Search for content in message boards

Interpreting DNA Matches

Interpreting DNA Matches

Posted: 1 Sep 2012 11:41AM GMT
Classification: Query
I just looked at a DNA match "Confidence: 95%" that compared my John White from Delaware (DE) to another tree that placed him in Germany, since "DE" is the country code for Federal Republic of Germany.
Now isn't that silly!
And it wasn't even a specific area of Germany, just "Germany".
So, obviously, AncestryDNA isn't comparing our DNA at all, it's just trying to pair up people from similar areas based on the family trees we've provided for them online at Ancestry.com.
That's useless and has nothing to do with DNA testing.

Either I get my raw data so I can use it to compare it at a legitimate DNA site or I want my money back.
I feel totally ripped off.

Re: Interpreting DNA Matches

Posted: 1 Sep 2012 12:22PM GMT
Classification: Query
I don't think they are just comparing trees. Many of my matches have no matching surnames, so they must be showing matching DNA. Their ability to match surnames leaves much to be desired, though. I don't see that it could give you a "match" unless you have two people with the exact same name and close to same date of birth. My only matches have been with trees that both have our common ancestor. I had a surname match for one person and knew that surname could be traced back to a single ancestor in that area. I was able to connect her ancestor with my ancestor in MY tree, but because she does not show our common ancestor, Ancestry cannot make the match.

Depending on how long your White family has bee in America the match may not be so funny. What is the relationship they predict for you and the other member? White it a very common name. Do thy give you any other surnames in common?

Re: Interpreting DNA Matches

Posted: 1 Sep 2012 12:47PM GMT
Classification: Query
After going through many of the "matches" that Ancestry listed, I started wondering if they were really comparing DNA or just people in our trees with similar locations. I got many unhelpful matches like- two different people lived in the Bronx, or Pennsylvania, not even specific counties or towns.

The "95% confidence" match of John White in Delaware and someone in Germany confirmed my suspicions. That was a fluke in confusing "DE" (Delaware) with "DE" (Federal Republic of Germany). And, again, no specific area or town. Yes, thousands and thousands of years ago we ALL came from similar migration paths from Africa, and we're all related if you go back far enough, but that isn't "recent enough to be important parts of your family story" (according to the AncestryDNA hype).

Before I go on any more wild goose chases, I'd like some hard facts from the AncestryDNA team about my actual DNA results, not their feeble attempt to match up people we've already provided them with.

Re: Interpreting DNA Matches

Posted: 1 Sep 2012 1:12PM GMT
Classification: Query
Two things struck me in reading your post that may or may not be helpful:
1) The person whose tree you are reviewing may have put in the wrong location. As others have pointed out, there is a lot of garbage out there in people's trees.
2) The name/location match is totally dependent on the information provided in your and the other person's tree. It doesn't mean that the person named in a location match is your DNA link to the other person. It is up to us to review the document trail to determine where the match may be. I have had to extend my tree to siblings of 9th g-grandparents in order to find matches.

Re: Interpreting DNA Matches

Posted: 1 Sep 2012 1:18PM GMT
Classification: Query
You cannot rely on Ancestry to make your matches. I just found a match this morning and it appears to be correct, but I also looked at the surnames and places suggested and found we had another ancestor in common in another line. Ancestry may have made an error linking your ancestors with a very common name, but your DNA matches and the confidence level should be related to the DNA match, not the tree. You need to look at the surnames and places and see if you see a correct match if you feel the on generated by Ancestry is wrong.

Re: Interpreting DNA Matches

Posted: 1 Sep 2012 1:53PM GMT
Classification: Query
That silly "DE" match really shook my confidence in Ancestry's results. I was really hoping for my very own cute little haplogroup to compare to other haplogroups. I really didn't want to rely on Ancestry trees. I'm sure I have mistakes in mine and a bunch of connections I tag as "SPECULATION" (needs further research) in my notes. Of course, you can't see my notes on Ancestry so the connection is still there, but that's the framework I start researching with.

If I have to go back several generations in a match to find the common ancestor, isn't that what I'm doing by trying to document my family? If both parties in a match knew that common ancestor, we would have that in our tree already. If I had that information already, I wouldn't need AncestryDNA to tell me...well, maybe to confirm it.

But seriously, that connection between Delaware and DE-Federal Republic of Germany? That just shows me that (some of?) the matches are haphazard.

OK, I'll explore some matches, send messages. MAYBE I overreacted.

Re: Interpreting DNA Matches

Posted: 1 Sep 2012 2:05PM GMT
Classification: Query
I just opened a surname match that said his ancestor was from Colony, Anderson, Kansas, born 1689 which is impossible since Kansas was not settled that early. When I opened his tree, the location was actually Colony, Virginia, and I recognized the family as one that came from Virginia to SC with some of my ancestors and some of my collateral lines intermarried with. If the surnames and place names given are of interest, do not just rely on ancestry to match you. Look at the actual tree to see if there are families of interest. It is going to take a little (maybe a lot) more work than that to find unknown common ancestors.

Re: Interpreting DNA Matches

Posted: 2 Sep 2012 1:28PM GMT
Classification: Query
The 95% confidence number means that it is likely that you have *some* common ancestor in the range listed. When you look at the details, Ancestry is highlighting surnames that are found in both pedigrees as a visual aid, not claiming that the person listed is *the* common ancestor responsible for the match.

I don't have an example in my own results, but I've seen a screen print of a case where Ancestry did identify a specific person as a candidate for the common ancestor. It showed lines of descent for the two sides. I can't dredge up the URL right now, but maybe someone else has an example.

Re: Interpreting DNA Matches

Posted: 2 Sep 2012 1:50PM GMT
Classification: Query
Edited: 2 Sep 2012 1:50PM GMT
I agree the confidence level relates to the probability that you match the person with a relationship as stated even if they got the match wrong. Attached is a screenshot of a match found in my tree by ancestry for Adam Wylie. The match is for my maternal line which was done long ago and I am working on my paternal side. I sent this to my maternal cousin and he was at first excited because we cannot prove our relationship to Adam Wylie and likely this person cannot either.

If you look below the match, there is a list of other surnames. when I checked this list, I found another match in the Patterson line. Mu cousin was able to confirm this match as correct and is most likely the correct match.

Now if we had our raw data and could see which segments we matched, we might be able to verify this as a double match if I matched other Wylies on one of the same segment. Ancestry, PLEASE let us download our raw data.

Re: Interpreting DNA Matches

Posted: 2 Sep 2012 1:51PM GMT
Classification: Query
oops forgot attachment
Attachments:
per page

Find a board about a specific topic