One example is anecdotal and does not always prove as much as we would like, but in this case it does make the point that the compiled trees on the AncestryDNA circles are, well, suspect.
The “James Rampley DNA Circle” indicates that his mother’s year of birth is known: it is not.
It is indicated that James had 11 or 13 children with Elizabeth Chaney. He did marry Elizabeth Chaney–they did not have anywhere near 13 children.
There is absolutely no evidence that James and his family were ever in Texas during the Civil War as this page on James states. They were enumerated in the 1860 census in Hancock County, Illinois, still owned property there (not direct proof of residence admittedly), paid property taxes there (not direct proof of residence, either), all three sons who enlisted in the US Civil War were Illinois residents at the time. I’m not certain where this little residential detail came from.
As I’ve mentioned in earlier posts, the AncestryDNA circles are helpful for sorting matches. But I just wish that Ancestry would not tie unsubstantiated compilations from user-submitted trees to their DNA results. DNA is supposed to be irrefutable–we’ve all heard “DNA doesn’t lie.” DNA does bring out the secrets no one wanted to tell you. DNA is science–based on sound research, evidence, and analysis. You can algorithmically determine if two individuals are related based upon their DNA. It’s essentially a matter of a computation.
And that’s where these algorithmically created compilations fail.
You can’t just take a bunch of compilations on one dead person and algorithmically create a summary on that dead person that is correct. Genealogical research is a little more nuanced than that.
Keep the circles. Throw out the compilations. It’s hard enough to fight this stuff without Ancestry.com generating more of it and the appearance of it on the DNA pages makes it seem more credible to some researchers.
After all, you can’t argue with DNA.
But you sure as hell can argue with the ancestral compilations on the AncestryDNA circles.