Disappointed by DNA

[JennyMB]

In the New Year 2020 Newsletter Peter writes of people being disappointed by DNA and suggests that they have not read his Masterclass on the subject. I am one of the disappointed, but I have read the Masterclass and I'm afraid it doesn't address my problem. The problem I have is that so many people who have submitted their DNA for testing seem to have no wish to take the matter any further!

I tested with Ancestry and of course have been told about hundreds of potential cousins. Several of the matches are quite close cousins with whom I am already in contact, but of course the vast majority are more distant people and they are the ones who are most likely to be useful in breaking down my brick walls. So I contact them and ... NOTHING! I have sent messages to dozens of people and very, very few have even acknowledged my contact. Of those who have replied it has usually been a case of something like "I don't know where we match but I will look into it" and that is the last I hear of them.

Incidentally, in all cases except one of a possible match it has been me who has made the first approach. The one person who contacted me first has not responded to my reply.

I really do have to wonder why people are going to the trouble and expense of DNA testing if they have apparently no interest in pursuing the results. I'd be interested to know what has been the experience of others.

 

[peter]

Jenny, I didn't say that people hadn't read the Masterclass, I said they weren't following the advice. Your disappointment is actually predicted in the Masterclass!

You talk in your post about hundred of potential cousins - considering that everyone has tens of thousands of matches I assume you're talking about the matches with whom you share the most DNA (who, it is important to note, are NOT necessarily the closest cousins). Here's what the Masterclass says about this:

"At Ancestry you'll typically have over 20000 matches with genetic cousins, and of those all but about 1% will be with 'distant' cousins, ie where the estimated relationship is 5th cousin or more distant. So you might think that the best strategy might be to focus on the 1% on the basis that if you can't make head or tail of those matches, your chance of resolving the more distant matches is negligible. Wrong, totally wrong - that approach will lead to frustration, disappointment, and disillusionment!"

The strategies I set out in the paragraphs that follow involve searching the trees of your matches. So any cousins who don't have a tree are out of contention, as are almost all of those who have very small trees. Although this might rule out 80% of your matches, 20% of 20,000 plus is still a large number!

Most of those who do have trees have public trees, so even if they don't reply (and most will) you've got a lot to work on. (Even I have a public tree, although it is direct ancestors only, to stop the name collectors harvesting it.)

So go back to the Masterclass and do what it says - you'll be amazed by the difference it makes!

 

[Kane133]

I'd be interested to know what has been the experience of others.

When I first started I too used to be frustrated with the lack of trees and replies. Then I realised that there was little point in being frustrated and changed my attitude and now have no expectation that someone will reply never mind help me solve a 'problem' that is essentially of my own making. So now I can be grateful for any replies, snippets of information in a tree and any help that may come my way.

It is outside the scope of Peter's Masterclass but after waiting a year for a reply from a young lady with a one living person 'tree' who at 198cM was my 'best' DNA match on my mother's English side, I got inventive. Besides knowing from the shared matches that she was on my maternal grandmother's line, the only 'information' I has was the initial for her first name, her surname, her age-range and a close-up profile photo. On FMP there were about 10-15 females with a first name starting with the initial for her first name and having the same surname. Next step was to search on Facebook using the names from FMP looking for a young lady whose photos looked that same as the profile photo on Ancestry. Besides finding someone whose broad features looked the same, with the close-up profile image I was able to positively identify her using dimples, freckles and beauty spots that were exactly the same on Ancestry as her Facebook selfie photos!

By scrolling/trolling through her Facebook photos and friends I was able to work-out who her mother, brother and cousins were. Then it was back to FindMyPast BDM records and Ancestry to start building a tree and exploring possible connections. Took 4 days but when an unusual but familiar (to me) surname came-up as the mother's surname on a birth record, everything fell into place. So now I know who my 2Cx1R is, and that the girl with the unusual surname standing next to my 6 year old mother in a photo taken in 1927, did have a family of her own. With my mother emigrating with her parents in 1928 she lost contact with her first cousin, who is my DNA matches grandmother.

All this was done without making contact and no, I do not intend to try again via Facebook. When her family is interested/ready their part of the tree is waiting in Ancestry and of-course I will reply if they send a message.

 

[Pauline]

I have tried something similar with some of my very close matches, though for ages I didn't have any shared matches with my (initially) closest nor do they have a profile photo.

Fast forward nearly 3 years and although this mystery person is no longer my closest match, I am still none the wiser as to how we are related.

 

[peter]

I suspect we all have had similar experiences - sometimes it's difficult to rationalise why our cousins tested at all. Perhaps it was just a whim? Or perhaps after reading scare stories in the tabloids whatever initial enthusiasm they had has waned?

I try to focus on the reason why I tested in the first place, which was to knock down 'brick walls' - finding cousins is an essential part of the process but it's only a stepping stone. Sometimes our cousins will want to join on us on the journey, but most times they won't, and how closely they're related to us probably won't be a key factor in their decision.

 

[JennyMB]

I apologise for sloppy wording in my previous post. I should have said "following", not "reading."

I can and do, of course, follow the advice in the Master Class and it is very useful. I think perhaps I need to take a more relaxed attitude to the lack of contact with DNA matches! I have been lucky enough to become part of a lovely little group (mainly 4th cousins) which has come together over the years to share our findings about one branch of our family but I have to accept that this is exceptional, rather than the norm.

 

[Pauline]

Or perhaps after reading scare stories in the tabloids whatever initial enthusiasm they had has waned?

I have wondered if some of the DNA scare stories (not only in the tabloids) have significantly reduced enthusiasm for testing altogether. Certainly in recent weeks and months, the number of matches I have - both close and distant - has been increasing at a much slower rate. Previously I was getting new 'close' cousin matches almost every day, whereas now it seems to be only 1 or 2 a month.

 

[Pauline]

Previously I was getting new 'close' cousin matches almost every day

Just to clarify that, I haven't been getting new close matches nearly every day since the day I first got my results (I'd have over a thousand close matches by now if I had), but the rate of increase had got to that level before the recent drop-off to only one or two a month.

 

[canadianbeth]

I tested 15 months ago in hopes of finding out who my paternal grandfather was. I am no closer to learning that, but I have learned that just about everything I was told about my paternal grandmother was incorrect. I have learned a lot about her and my ancestors on that side, all the way to 5x great-parents. A lot of unknown matches that probably link to that grandfather, but no one to whom I have written has bothered to reply. I had two new matches at Ancestry yesterday but both were for my maternal ancestors.

 

[peter]

I tested 15 months ago in hopes of finding out who my paternal grandfather was. I am no closer to learning that, but I have learned that just about everything I was told about my paternal grandmother was incorrect. I have learned a lot about her and my ancestors on that side, all the way to 5x great-parents. A lot of unknown matches that probably link to that grandfather, but no one to whom I have written has bothered to reply. I had two new matches at Ancestry yesterday but both were for my maternal ancestors.

Roughly a quarter of your matches are linked through each grandparent. Even if only one person in 10 has a public tree of any substance, that's still over 500 matches per grandparent that you can analyse without needing to hear from the other person (though in my experience the people who do have trees are most likely to respond).

So there must be a wealth of evidence that you can draw on (assuming you have an Ancestry subscription) - the challenge for you is to find an effective and efficient way of analysing the data in the trees of your matches.

 

[Pauline]

Even if only one person in 10 has a public tree of any substance, that's still over 500 matches per grandparent that you can analyse

I agree in principle, but your figure of 500 matches does pre-suppose that we can identify which of those 2000 matches with useful public trees belong to which grandparent. I know that takes us back to gathering more documented cousins (and that is a very important part of it) but it's much harder identifying the 'distant' cousins matches (which most of those 500 per grandparent will be) since the shared match feature only finds the one who are 'close' cousin matches.

When it comes to identifying which grandparent each of my matches relates to, I assign them one of two colours - the stronger colour indicates that the match is, one way or another, a documented cousin, the lighter colour indicates they are a shared match of a documented cousin but their exact connection is unknown. The light coloured matches are the ones who are perhaps most likely to provide new clues towards breaking down brick walls, although their trees obviously need to be analysed in conjunction with those of the documented cousins.

The number of coloured matches, both strong and light, that I have is considerably fewer that 500 per grandparent, and includes quite a few who don't have trees. There is also something of an imbalance as I can identify far more matches on my mother's side than on my father's, but I expect is not unusual.

