at closest and would then be towards the limit at which this test could confirm what particular relationship.

However, there is no reason to suppose that we have any other lines in common, so if there is a relationship strong enough to show up on this test, then it is only likely to be on this one line.

Of course, to demonstrate that the test parameters quoted by the company are correct, it would be nice to see the results from a sibling and one or two first cousins, and some second cousins across a variety of lines, and other third cousins too for which I do have full documentary evidence, but that's wishful thinking, though I might suggest it to a few others among my contacts, if we appear to get a result here.