Tuesday, June 19, 2018

Ancestry.com Suggests "Potential Ancestors" For your Ancestry Member Tree

Kris Hocker, on her A Pennsylvania Dutch Genealogy blog, today alerted us to "New in Ancestry -- Potential Ancestors."

1)  I went looking in my Ancestry Member Trees and, sure enough, there are some "Potential Father" and "Potential Mother" boxes.  Here is one instance of it:


2)  Ancestry is trying to help me by suggesting parents for my 4th great-grandfather, Philip Row (1752-1817).  I clicked on the box for "Potential Father" and saw:


3)  And then I clicked on the green "Review Details" button and saw the information for the potential father:


And further down the sidebar for the potential father:


4)  At the bottom of the information for the potential father is a question "Do you think this is the father of Philip Row?"  I clicked "Yes" to see what would happen, and saw:


Poof. Just like that I now have a father for my 4th great-grandfather.  But wait, where are the records for this relationship?  Is there a birth or baptism record, a land or probate record that mentions his son Philip in New Jersey?  Well, not that I've found in the last 30 years of researching this family.

5)  There are plenty of Ancestry Member Trees, and other online trees, that claim this relationship is true.  But I don't believe that there are ANY records that support the relationship.  In fact, the "Potential Father" information above lists other sons named Philip - there's one named Philip Rowe (1761-1844).

6)  There are Ancestry record collections that suggest the relationship - here's one that's on Philip Row's Hint list:


So this Philip J. Rowe was born in Hunterdon County, New Jersey in 1757 to Johannes and Catherine (Lescher) Rowe.  But Johannes Rowe's family resided in Dutchess County, New York.  And this record says that Philip J. Rowe married Mary Smith on 4 Dec 1791 and died in 1817.

7)  However, the Revolutionary War Pension File for Philip Row of Hunterdon, New Jersey indicates he was born in December 1752, not 1757.  The Pension File also listed his marriage date as 9 July 1772.  The Pension File also lists his death date.  If the "Family Data Collection" record above is to be believed, he would have been 15 years old when he married and 16 years old when he had his first child in 1773.

8)  The reality is that a user cannot trust this "Family Data Collection" record because some of the information is wrong and it doesn't provide any source citations.  

So I quickly deleted the "Potential Father" from my tree and Ignored the Hint for both the potential father and potential mother.

9)  I looked through my tree for other "Potential Parents" and found some, and I have not found one yet that is supported by records and that I might add to my tree.  

10)  How does Ancestry.com determine the potential relationships?  There are 17 other Ancestry Member Trees for Philip Row (1752-1817), and ten of them postulate these parents.  As shown above, there are some record collections that are family compilations that show the relationship.  So there is a preponderance of evidence, right?  It is very likely that Ancestry.com creates "combined profiles" for many historical persons and claims that they are accurate.  Ancestry's We're Related mobile app is full of "combined profiles" that are inaccurate.  I have lots of them.

The bottom line for me is that I don't trust Ancestry Member Trees, combined profiles, family compilations from personal trees, etc.  I do use  online family trees with source citations to actual records and relationships that make sense to me as hints or clues.

11)  The FamilySearch Family Tree does not show this relationship, mainly because I have modified the profile for Philip Row and no one else has edited the profile.  The WikiTree profile reflects my research also, and no one else has modified it.  The Geni.com profile for Philip Row reflects my research, and no one else has modified it.

12)  Thank you to Kris Hocker for alerting me to this new Ancestry.com feature.

                                      =============================================

The URL for this post is: https://www.geneamusings.com/2018/06/ancestrycom-suggests-potential.html


Copyright (c) 2018, Randall J. Seaver

Please comment on this post on the website by clicking the URL above and then the "Comments" link at the bottom of each post.  Share it on Twitter, Facebook, Google+ or Pinterest using the icons below.  Or contact me by email at randy.seaver@gmail.com. 


7 comments:

Elizabeth Handler said...

I noticed this a couple of days ago - a known 4th cousin of my husband has their common third great-grandmother's name as "unknown" which appeared as a Potential Ancestor in my tree. This is a Jewish woman who lived and died in Galicia (now Ukraine) in the early 19th century. I don't know if we'll ever determine her name. Not helpful if Ancestry is just sending you to someone else's tree.

Kris said...

You're welcome. I've been watching for it after seeing mention of it on Facebook.

It's an interesting idea. I don't know how well it will help those whose trees stretch back into the 1700s already. Maybe it works better for later potential parents from the 19th and 20th centuries. I sure would like to see source information or attached records, though.

~Kris

Diane Gould Hall said...

I saw this last week, Randy. I checked one that they suggested in my tree and found it to be highly unlikely, if not completely wrong. I don’t trust these hints either and will review them carefully, as you did.
Wouldn’t it be nice, though if they added a parent or parents that broke down one of my brick walls?

Gayle said...

I too caught this last week and had an immediate negative reaction. Just too much potential for error resulting in many many more unproven trees.

Dennis Lohr said...

10) How does Ancestry.com determine the potential relationships? There are 17 other Ancestry Member Trees for Philip Row (1752-1817), and ten of them postulate these parents.


Carry this forward several years, and those 10 Member Trees will grow exponentially as others accept this same suggestion. Then what may have been a relatively small number of trees that gave one pause before selecting the relationship, now seems much more plausible because 50 or 100 trees have accepted it. And the ease with which someone can add this “potential” is what makes it so dangerous, Genealogical Proof Standard be damned.


It didn’t take me long to remove other trees from my hint results. This change only strengthens my decision to do so, but I am frustrated by the prospect of having to do additional vetting of poor research in the future.

James Alexander Knighton said...

I was immediately sceptical when I saw this feature appear, and I still dislike it, but funnily enough the two times I tried it out actually did lead me to discovering what I consider to be the actual parents. Regardless, I have disabled the feature since I don't like the way it looks.

Randy Seaver said...

Dennis Lohr, excellent comment and you're spot on.

We can only do the best we can, and try to convince other researchers of the accuracy of our work. That's one reason I write my blog, because someone may be searching for the information I have that many of the others don't have. For instance, a researcher might Google Philip Row and find this blog post, or the blog post that gives his Revolutionary War Pension file transcriptions, and see the available evidence.

And Ancestry.com will continue to promote their One World Tree used for the We're Related mobile app with "consensus" relationships. At least on FamilySearch Family Tree, we can edit the profiles and try to convince other users based on the available evidence.