Archiver > GENEALOGY-DNA > 2005-06 > 1118868031

From: <>
Subject: Re: [DNA] Heresy!
Date: Wed, 15 Jun 2005 16:40:31 -0400


Very well put. I know nothing about these Scottish clans, but I do know it's human nature to get too
wrapped up in wanting to believe something. I think we are all guilty to one degree or another of getting
the results we want from the numbers.


> From:
> Date: 2005/06/15 Wed PM 04:30:34 EDT
> To:
> Subject: [DNA] Heresy!
> First let me state that it's my position (and I believe Clan Campbell's) that the origins of the earliest
Clan Chiefs are shrouded in mystery. I don't think that the Clan has taken a definitive position one way
or another.
> Peter essentially asks "how can that be when DNA data indicates a strong tendency toward R1b?"
> Well, in my opinion the ... (here it comes) ... the DNA evidence is wrong! Or more precisely, our
interpretation of the DNA evidence is specious. Now before everyone gets wound up, let me explain.
> First, I believe that the many surname Group Administrators tend to over analyze their data. For
instance, the Campbell project -- which at 76 results returned has become a sizeable project -- has
the following breakdown.
> 55 samples (72%) cannot trace their oldest proven ancestor beyond the USA
> 10 samples (13%) have a proven line to Scotland
> 5 samples (7%) have a proven line to Ireland
> 6 samples (8%) have a line traceable to Canada
> Clearly this sample size is biased towards the USA and not necessarily representative of the lines
that existed a thousand years ago. In addition, I believe that many of the studies based on
YSEARCH.ORG data base are probably biased in the same manner. I suspect that the vast majority of
FTDNA participants live in the US.
> Second, the samples in most studies are not independent. For example I believe that 4 of my 6
Canadian samples (which have a documented common ancestor) were collected because a single
researcher reached out to his known cousins.
> With all due respect to some of my fellow researchers -- while publishing DNA signatures of
Somerled and Colla Uais is fun, we have to take this information with a grain of salt. In most cases
the written record of these individuals is suspect at best.
> I also have to admit that I am as guilty as the next fellow. Based upon some of the Campbell
Project's "ancient" lines I have publicly speculated about the DNA signature of Sir Colin "Iogantach"
Campbell of Lochawe who died in 1412. Such assertions are fun but not scientific fact. In most cases
the supporting analysis has not been published nor has the analysis stood up to a juried peer review.
> I only mention this because I believe that many of us take this analysis too seriously. I am frequently
accosted by distraught Campbell study participants who think that because of a match with another
surname that they aren't Campbells or get overly excited about a high resolution match. In the end, we
must remember that DNA analysis is a statistical technique that is subject to uncertainty.
> Which brings me to my final point. There's an old saying in statistics -- "If you torture the numbers
enough, they will tell you whatever you want to hear!". Sometimes I feel that list members are making
sweeping generalizations based upon very small data sets. I guess my heart goes out to the
numbers, because I can hear them screaming.
> ==============================
> Search the US Census Collection. Over 140 million records added in the
> last 12 months. Largest online collection in the world. Learn more:

This thread: