I receive this question regularly from people who have taken one of the autosomal DNA tests and who expected to match a cousin, but don’t.
Of course, the Jeff Foxworthy in me wants to say, “Because he’s not your cousin,” but fortunately, I never let my inner Jeff Foxworthy out in public.
Actually, that’s often their biggest fear – that they are uncovering a very unpleasant family secret – but Jeff Foxworthy aside – that’s generally not the case.
Let’s take a look at why.
|Relationship to You||Likelihood of a Match||% of DNA Shared|
|1st Cousin (common grandparents)||100%||7-13|
|2nd Cousin (common great-grandparents)||>99%||3-5|
|3rd Cousin (common great-great grandparents||>90%||.3-2|
|4th Cousin (common ggg grandparents)||>50%||<1%|
|5th Cousin (common gggg grandparents)||>10%||Sometimes none detectable at match threshold|
|6th Cousin (common ggggg grandparents)||<2%||Often none detectable at match threshold|
If you don’t match your first cousin, then you need to start thinking about Jeff Foxworthy or you’re simply extremely lucky, or unlucky, depending on your perspective. Buy a lottery ticket.
In all seriousness, if you don’t match a first cousin, consider having your sibling (or parent) or your cousin’s sibling or relevant parent test as well. In some cases, two people simply inherit different DNA and even though they don’t match each other, they do match other people in the same family.
However, if you’re going to go down this path, be prepared that the answer may be that you really aren’t genetic cousins. By the time you get to this point, you’ve already peeked into Pandora’s box though, so it’s kind of hard to shut the crack and pretend you never looked.
Another option for determining whether or not you really match that cousin is to download both of your results to GedMatch. The testing companies have pre-set match thresholds that determine what is and is not a match. That’s a good thing, but what if your match is just slightly under that threshold, and there aren’t other relatives to test? GedMatch allows you to match at very small segment levels that would generally be considered population matches and not genealogy matches.
Judy Russell had the perfect example of just this situation in her Widen the Net blog. Her mismatch was with a 3rd cousin. According to this the chart above, she stood a greater than 90% change of matching, but she didn’t, so she’s in the special 10%. And that 10% gets left wondering. Fortunately, Judy had tested aunts, uncles and another first cousin, and her cousin who did not match her did match them.
The moral of this story is:
- Ignore Jeff Foxworthy when he starts to whisper in your ear, at least initially
- Test as many family members as you can
- Don’t jump to conclusions
- Utilize third party tools like GedMatch if necessary
- Understand that if you test enough family lines, you will eventually find an undocumented adoption someplace