On 5/29/2016 11:09 PM, joecook via wrote: > On Sunday, May 29, 2016 at 11:58:26 PM UTC-4, Thomas Milton Tinney, Sr. wrote: >> Ignorance, False Promises and Pseudoscience: Is This >> Profit Promotion of DNA Fiction by Senior Genealogists? >> >> In 2013, The Church of Jesus Christ of Latter-day Saints suggested "New Technology Makes Family History Easier, Even Fun", noting "An interesting development in family history research is the use of DNA testing to discover one’s ethnicity." >> https://www.lds.org/church/news/new-technology-makes-family-history-easier-even-fun?lang=eng >> > <snip> > > Your conclusions are flawed and the primary reason is that your references are seven to nine years old, which is significant when dealing with new technology. Ancestry.com has built up an enormous database that gives a surprisingly good ethnicity estimate that attempts to reach back about 300-500 years. > > I am English, German, French, Italian, Irish, Dutch, etc, etc. and my ethnicity (country of origin, really), percentages for each are extremely accurate on ancestry.com. The same is true for my English, Irish, Polish, French, German, (etc), wife. > > They have built up a mathematically significant sized database, and dedicating increasingly larger amounts of processing power (mainly an iterative algorithm) to the problem and making very good progress. One of the problems with the concept of "nationality" is that it is fuzzy. With regard to ancestry, it is also very time-dependant. I can state with near certainty that measuring from 100 years ago, I am of 100 percent American ancestry, because all of my ancestors then living were natural born American citizens. I can also state with high confidence that measuring from a couple of million years ago, I am of 100 percent African ancestry (as are we all). With reasonable confidence (but more ambiguity), I can state that measuring from the year 1800, my ancestry was about 1/2 (or a bit more) American, 1/4 German, 1/8 English, and 1/16 "Irish" (i.e., Scotch-Irish, so presumably Scottish a century or two before that), perhaps also with a smidgen of Dutch (the latter a guess based solely on a surname). In the last case, gaps in documentation produce some uncertainty, but the percentages should still be close. Based on my autosomal DNA, the "My Origins" page at FamilyTreeDNA shows my ancestry as 60% British Isles, 31% Western and Central Europe, 6% Eastern Europe, and 3% Middle East, which isn't that far from what I would have guessed from my own research, although I have no idea where the Eastern European or Middle Eastern would come from. For my ca. 1600 ancestry, I would have guessed something like 2/3 British Isles and 1/3 Western and Central Europe, and other models give some different percentages. Such differences are not too surprising, given that we are trying to answer rather ambiguous questions using a science which by its nature is inexact and still in its infancy. However, it is not clear that the testing companies are adequately explaining these limitations to their customers. Stewart Baldwin