GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2007-10 > 1191843474


From: Jerry Lobdill <>
Subject: Re: [DNA] 60 Minutes Genetic Genealogy Segment Up Next
Date: Mon, 08 Oct 2007 06:37:54 -0500


I have the entire segment recorded on my TiVo. It seems to me that
Leslie just couldn't produce a segment without some hint that there's
impropriety involved in genetealogy. The professor seemed to know
nothing about genealogy, science, or mathematical probability.

As a scientist with extensive experience in statistical methods it
was not difficult for me to ramp up pretty quickly when I first got
into genetealogy a couple of months ago, but the jargon of genetics
used on this list and elsewhere in this field is a formidable
impediment to the simple application of my math knowledge. I'm not
complaining, mind you, but I can hardly read this list without
extensive reference to my Kerchner's dictionary. Whew!!

I find that most people I've met who have purchased testing are much
like the Wests on the 60 Minutes segment. They are not statisticians
and do not know much about probability. We are not doing a very good
job of educating them in our discussions on this list. Perhaps that
is not the purpose of the list, but nevertheless, the average tested
individual doesn't find much help here or anywhere else that I've found.

The usual approach I see is one in which the explanation of a
person's test results is confined to giving a diagnosis with no
tutorial discussion of how it has been decided that if there is an
11/12 match for this particular case, that gives an X% confidence
level that the two people have a recent common ancestor. It's like
giving someone fish but not teaching them how to catch them.

Both Leslie and the law professor exhibited a complete lack of
understanding of the results of the tests from the different
companies, but the results seemed not at all surprising to me as a
person who is schooled in the math behind the results. They seemed to
think the results were inconsistent and perhaps dishonestly
interpreted by the companies involved. This is unfortunate, but a
common attitude that can only be changed through education.

The professor's graph and his complaint that DNA is not giving a
complete story of the entirety of a person's ancestry showed a lack
of understanding of how genealogy is pursued and the kind of
questions that DNA testing can answer.

The author of the most recent genealogy of one of the surname
families I am researching recently refused to help me spread the word
about a DNA project I administer because the technology is not yet
good enough to answer lineage questions definitively. Duhhh!! We
have to find a way to educate people like this.

As one of my math profs used to say, "When you're dealing with
probability, you'd better be willing to take a chance." :-)

Jerry







>Date: Sun, 7 Oct 2007 21:11:03 -0500
>From: "grandcross" <>
>Subject: Re: [DNA] 60 Minutes Genetic Genealogy Segment Up Next
>To: <>
>Message-ID: <003f01c80950$7af7c130$>
>Content-Type: text/plain; format=flowed; charset="iso-8859-1";
> reply-type=original
>
>
> >
> > While I appreciate the publicity, I thought the slant was negative and the
> > information not adequately explained nor completely accurate. I have
> > emailed
> > CBS about it.
> >
> > Anne
> >
>
>
>It was unnecessarily critical and clearly biased. I don't know what the
>Stanford Law professor was trying to prove. His comments were so filled with
>conditional language and unexplained innuendo that they were useless. I
>guess his point was that companies selling their services to the genetic
>genealogy community were guilty of hype and this was intolerable because
>they were "using science" to further their sinister goals. The problem is
>that nobody else was complaining, certainly not those who were clients of
>these testing labs. We were never told what constitutes hype.
>
>Leslie Stahl flatly asserts Thomas Jefferson fathered Sally Hemmings' child.
>Then, "....or it could have been...." Thanks for reinforcing the myth of
>paternal certitude.
>
>Unfortunately, this is SOP for 60 minutes.


This thread: