Announcement

Collapse
No announcement yet.

Help Understanding Mysterious "Close" Matches

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Help Understanding Mysterious "Close" Matches

    My closest FF matches are the most mysterious, and I need some help understanding what’s going on.

    A brief synopsis of my scenario:

    My ancestry is ½ Italian (paternal grandparents were Italian immigrants) & ½ Colonial New England (maternal family all lines go back to 17th century English settlers).

    I tested and my father tested through both Ancestry & FTDNA. Dad’s results came back as expected. His near matches are all Italian descended people whose ancestors come from the same general area as my grandparents. No surprises here.

    I match several of Dad's close matches. OK, makes sense. And I have several close matches (2nd and up to 5th cousins) where I can easily identify genealogical connections to all four of my maternal great-grandparents, etc. In other words, Mom didn't test, but I'm finding matches to other folks out there who are related closely to her four grandparents. OK, that makes sense too.

    But then I have several other close matches (2nd to 4th cousin range, etc.) that make no sense at all (and are completely inconsistent in relation to each other as well). One close match is to a woman who was born in Norway to Norwegian parents and all Norwegian ancestry (I have no Norwegian ancestry). Another is to a gentleman whose great-grandparents were all German immigrants to the Pittsburgh, PA area (I have no German ancestry, and no ancestry in Pennsylvania). And there are several matches to individuals with completely southern (U.S.) ancestry. I have no southern ancestry. Another close match is to a woman in Australia whose father was a Scottish immigrant and whose mother was from English families who go back to early 19th century Australia. Nothing familiar in any of that either.

    Forgive my obvious lack of expertise in all of this, but my question basically is: why are these people who are clearly NOT related to me showing up as close DNA matches (2nd to 4th cousins, etc.)?

    Would appreciate any insight or thoughts on this. Thanks muchly!
    Last edited by MaineDNA; 29 September 2014, 04:43 PM. Reason: clarification

  • #2
    The 4th cousin is not THAT close Anyway, here is the explanation:

    The matching program found that the segments you shared with the other person were likely Identical By Descent (IBD) . It was only a high likelihood, and not certainty. As in the reality they are only Identical By State (IBS).

    If you search for the above two terms, some people wrote very good explanations.

    Comment


    • #3
      You can't be sure these aren't your relatives, no-one's tree is ever that complete.

      The information that you are using to say ''these aren't my relatives'' is incomplete. You know about the descendants on your side of the family, however you don't know every step/generation on the sides of the family that are unknown to you.

      Your 2nd-4th cousins could be perhaps 4th cousins twice removed, that means you would share your 3xgreat grandparents - any one of them could have had a child (perhaps by a previous marriage?) and one of their descendants could have moved at any point to one of the locations you mentioned.

      DNA tests can throw up this unexpected matches but it doesn't make them any less valid, if you share enough DNA to be predicted as a 2nd-4th cousin I can't see this being a case of IBS. Plus you can't rely on others trees - I have seen the standard of some trees and their aren't always reliable.

      Maybe at some point someone else will test who is a strong match to you and the other cousins you are confused about. That match might make things a little clearer so it may be a case of 'wait and see'.

      Comment


      • #4
        Originally posted by Lucky View Post
        [----] DNA tests can throw up this unexpected matches but it doesn't make them any less valid, if you share enough DNA to be predicted as a 2nd-4th cousin I can't see this being a case of IBS. [----]
        In addition to IBS (Identical By State) versus IBD (Identical By Descent) segment match, there is another issue at play. Read on.

        There are limitations of the current DNA testing technology, resulting in some SNP values not being read and even a couple of values (out of hundreds of thousands) potentially misread.

        It is my understanding, that due to the above limitations, all the companies that offer autosomal DNA testing (tests like the Family Finder), decided to use the software responsible for the interpretation of the results in such a manner that it allows for inclusion of false positive matches. The alternative strategy would potentially remove matches even from close family members. (There are no strategies in-between.)

        Consequently, for any endogamous population, the number of false positives matches can be very large, in thousands. Those outside of endogamous populations might still get hundreds of false positives to sift through...

        Comment


        • #5
          Thanks for your responses DNA & Lucky, I appreciate your thoughts.

          To clarify a couple points made by Lucky:

          “You can't be sure these aren't your relatives, no-one's tree is ever that complete.”

          I understand what you are saying here, and I agree it’s possible to share ancestry with just about anyone if you go far back enough in time. And sure, no one’s tree is ever complete enough to say anything for sure.

