Archiver > GENEALOGY-DNA > 2010-02 > 1266947674

From: Michael Maddi <>
Subject: Re: [DNA] : low variance MRCA dates for P310 clades in ItalyandSEEurope
Date: Tue, 23 Feb 2010 09:54:34 -0800 (PST)

Ken Nordtvedt wrote:

[[ I have my doubts the real statistical confidence intervals for these
TMRCAs permit reaching conclusions of the type you suggest. Remember also;
one can serially try dividing the map E versus W, then N versus S, then NW
versus SE, then SW versus NE...... Eventually, because of TMRCA noise in
the regional numbers, one might eventually "see" a trend in one of the
groupings. Then throw in the fact that if just a not too large sampling of
a haplogroup migrates to a new area, it's variance in the new area should
not necessarily decrease; it could just as well increase; its statistical
confidence interval about expected value goes up even more. Ken ]]

My reply:

This is exactly the point I would make in answer to Vince's reply to Alan just before Ken posted. Vince noted that the variance of the 6 R1b-U106 members of the Italy Project is higher than that of 35 Central European members of the R1b-U106 Project.

Ken's point above in regard to "a not too large sampling of a haplogroup [that] migrates to a new area" having equal or greater variance compared to the haplogroup in the original area where it migrated from is how I would explain what Vince noted in his posting. I regard R1b-U106 as a mainly northern European subclade. So its presence in Italy, at low levels, is the result of migrations/invasions into Italy from various parts of northern Europe, for instance the Lombards or the Normans.

Mike Maddi

This thread: