SANDERS-DNA-L Archives

From: "Justin M. Sanders" <>
Subject: [SANDERS-DNA] Valuable Lesson
Date: Thu, 13 Mar 2008 23:02:45 -0500

This past week an event happened in the Sanders YDNA project that holds
a valuable lesson for all of us DNA folks.

The lesson is that DNA should be treated as one kind of evidence among
many that we must use in genealogy. It can supplement, but can't
replace, traditional genealogical research. The particulars of the
lesson are these (as I understand it):

One of our members (call him "A") was tested a couple of years back, and
suspected that he was a descendant of Moses Sanders of Georgia. The
"Randolph Group" also suspected that Moses was kin to them. But when
"A" was tested at the Y37 level, he did not match with the "Randolph
Group". We assumed that we just weren't very close kin and worked on
other avenues of research. However, over the last year, a couple of
reasonably-well-documented descendants of Moses were tested and they
*did* match the Randolph Group. So Moses was related to the Randolph
Group after all, and we could conclude that "A" was just unlucky and
that despite the hints in the documentary trail, he wasn't really a
descendant of Moses. Up to here, all this is the normal story in the
YDNA business.

What's new is that last week, we learned that "A's" test back in 2006
really did match the Randolph Group, but it had been entered into the
database (by the laboratory) incorrectly. The "new" (not really new,
just correct) numbers matched "A" with the Randolph Group, and
therefore, his documentary evidence that pointed to Moses as his
ancestor was correct along.

My purpose is not to complain about this, but rather to show that DNA
testing involves human beings, and people do make occasional mistakes.
FTDNA has over 100,000 kits, and if they average 25 markers, that would
be 2.5 million markers. Even if they were 99.99% accurate, there would
still be 250 incorrect markers out there. The lesson is that if you
have some decent documentary evidence that indicates a certain line of
descent, and if the DNA is off by a little bit, there is the possibility
that the problem is the DNA test and not the traditional genealogical
evidence. In such a case, it is not unreasonable to ask for FTDNA to
double check the results-- my understanding is that they will work with
you to do this, if you have convincing reason to believe that something
is wrong with the DNA test. In our case, we should have done this with
"A"'s test, rather than sit back and assume that the DNA test had to be
right and the other evidence had to be wrong.

Justin M. Sanders

This thread:

• [SANDERS-DNA] Valuable Lesson by "Justin M. Sanders" <>