Blog post

Monday, October 24, 2016

Ancestry hints: The good, the great, the horrid

My Ancestry tree has a lot of shaky leaf hints which I leave unpruned and untended. I usually only look at them when I am concentrating on an individual. The more I do look at them, the more I worry about new researchers who might suppose that the hints are more reliable than they actually are.

I was considerably surprised when Randy Seaver of the Geneamusings blog recently looked at his first 400 hints and concluded that over 85% of the hints were for the specific person cited. I wondered for a minute if it was ‘opposites day’. My children often tell me it is, it’s so hard to keep track. 85% correct just didn’t feel like my experience. Not at all.

Commenters on his post suggested that the quality of hints received depended on the amount of relevant information in the tree, particularly relationship information. They also suggested that the hints are often for information that is already in the tree.

I’m not quite up for a review of 400 hints, but I could manage a review of one person and by good fortune the first person I looked at was a great candidate for the exercise.

My ancestor Ellen Fearn, who emigrated from Derbyshire, England to Australia in 1852, has 15 unreviewed hints to her name. She has good basic information in the Ancestry tree, including relationship information:

  • parent names
  • birth date and place
  • spouse name
  • marriage date and place
  • 1851 census
  • birth of her children including dates and places
  • death date and place
  • burial date and place

She also has some missing items that I know are available in Ancestry. These are things that I would hope to see in hints:

  • baptism
  • 1841 census
  • immigration records

I am going to classify each hint as bad if it relates to a different person, good if it relates to her but duplicates existing information, or great if it contributes a fact that is not already in the Ancestry tree, such as one of the three I’ve identified. Records that I’m not sure about I’ll classify as as maybe. I’ll ignore any Ancestry member tree hints (which is my usual practice) and exclude them from the tally.

The ratings only reflect if the hint is ‘accurate’ in that it relates to the right person, and if it adds anything new. I’m not looking at the quality of the sources themselves or whether images are available.

Hint

Rating

England, Select Marriages, 1538-1973

Good

Victoria, Australia, Assisted and Unassisted Passenger Lists, 1839-1923

Great

Victoria, Australia, Assisted and Unassisted Passenger Lists, 1839-1923

Great

England, Select Births and Christenings, 1538-1975

Great

Derbyshire, England, Select Church of England Parish Registers, 1538-1910

Great

England & Wales, Civil Registration Marriage Index, 1837-1915

Good

1871 England Census

Bad

1861 England Census

Bad

Bristol, England, Select Church of England Parish Registers, 1720-1933

Bad

1871 England Census

Bad

1881 England Census

Bad

1881 England Census

Bad

1841 England Census

Great

1881 England Census

Bad

1871 England Census

Bad

Overall, a third of the hints added new information. I was very pleased to see that all three of the known missing facts were included among the hints.

On the other hand, around half of the hints were just plain wrong. I could see why most of them came up – the people had similar names and, usually, were from Derbyshire. If I had accepted the hints I would have grafted together a franken-family. How horrid.

The final tally was:

Great 5
Good 2
Bad 8
Total 15

I don’t think that the ancestor I chose is an oddity. My gut feeling is that she had better than average results compared to the rest of my tree. A quick spot check of the hints for several of my great-grandparents supports this – I had “ignored” up to 80% of the hints that had been offered to me for some ancestors. Hinting may work better for some populations than for others.

My conclusion about Ancestry hints?

When they are good
they are very, very good
but when they are bad
they are horrid!

Look at those Ancestry shaky leaf hints by all means, but take care.

I feel happy with my current approach. I generally review hints only when I am concentrating on the person in question. This allows me to review them with a critical eye, as my mind is primed with what I know about the person and their family.

4 comments:

  1. Franken-family - love it!

    Tempted to replicate your experiment but time poor.

    ReplyDelete
  2. Franken-family another,for the geneadictionary? I agree that it seems understandable how newbies can wind up with such a tree.

    ReplyDelete
  3. I just Googled "Ancestry Hints" and came up with this blog.

    I have been on Ancestry for several years now, and in the past several months have been getting a lot of atrocious hints. Duplicates---several at the same time, and totally irrelevant hints (names don't even match), to name a few. I try to keep a well documented tree, but its getting so bad that I almost hate to get on Ancestry now.

    ReplyDelete