Archiver > GENEALOGY-DNA > 2010-11 > 1289239062

From: Jim Bartlett <>
Subject: Re: [DNA] "Relationship Range" Confidence Level
Date: Mon, 08 Nov 2010 12:57:42 -0500
References: <><>
In-Reply-To: <>


So far I've confirmed 11 new cousins back to common ancestors, and am working on the 12th; all with Family Finder. The ancestral paths go through 34 different surnames.

Predicted vs actual:

FTDNA said 2C - turned out: double 3C (2 brothers m 2 sisters)

FTDNA said 4C - turned out: 10C-1

FTDNA predicted the other 9 as 5th or more - turned out: 6C-1, 6C-1/7C-1 (double cousin); 7C; 8C-1; 8C; two 9C; two 10C.

All but one of these is from Colonial Virginia. None of these has any block over 7.7cM in common with or overlapping any others - all large blocks are on separate chromosomes and/or at separate places. Many of the smaller blocks are shared by several matches.

Like you, I'm interested in what others are finding.

Jim - Sent from my iPhone - FaceTime!

On Nov 8, 2010, at 10:47 AM, "E. Harold Vannoy" <> wrote:

> As you mention below, they also suggest "a
> range of possible genealogical relationships." This is the "relationship
> range" that I was asking about. Specifically, I am wondering if we can
> treat this relationship range as a "confidence interval" as is sometimes
> used when evaluating statistical data? If so, do we know what the
> "confidence level" is for this interval (over several trials how often
> has this "relationship range" been found to contain the ACTUAL
> relationship - 9 times out of 10 would be 90%, 19 times out 20 would be
> 95%, etc.).
> Does anyone have enough data to begin to determine how often the actual
> relationship between two individuals is found to be within in the
> "Relationship Ranges" suggested by "Family Finder" and "Relative Finder"?
> Harold Vannoy

This thread: