I have had the rare privilege of having four generations of family tested through AncestryDNA. I have no regrets at all about this testing, and I am grateful to all my family members who tested. It’s been fascinating to see the numbers behind our relatedness, and our participation in this test helped draw my family closer. Our test results also helped confirm the identity of one of my paternal great grandfathers, which sparked a family reunion eighty years in the making. While I would not want to discourage anyone from taking the test, I also want others to know that there are significant limits to the precision of matching. I hope that the science behind matching improves over time, but there will necessarily be limits to the matching of distant relations, no matter how accurate a company’s matching algorithm may become. At the level of third cousin and beyond, there is a 10% and growing chance that you may not share enough DNA with a relative to be matched to him or her.
After both of my parents had been tested, my confidence in AncestryDNA’s matching declined. I have 240 high confidence matches and more than 7,000 matches overall. Ancestry does not provide shared match information beyond the high confidence range (“4th cousins or closer”). All of my high confidence matches should be shared with my mom or my dad. There would be the rare possibility that a few could be closely matched to both of them, i.e. distant double cousins to me. I have been disappointed to discover that 31% of my close matches aren’t matched to either of my parents. I have also noticed that there have been some matches shared between one of my parents and my daughter, where I am “skipped” as a match. In a few cases I have found that a close match to me is only a distant match to one of my parents. While there is a remote possibility that a close match to me is distantly related to both of my parents, it seems impossible for example that one of my maternal relatives would actually have more DNA in common with me than with my mother. Overall, I’m seeing significant amounts of missed and coincidental matches.
This test can help confirm close family relations, second cousins and closer. For more distant connections, I suggest that such a test be used for leads for further traditional research instead of proof alone of relatedness.