Wednesday, March 19, 2014

Dear Randy: Why Don't You Attach All Pertinent Records to Your Ancestry Member Tree?

I knew someone would ask this question - and a reader did it after I posted Mining the Ancestry.com "Massachusetts Town and Vital Records, 1620-1988" Collection on 17 March 2014.

I used Ephraim Sawtell (1735-1800) in my blog post, but it could be almost anyone in my Ancestry Member Tree.  


1)  I had 9 Hints for Ephraim Sawtell - I accepted 2 Hints and ignored 7 of them.  The 7 Hints that I ignored consisted of:

*  "Ancestry Family Trees" -- this is other person's work, and I don't attach them because they are compilations rather than actual records or sources.  I do explore them (sometimes) to see if there are links to records I haven't found.  For Ephraim Sawtell, I had the birth and marriage record, and nobody has a death record.

*  "Massachusetts Marriages, 1633-1850" -- this collection is an index, taken from an FHL microfilm.  I have the Groton, Mass. Vital Records book page as the source and an image of the page already in my tree.  I don't need this index, it's superflous.  At best, it's a "finding aid" to lead me to the FHL microfilm, the vital record book or the actual town record book. 

*  "New Hampshire Marriage Records, Index, 1637-1946" -- this collection is an index, and the Hint is for a marriage of an Ephraim Sawtell to an Abigail Farnsworth in 1822.  My Ephraim died in about 1800, and I really doubt that he married in 1822 at age 87 (although stranger things have happened).

*  "American Genealogical-Biographical (AGBI) Index Record" -- this collection is a name index which lists the birth of Ephraim in 1734.  It applies, but I have better records than this which shows only the year, not a date or place of birth.

*  "Massachusetts, Town Birth Records, 1620-1850 Record" -- this Hint is an index of birth entries that gives Ephraim's name, his parents, a birth date and a birth place.  Again, I have a published vital record for the event and an image of the book page.

*  "American Genealogical-Biographical (AGBI) Index Record" -- this collection is a name index which lists the birth of Ephraim in 1734.  It seems identical to the previous Hint from the same database.

*  "Public Member Photos" -- this is an image of a 13-star flag that denotes an "American Revolutionary Ancestor."  I choose to not use it because it's not a record of service, and is not a photograph of the person or anything connected to the person.

My opinion is that all of those ignored Hints are Derivative Sources (e.g., indexes) or Authored Works (e.g., family trees) and I prefer to not attach them to my Ancestry Member Tree persons.  I would rather attach actual records (preferred is an Original Source, but Derivative Sources are acceptable) that provide Primary Information and Direct Evidence (e.g., a birth date and place name, rather than just a year and a state) to persons in my tree.  

I do make the exception for some Authored Works, such as an article in NGSQ, NEHGR, TAG or an authored book that I deem to be trustworthy.  I use them as sources, and will attach them when I have the opportunity.

2)  There are other Ancestry.com databases that I don't use - they weren't on the list for Ephraim Sawtell, but they are on the list of Hints for many of my ancestors.  They include:

*  "Family Data Collection - Births Record"
*  "Family Data Collection - Marriages Record"
*  "Family Data Collection - Deaths Record"
*  "Family Data Collection - Individual Records Record"

*  "U.S. and International Marriage Records, 1560-1900"
*  "Millennium File Record"
*  "American Marriages Before 1699 Record"
*  "U.S., New England Marriages Prior to 1700 Record"

Some of the indexes on this latter list are very useful as "finding aids" - if an actual date and place are listed in the index, that usually means that I should be able to find a better source for the event.

3)  I want my tree to reflect names and events from actual records and relationships that are verified through evaluation of a number of sources.  I don't need 4 or 5 birth records from indexes when I have a definitive record image - that's just duplication of effort and craps up my tree.  They are not independent sources - they are Derivative Sources from an Original Source.  Quantity doesn't count, Quality does.  I try to find the Original Source, and use a Derivative Source when I can't find one.  