          However, I do believe I can say with reasonable certainty that I do not share a 2nd to 4th (or even much further) degree of genealogical kinship with some of my matches that are reported as such. The vast majority of my matches make perfect sense. And there are also several where no ancestral match is apparent but I think can be resolved with further research, because the pedigrees of the matching person make sense in relation to mine (similar backgrounds, geography, etc.).

          To more specifically recap my personal scenario: my paternal (all Italian) ancestry has been well documented in all lines going back at least 8-9 generations (and more in some cases). My maternal (all colonial New England, and before that English) ancestry is well documented back to the Great Migration (early to mid 17th century) in all but a couple or so lines that end with a female ancestor in the mid 18th century.

          The vast majority of my near and more distant matches support all of this. Dad’s matches (and the matches I share with him) are to Italian people from the same area of Italy that his family came from. Most of my other (maternal) matches are to families that are related to each of my four maternal great-grandparents. This all makes sense.

          And I agree that any one of my 3rd great-grandparents could have had children by a person different than their spouse, but it’s not realistic to envision any of my 18th century maternal grandfathers in Maine or in Italy slipping away to Norway to impregnate a local girl (etc.).

          These are the 2nd to 4th cousin matches that I question.

          The gal in Norway shared with me a remarkably complete pedigree going back 8 or 9 generations in nearly all family lines – all Norwegian ancestry - from the same little area of Norway to boot. Similar story with the German guy and the Australian lady.

          My (probably very poorly made) point is: I can understand that it’s possible that unexpected “matches” can turn out to be true genetic cousins with further research. But what about those where that just doesn’t seem possible (because the match is from a different country, or has completely different ethnic ancestry).

          Thanks again for your insights.

          Comment


          • #6
            Originally posted by MaineDNA View Post
            [----] But what about those where that just doesn’t seem possible (because the match is from a different country, or has completely different ethnic ancestry). [----]
            My previous post described the process by which such bogus matches appear in the Family Finder results.

            I am only a part time hobbyist, but the technology limitations are real (and described in the detail in research papers and technology announcements on Internet).

            P.S. Definitely not in your case, but some researches overlook adoptions that before times when they were institutionalized (varies between the countries) often left no paper trail. Most common case being of a widow with very young children marrying for the second time and her children acquiring their step-father family name. And I have had once a widower with two children, that died not too long after marrying for the second time, and she kept his children, married for the second time, and her second husband gave his name to the children... Luckily, we discovered that before any DNA testing...

            Comment


            • #7
              Originally posted by dna View Post
              I am only a part time hobbyist, but the technology limitations are real (and described in the detail in research papers and technology announcements on Internet).
              I appreciate your level minded approach, thanks for sharing your thoughts and experiences. As I've read more on this, I've seen many other examples and shared experiences that further illustrate your points. I dare say that as this new technology develops, perhaps some of these limitations will become better understood and - hopefully - resolved.

              I am not ignorant to the fact that things sometimes happened that are not necessarily reflected in the genealogical paper trail. The possibilities for those types of events are numerous, however without direct evidence to explain what seems to be an unexpected DNA match (or mismatch), we are really just jumping to conclusions or, perhaps at worse, succumbing to wild fantasy. I'm obviously not talking about what should be easily identifiable matches with immediate family, but it's clear to me that at least a handful of my 2nd to 4th cousin matches (and beyond) could not be related to me within the past dozen generations.

              Thanks again.

              Comment


              • #8
                Originally posted by MaineDNA View Post
                I appreciate your level minded approach, thanks for sharing your thoughts and experiences. As I've read more on this, I've seen many other examples and shared experiences that further illustrate your points. I dare say that as this new technology develops, perhaps some of these limitations will become better understood and - hopefully - resolved.

                I am not ignorant to the fact that things sometimes happened that are not necessarily reflected in the genealogical paper trail. The possibilities for those types of events are numerous, however without direct evidence to explain what seems to be an unexpected DNA match (or mismatch), we are really just jumping to conclusions or, perhaps at worse, succumbing to wild fantasy. I'm obviously not talking about what should be easily identifiable matches with immediate family, but it's clear to me that at least a handful of my 2nd to 4th cousin matches (and beyond) could not be related to me within the past dozen generations.

                Thanks again.
                I'm not sure if my case is similar to yours, because I have FTDNA customers claiming to be 5th to remote cousins and one came from a different country and we have no paper trail of anyone from a different country. I don't think any of these supposed matches are reliable. But since no one suggested an IBS for my case, I'm not sure it applies to me. At anyrate, if it did apply to me, then the case for IBS would be even stronger since my tests match with customers as 5th to remote cousins.
                Check it out and see what you think.
                http://forums.familytreedna.com/showthread.php?t=35822

                Comment


                • #9
                  What IF

                  What if you have/had close male relative who had a one night stand that resulted in a child they don't even know they fathered?

