Just to clarify, I think Elizabeth Warren is a complete fraud here, because she claimed she had Cherokee ancestry and she claimed that her mother was recognized as Cherokee and suffered discrimination for it...which on the evidence is completely untrue. Warren is no more Cherokee than your average US non-Cherokee citizen, and there has been no evidence that she or her mother had any Cherokee ancestor much less one that was recent enough for her story to be plausible. My issue here is purely with how to do a math problem.
Wrong! There was no exact generation reported for her suspected “Indian” DNA.
I did not dispute that there is no exact generation reported, nor do vouch for the accuracy of the report, I merely ran the numbers. Nor do I vouch for the alleged ancestor in the report being native to North America rather than South America. I was only doing the math based on the reports claims. Based on the report at face value the math the range is 1/64th to 1/1024th.
It was a range of generations, and so the original “Indian” ancestor would be one of 992 possible persons or approximately a 0.1 percent chance that she is part Indian, assuming that her DNA from Mexico, Peru, and Colombia is actually Indian DNA.
Yes it was a range. However your math used generations outside of that range including fourth generation when the reported range was 6th to 10th and average of 8th. Moreover, summing the number of people in different generation as a denominator for DNA contribution of an individual is just an entirely incorrect approach to the problem. While all the ancestors in a particular generation with the same weight, ancestors in different generations have different weight. For example our parents each have a 1/2 contribution to our DNA by contrast our 10th generation ancestors each have a 1/1024th.
However your math used generations outside of that range including fourth generation when the reported range was 6th to 10th and average of 8th.
I don't appreciate you lying about what I used.
These are the first 5 generations:
What I originally wrote was:
32 4th great grandmothers +
64 5th great grandmothers +
128 6th great grandmothers +
256 7th great grandmothers +
512 8th great grandmothers = 992 women or 0.1 percent
and those are the 6th through 10th generations!
You then go on to say,
Moreover, summing the number of people in different generation as a denominator for DNA contribution of an individual is just an entirely incorrect approach to the problem. While all the ancestors in a particular generation with the same weight, ancestors in different generations have different weight.
and once again, you are mistaken. One and only one of those women (or men), was the original ancestor that provided the DNA. You cannot take the middle number between 64 and 1024 to determine what the odds are. That's nonsense.
But in any case, even the 0.1 percent is just an approximation, because the DNA percentages from each grandparent and/or great grandparent are not consistent from generation to generation. DNA gets passed on in chunks or segments of varying lengths. It is not always exactly 50/50.
Another fact:
You can send the exact same DNA file to three different genealogy sites who do DNA analysis and get completely different results on where your ancestors came from.