GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2010-11 > 1290210469


From: "Alister John Marsh" <>
Subject: Re: [DNA] Odds Are, It's Wrong
Date: Sat, 20 Nov 2010 12:47:49 +1300
References: <F9C440A2-FC59-4A9E-AAAC-85DEE9D2FAB0@GMAIL.COM><4CE70443.7050108@san.rr.com><000e01cb883f$7e070ea0$c2482dae@Ken1>
In-Reply-To: <000e01cb883f$7e070ea0$c2482dae@Ken1>


Ken,

You said
>>>>>>>
> [[ and consider the alternatives? ]]
<<<<<<<

...... you mean like the truth?

"Don't know" is often a better answer than placing reliance on an answer
based on so many not well understood variables that the answer has a high
chance of being wrong.

You can say that if dozens of underlying assumptions are correct, there is
95% chance the statistical answer is correct. But if the underlying
assumptions have (sometimes high) possibilities of errors, then it is a bit
of a scam to promote the "95% confidence leve"l to the masses who are not
aware of the perhaps fragile underlying assumptions.

Closer to home, I think it is risky to assume DNA mutations are entirely
random when statistically calculating the age of haplogroups etc. They may
be largely random, but I don't think we can be confident of that yet.

John.

This thread: