Ancestry Matches and Half Cousins.

[BristolClark]

It occurs mostly where some of the other trees from which Ancestry is taking its thrulines have:
a) only got one half of the couple in their tree
2) both people are there but not spelt in exactly the same way as you have so Ancestry is unsure whether they are the same people

Pheno

Thanks for responding. I had noticed a few had one parent and not the other, or both parents but with different spellings, and even locations, but really I hadn't thought Ancestry would not compensate for spelling differences.

You would be better going by the amount of cMs, % (Percentage) & Segments to try & determine the relationship with your match & any Shared Matches.

It may be worth trying the LEEDS Method...

https://www.yourdnaguide.com/leeds-method

Annie

Thank you, Annie. I'll give that guide a go.

Does that help ?

It certainly has. Thank you.

I have noticed a couple of mispellings, or add-ons to the names, and even some stars and other emojis which must puzzle the algorithm. Having looked back on a couple of lines that weren't matching correctly, the system has now altered their guess to including both parents, and these were on lines I started sorting out and adding descendants to, so ultimately it's on me to check, check and add to help the system auto-correct.

One of the ones that I found most surprising is that by default Ancestry uses the genealogical standard format for dates, (ie dates in the form 22 Sep 2022). Not that it is surprising that a genealogical programme using the standard genealogical format (though ironically the "British" Find my Past does not). What was surprising to me at least, was that most importantly she stressed that any other format can, as she put it, "confuse" the algorithm they use, not just in cases like 5/12/22 (is it 12th May or 5th Dec) but even for example 22 SEPTEMBER 2022 or even 22nd Sep 2022.

That is interesting! Something so minute has a potential to confuse. I admittedly make an effot to format my date as 22nd Sep 2022, and notice a lot of other users stick by what Ancestry defaults. Perhaps my "tidying up" of the dates has hindered me.

Thanks for the reply, and thank you to the others who replied. It looks like I have a bit of work to get on with. 

[LizzieL]

It occurs mostly where some of the other trees from which Ancestry is taking its thrulines have:
a) only got one half of the couple in their tree
2) both people are there but not spelt in exactly the same way as you have so Ancestry is unsure whether they are the same people

Pheno

Totally agree with Pheno

I have have numerous half cousin matches according to Ancestry, which the paper trail shows irrefutably are full cousins, and some the other way round. In a few cases it is because there is a mismatch between what I have in my tree for one of the pair of ancestors and what other people's trees have, so Ancestry thinks John Smith married two different Mary Browns in the same place on the same date, because my Mary Brown is annotated with her exact dob and place, whereas other have taken year of birth from censuses (which may be one or more years out) and birthplace from a transcription which may give a meaningless place name.
In a few cases it is because one of the pair of ancestors has married twice and other trees have connected a child (from which my line descends) to the wrong wife or husband. I have lost count of the erroneous "half cousins" and "full cousins" from my Annetts branch, because so many trees have my 2 x great grandfather as the product of Thomas Annetts and his first wife  - quite an achievement for the lady since she had died three years before he was born!