Sue wrote: >The whole project would shrink >to the point where it would be quite difficult to draw any conclusions at >all. My point is exactly this. We can't make any conclusions yet because we don't any non American samples to support the AAV (apparent ancestral value) suggested by the superior number of American samples. The AAV of Group A as suggested may in the end be the true ancestral value but we can't say for sure yet. We need more Old World samples. I think (this is my opinion) that there are a lot of participants in our project descended from the same 18th century family which is skewing results - just as results would be skewed by the descendants of David Kincaid of Londonderry, Northern Ireland if we just considered non American samples. I have before that stated before that, for one, it is still uncertain if the true ancestral value for DYS 459a,b is 9,10 or 9,9. I still think it is 9,9. The implications are this. If it is 9,9 then one can look at the bulk of set 1a being of the same 17th or 18th century family. If it is 9,10 then one can't make any conclusions about any of them being related to each other (without supporting paper records) as they are simply carrying the values for the tested 67 markers that the original Kincaid had. One other point to remember that is not DNA related is this. Before DNA 'the bulk of' American genealogies had their Kincaids descended from the so called sons of Alexander of edinburgh, the alledged three brothers of Albermarle County (David, Joseph, and James Kinkead) and their related Rev. John Kinkead (John the Clerk), or John Kincaid, the patriarch. If the bulk of American Kincaids claimed descent from this small group of Kincaids before DNA testing, why would one not think there would likely be a bias in DNA testing results for American Kincaids - and thus treat results cautiously until non American DNA tests support the American data. I would like people to think in terms of sampling. If during an election we sampled mostly Republicans, what do you think the survey would suggest about who will win the election. If we sampled mostly people from the state of Michigan, what do you think the survey will suggest about what Americans think about a bailout of the US automakers. You can appreciate a bias will suggest different things about the whole population. It is only natural to think that a strong sampling bias of American participants in our DNA project may be suggesting something different about the whole Kincaid population (hence our Kincaid patriarch) than if we had a better balanced project. Best wishes! Peter P.S. My comments were given as a reaction to the comment that the 67 marker testing is of little value. I think one should at least look at upgrades for the non American tests (i.e. 2617, 23547, 64507, 94749, and 98268 for starters) before we can rule out whether there is any further markers of significance in the 37-67 group of markers.

But we do have samples from some whose ancestors did not immigrate during the Colonial period and therefore are not direct descendents of the PA/VA families. Would it have made a difference in your thinking if 23547's sample was from a line down from James/Helen Scott m 1669 who had not immigrated? 94749's ancestors didn't immigrate during the colonial period. His results do not represent a colonial line. Would these lines then meet your criterea for non-US colonial lines? If more of Rev. Joseph/Francis Cochrane's non-immigrant line had tested would 2617's results weigh more in your mind? The testee was certainly non-US. Granted there is a disportionate number of participants from the colonial immigrants who entered VA and PA because they have been breeding in the US for almost 300 years. Since the US is where the most interest in DNA testing occurs and DNA can help untangle the various family lines, it is natural that these lines are over represented within the project. But there is also a disportionate (given the c1800 birthdates) representation of your David's and his possible brother George's line. We have 2 branches based on 459b (set 1 and set 2/4) of the same family (A). Both branches had colonial immigrant ancestors and those who remained in Scotland or Ireland past the colonial period. Arguing which branch is more ancient is futile at this point and may be futile at any point as the split probably occured well before historical data exists. The high rate of mutation, as exhibited in set 4 and as shown where we have confirmed lineage to a given individual in both sets suggests that deciding which is the oldest lineage on the number of mutations may also be problematic. It would be interesting as an excersize to shrink the chart by allowing only 1 representative of each identical result string then group together those who have a vetted or suspected common historical ancestor. Decide on an AAV for each ancestor grouping (this may have to be subjective) and let that AAV represent the ancestor. This will illimate the over-representation of any given line and should give a better picture of the overall rate of mutation for each set. If I had time, I would do this. Perhaps in January or February. Right now I have a Christmas tree to trim before heading for work. Sue Liedtke ----- Original Message ----- From: "Peter A. Kincaid" <7kincaid@nb.sympatico.ca> To: <kincaid@rootsweb.com> Sent: Friday, December 12, 2008 10:59 AM Subject: Re: [KINCAID] 67 marker DNA test analysis & infoonTheGatheringoftheClans in Scotland > Sue wrote: > >>The whole project would shrink >>to the point where it would be quite difficult to draw any conclusions at >>all. > > > My point is exactly this. We can't make any conclusions > yet because we don't any non American samples to > support the AAV (apparent ancestral value) suggested by > the superior number of American samples. The AAV of > Group A as suggested may in the end be the true ancestral > value but we can't say for sure yet. We need more Old > World samples. I think (this is my opinion) that there are > a lot of participants in our project descended from the same > 18th century family which is skewing results - just as results > would be skewed by the descendants of David Kincaid of > Londonderry, Northern Ireland if we just considered non > American samples. > > I have before that stated before that, for one, it is still uncertain > if the true ancestral value for DYS 459a,b is 9,10 or 9,9. > I still think it is 9,9. The implications are this. If it is > 9,9 then one can look at the bulk of set 1a being of the > same 17th or 18th century family. If it is 9,10 then one > can't make any conclusions about any of them being > related to each other (without supporting paper records) > as they are simply carrying the values for the tested 67 > markers that the original Kincaid had. > > One other point to remember that is not DNA related is > this. Before DNA 'the bulk of' American genealogies had > their Kincaids descended from the so called sons > of Alexander of edinburgh, the alledged three brothers of > Albermarle County (David, Joseph, and James Kinkead) and > their related Rev. John Kinkead (John the Clerk), or John Kincaid, > the patriarch. If the bulk of American Kincaids claimed > descent from this small group of Kincaids before DNA > testing, why would one not think there would likely be a > bias in DNA testing results for American Kincaids - and > thus treat results cautiously until non American DNA tests > support the American data. > > I would like people to think in terms of sampling. If during > an election we sampled mostly Republicans, what do you > think the survey would suggest about who will win the election. > If we sampled mostly people from the state of Michigan, what > do you think the survey will suggest about what Americans > think about a bailout of the US automakers. You can appreciate > a bias will suggest different things about the whole population. > It is only natural to think that a strong sampling bias of American > participants in our DNA project may be suggesting something > different about the whole Kincaid population (hence our Kincaid > patriarch) than if we had a better balanced project. > > Best wishes! > > Peter > > P.S. My comments were given as a reaction to the comment that > the 67 marker testing is of little value. I think one should at least > look at upgrades for the non American tests (i.e. 2617, 23547, > 64507, 94749, and 98268 for starters) before we can rule out > whether there is any further markers of significance in the 37-67 > group of markers. > To see the Kincaid of all spellings DNA chart in Excel: > http://freepages.genealogy.rootsweb.com/~adgedge/Research/April%202004/Kincaid%20%20DNA.xls > > > ------------------------------- > To unsubscribe from the list, please send an email to > KINCAID-request@rootsweb.com with the word 'unsubscribe' without the > quotes in the subject and the body of the message