How unique are we?

[Pauline]

My understanding is, with English names at least, that a name along with full date of birth makes us, to a large exrent, uniquely indentifiable. Of course, it's not invariably the case but is nevertheless generally so.

This came up when investigating one of my grandmother's step sisters - Edith Smith - in the 1939 register. Previous research suggested she had married a Mr GPH in 1913 but I came across descendants of Mr & Mrs GPH who were adamant that their Edith Smith could not be the same person as mine.

Now I find that the dob of Mrs GPH matches that of my Edith Smith exactly, which seems to confirm my earlier thinking.

Any thoughts?

 

[Tim]

What was their reasoning or proof that your Edith and their Edith were not the same person?

 

[Pauline]

Largely, I think, because they didn't recognise any of the names I gave them from my family. But that doesn't surprise me - my grandmother and her siblings did not get on with their stepmother, and to the best of my knowledge didn't keep in touch with their half siblings, let alone their step siblings.

Before the 1911 census was released, I was unaware there had been any half and step siblings, and because there was no actual marriage, even of the name of her step mother.

 

[Tim]

All you can do is to show them your proof and reasoning. If they refuse to accept or believe it then you've tried your best. They will obviously be happy in their ignorance.

 

[peter]

With a surname like Smith there are bound to be people with exactly the same name who were born on exactly the same day. For example, when I picked 1891 (at random) - I found there were 73 Edith Smiths (with no middle name) whose births were registered in that year and, as we all know, with 23 or more people in a room there's a better than 50% chance that two of them have the same birthday. Another way to look at is to say that if you take one of those Edith Smiths at random, there's about 1 chance in 5 that there's another Edith Smith who shares her birthday (because 72 * 5 = 360).

This is simply an extension of the problem faced by those who rely only on IGI data, or only on free information - it's inevitable that there will be plausible, but incorrect, answers to many of their questions. We faced the same problem before indexed parish registers started to be made available online - and in many counties we still do.

 

[Tim]

There's nothing wrong with your reasoning and statements Peter, but if you factor in a marriage as well?

What are the odds of 2 Edith Smiths, with the same birthday who both married a Mr PGH?

Pauline, so perhaps addresses and location may help these people believe?

 

[Pauline]

All you can do is to show them your proof and reasoning. If they refuse to accept or believe it then you've tried your best.

I think they already have refused. I have my suspicions that Edith's illegitimacy and family background made her persona non grata on their tree. My attempts to continue the discussion were met with a deafening silence.

 

[Pauline]

there's about 1 chance in 5 that there's another Edith Smith who shares her birthday

That would mean that even with a common name like this it's 4 times more likely that there won't be another Edith sharing her exact dob than that there will.

But even with a much rarer name I certainly wouldn't suggest using the date of birth as the sole evidence of a person's identity. I was thinking more of it supporting other evidence which in itself had made a person's identity highly likely, but was not wholly conclusive.

 

[Pauline]

I've just followed this up with one of Edith's sisters - an Emma Ann Smith. To the best of my knowledge Emma did not marry, so in the 1939 search I entered Emma Smith and her exact date of birth. Up popped just one intriguing entry showing her as single but living with a Frederick G Smith, and in the right sort of area.

Further searching on Frederick using his exact date of birth shows a match with Emma's father's younger brother. (Perhaps it would be more correct to describe him as brother of her mother's husband as I don't know for certain who her father was.)

Anyway, all things considered, it seems quite likely I have found the right Emma by searching with name and birth date.

 

[peter]

What are the odds of 2 Edith Smiths, with the same birthday who both married a Mr PGH?

Very small - but that isn't what we're talking about, surely? I thought the dispute was over the identify of the Edith Smith who married PGH.

 

[peter]

That would mean that even with a common name like this it's 4 times more likely that there won't be another Edith sharing her exact dob than that there will.

That's not a helpful line of reasoning - I wouldn't put something in my tree if I was only 80% convinced it was correct.

 

[Tim]

Very small - but that isn't what we're talking about, surely? I thought the dispute was over the identify of the Edith Smith who married PGH.

Good point, and well made. It's difficult for us to say much as we didn't do the initial research.

 

[Pauline]

That's not a helpful line of reasoning - I wouldn't put something in my tree if I was only 80% convinced it was correct.

Nor would I!

As I said above, I was only thinking of it as supporting evidence where other evidence had already made the person's identity highly likely.

Sometimes it just isn't possible to be 100% certain you have got things right so the more strands of evidence you can find in support of your supposition the better. And by quoting your sources and reasoning along with stressing the element of doubt, you give others the chance to make up their own minds.

 

[Bob Spiers]

Sometimes it just isn't possible to be 100% certain you have got things right so the more strands of evidence you can find in support of your supposition the better. And by quoting your sources and reasoning along with stressing the element of doubt, you give others the chance to make up their own minds.

I have only just discovered this quite fascinating topic and have little to add to what has gone before. However I must agree wholeheartedly with Pauline on the above quotation extract. I use my Tribal Pages (on offer to family members who are interested) to set out situations where I cannot be 100% certain I have alighted on the right ancestor. I give reasons why I'm as sure as I possibly can be, then as Pauline says, anyone interested can make up their own minds, or research further if so inclined.

With my Ancestry Public Tree I am prone to leaving extensive comments on the Ancestor page, or the nearest ascendant or descendant if applicable. People can read and make up their own minds, and hopefully get in touch if they agree or disagree and either adds to my knowledge base. I only recall one such contact who as it happens agreed they had reached the same conclusion.

 

[Bryman]

I wouldn't put something in my tree if I was only 80% convinced it was correct.

Where else would you put it? Surely that is fine as long as there is nothing better to replace it and the doubt is documented in the tree.

That gets a little awkward if there are several possibilities but then it is unlikely that any of them would be 80% certainty.

I have a few instances where my guess is that there is 5-10% doubt but with no other known alternative. At least having the known details in the tree makes it easier to investigate for other alternatives and lessens any liklihood of something being overlooked.