4)  I would love to see Ancestry.com institute some sort of "grading process" where I could select to see different types of Hints to accept, reject or ignore.  At present, there is no "reject" category - where a record doesn't apply to the person that it was provided for.  I think that there should be.

My experience with Ancestry.com Hints and Suggested Records is that they are 80 to 90% accurate, unless it is a really common name.  The bigger problem is that spelling variations hide some possible Hints and Suggested Records, but I guess we can't help that.

A "grading process" scale for a user to select from might be:

*  Original source records
*  Authored works derived from original records (books, periodicals, transcriptions, extracts, etc.)
*  Indexes created from derivative sources or authored works
*  Indexes created from other indexes (e.g., AGBI, Family Millennium File, etc.)
*  Ancestry Member Trees and Member Photos

My experience is that only 5 to 15% of the Ancestry.com Hints I find for my New England ancestors fall in the first category, and another 10 to 20% fall into the second category.  Those are very useful and helpful, in my opinion.  The fourth and fifth categories are pretty useless for me, but often are more than 60% of the total number of Hints and take significant time to mark them as "Ignore."  For some of my tree persons, I've had 30 to 40 Hints to accept or ignore, and many are a waste of my time to evaluate.

The URL for this post is:

Copyright (c) 2014), Randall J. Seaver


5 comments:

Dana Leeds said...

I really appreciate your post and learned quite a bit about some of the titles of Ancestry's sources. Some of these titles I already avoided, but now I'll know I can use some to find an original source! And, I'll know to avoid some others. Thanks... I learned a lot in just a couple of minutes reading your post!

Lauri said...

Good points. Now I am going to add a new clean up task to delete the duplicate source records that are crapping up my tree. I know I have a lot of them when I have the original source. Clean up and genealogy just never end.

Fax said...

Like you, I'm concerned that we have no options in responding to shaky leaves or other hints. Often I find erroneous information that I'd love to correct or at least comment upon, rather than simply rejecting without cause.

Shirley said...

Great article Randy. Thanks.

To your list of non-records for New England I would like to add these compilations:

Mayflower Births and Deaths, Vol. 1 and 2,
the profile part at "Profiles and Historical Records", AND
Passenger and Immigration Lists Index 1500s-1900s.

The transcriptions often have "before" and "after" that appear in the printed index but people seldom click on the image to actually read it, depending instead on a straight copy from the index that appears.

The discontinued LDS IGI (index) data is just being recycled in the form of the Green Leaf Hints. Most of the hints were compiled by a computer sweep through submitted trees and family group sheets from years ago when almost every beginner used the LDS site and copied straight off it because they thought "if the Mormons have it on their site it must be true."

These hints in many cases are turning out to match exactly (even the errors match exactly) the submittals in the IGI file which is still accessible on the new Family Search (LDS) site if you know where to look. I go there to check out duplicated mistakes. They've taken off the feature that allowed us to see what date the AF or PRF was first submitted which used to be a big help in seeing who the first person to invent a date or person was.

I have been adding comments in the left column on a lot of the green leaf hints I've been getting--trying to send people to the authorities or documentation for each event. If I find an image that doesn't belong to that person I click on it and leave a comment on the View Media page. That comments get sent by Ancestry to everyone who ever saved that image. It's about the only way to get the word out because beginners don't know how to use message boards anymore.

Unknown said...

>> A "grading process" scale for a user to select ...

That would be lovely. I think it's useful to set the bar low when you're trying to figure out who is who, so I go ahed and attach the source.

Now that I'm in "cleaning" mode, I want to start pruning. and ancestry.com is really just not helpful. Instead, I'm recreating my tree into wikitree with primary or researched-secondary references. It's slow going, but worth it. Having decided to stop treating my ancestry.com tree as authoritative, I'm liberated from a lot of the "wants" I would have if I were trying to clean it up, but I'm also at a point where I don't want hints that are closer to rumor than verifiably true...