                  Comment


                  • #10
                    Originally posted by Jim Barrett View Post
                    What if you have/had close male relative who had a one night stand that resulted in a child they don't even know they fathered?
                    These types of things happened, sure, but it doesn't really work as a realistic explanation for someone who (for example) grew up in Norway and has completely Norwegian ancestry for the last 8 or 9 generations ... and has among their otherwise Norwegian DNA matches a 2nd to 4th cousin match with a guy whose parents were Italian immigrants with Italian ancestry, and all his other matches are to Italian people from the same region of Italy.

                    There are some out there who are quick to suggest a non-paternal event for every unexpected match (or non-match), even when that scenario doesn't make sense in the context of everything else. My understanding is that DNA genealogy is a tool that should be considered in relation to real genealogical research before any definitive conclusions can be made.

                    Comment


                    • #11
                      Originally posted by MaineDNA View Post
                      These types of things happened, sure, but it doesn't really work as a realistic explanation for someone who (for example) grew up in Norway and has completely Norwegian ancestry for the last 8 or 9 generations ... and has among their otherwise Norwegian DNA matches a 2nd to 4th cousin match with a guy whose parents were Italian immigrants with Italian ancestry, and all his other matches are to Italian people from the same region of Italy.

                      There are some out there who are quick to suggest a non-paternal event for every unexpected match (or non-match), even when that scenario doesn't make sense in the context of everything else. My understanding is that DNA genealogy is a tool that should be considered in relation to real genealogical research before any definitive conclusions can be made.
                      Regarding my case, customer service says "an NPE is almost always the first thought until we know more" and "there is a possibility of an NPE in anyone's ancestry."

                      so in your case an NPE is almost certain.
                      But what sort of NPE? An adoption? Illegitimate birth?
                      An NPE into your line or into the Norwegian's line?
                      That's the real question.
                      Last edited by Hando; 2 October 2014, 12:32 PM.

                      Comment


                      • #12
                        In other words, you need to consider the possibility that one of your direct ancestors was adopted or an illegitimate child.
                        Last edited by Hando; 2 October 2014, 01:33 PM.

                        Comment


                        • #13
                          I am struggling to understand why some people seem to think that every unexplained relationship has to be the result of an NPE. At 4th cousin level we are contemplating a fairly extensive tree with many branches. I certainly cannot document my full pedigree back to that level and trace all of the branches down again to parallell generations. Maybe one legitimate offspring somewhere had a connection to another country but has not shown up yet in the paper trail. Am I missing something in my understanding here?

                          Comment


                          • #14
                            Originally posted by LynCra View Post
                            I am struggling to understand why some people seem to think that every unexplained relationship has to be the result of an NPE. At 4th cousin level we are contemplating a fairly extensive tree with many branches. I certainly cannot document my full pedigree back to that level and trace all of the branches down again to parallell generations. Maybe one legitimate offspring somewhere had a connection to another country but has not shown up yet in the paper trail. Am I missing something in my understanding here?
                            Lyncra, I'd like to agree with you, Elise F etc, but the following email from Michelle F suggests otherwise.

                            She states : "When someone contacts Customer Support and says that they have an unexpected match, the thought of an NPE is almost always the first thought until we know more."

                            And in her final conclusion from the same email, she states "So we get back to the ultimate question: Could there be an NPE somewhere in your ancestry?

                            The ultimate answer is that there is a possibility of an NPE in anyone's ancestry. That is why I cannot and will not tell you with 100% certainty that your family tree is 100% accurate and there can't possibly have been an NPE. You're asking me to read tea leaves, and that is not how our testing is designed."

                            To me this clearly indicates an NPE is possible and in fact the main probability. This is exactly the reason why I am having a difficult time accepting yours and Elise's etc suggestions and why MaineDNA should accept it too. After all customer service has to be reliable.

                            Comment


                            • #15
                              I have no intention of hijacking another person´s thread to discuss this NPE and customer service issue with you Handbo, so this will be my final comment here. I have no information on how FTDNA staff their customer service, but in many companies the first line of customer service personnel in the IT dept are fairly junior and inexperienced people who work from prepared scripts and refer issues they cannot resolve to more experienced people in 2nd line support. I personally would never base my whole research strategy on a comment from a customer service rep. Elise is the person who fronts FTDNA webinar services and moderates the forum. I would place more confidence in her responses.

                              Comment

                              Working...
                              X