 

[peter]

That is indeed the challenge - all that evidence, yet no obvious path to the solution.

So far you've only talked about looking for shared matches, which is a strategy that generally only works for close matches. But what about looking for other things that your matches have in common, such as surnames and birthplaces? Make the most of the evidence that you have; come up with different ways to analyse it in the hope of spotting patterns or similarities.

 

[Kane133]

But what about looking for other things that your matches have in common, such as surnames and birthplaces?

Personally, I think genealogy is a bit like real estate ... location, location, location. Knowing a location from where ancestors came from and doing a "Birth location in matches' trees" search of your DNA matches on Ancestry can be productive in finding cousins who share less than 20cM. When I search using Sigglesthorne, Yorkshire I get 9 matches below 20cM of which two were found to be cousins. One a 4Cx1R sharing 9cM and the other a 5C sharing 6cM. Unless a surname is unusual, surnames are a secondary consideration IMO. Then again a surname plus a parish or county location search is good for focusing on which matches to explore.

 

[peter]

Personally, I think genealogy is a bit like real estate ... location, location, location. Knowing a location from where ancestors came from and doing a "Birth location in matches' trees" search of your DNA matches on Ancestry can be productive in finding cousins who share less than 20cM. When I search using Sigglesthorne, Yorkshire I get 9 matches below 20cM of which two were found to be cousins. One a 4Cx1R sharing 9cM and the other a 5C sharing 6cM. Unless a surname is unusual, surnames are a secondary consideration IMO. Then again a surname plus a parish or county location search is good for focusing on which matches to explore.

Absolutely right, but with places you can have the same problem as with surnames - a small village is the equivalent of a rare surname, but looking for people with ancestors from London is like looking for Smiths. It also helps if your ancestors were in the same locality for several generations.

 

[Pauline]

I tend to adopt a multi-pronged approach, though I try to do so in a methodical rather than chaotic way. I have also adapted my methods as the available features at Ancestry have changed.

So I do periodic searching of all my matches (and those of my sister) by surname and by location, though I tend to restrict my use of location searches to smaller places - searching for all matches who have folk born in Islington, for example, brings up far too many matches to be useful.

I also regularly look for new 'Common ancestor' matches, of which I seem to have a lot, and thoroughly scrutinise the supposed connection, especially where a common ancestor match is based on extrapolating the match's tree using other Ancestry trees. So far I have found that most of my common ancestor matches can be confirmed with a bit of research, while a few can be definitely discounted. Others seem, from shared matches etc, to be connected in the same general area that is suggested, but clearly not via the route suggested, and need further investigation.

I also take a good look at all new 'close cousin' matches, checking for unlinked trees if there is no linked one. I look at shared matches with the new match, check to see if my sister is also match and if she gets different shared matches which shed a light on anything etc etc.

I now also make quite a lot of use of the newish facility to set multiple filters and, for example, can then use the search to identify common surnames and locations within a particular cluster of matches, which might then help me in identifying in which part of my tree I might connect to them.

Anyway, this has got rather long, so I'd better stop here, but this isn't a complete list - more examples of the sort of methods I use.

 

[canadianbeth]

My grandmother went from the workhouse to Woodford where she allegedly met my grandfather. I put Woodford into the search some time ago and got dozen upon dozens of names. One possible 3rd cousin, C.W., the same one who comes up every time right at the top - 198cM, 9 segments. I have written to him twice without reply. There are 36 in the 4th-6th list, many already matched to Joyce, which was my grandmother's name. C.W. has 37 names in his tree; I have looked at the ones with the possible dates and they are all linked to Norfolk. But he is not linked through shared matches with my first cousin once removed; she is on my mother's side, so he has to be paternal. I just cannot figure out how. All paternal matches would be half-relatives at best.

 

[peter]

All paternal matches would be half-relatives at best.

What do you mean by that?

 

[canadianbeth]

Since my Dad was illegitimate, any other children that his father had would be half-siblings, would they not?

 

[Kane133]

Since my Dad was illegitimate, any other children that his father had would be half-siblings, would they not?

Yes they would 'assuming' a different mother. But life is usually stranger than fiction and I have seen TV programs where a person has discovered full siblings having been adopted out as a baby because of some sad reason or difficult situation the parents found themselves in. So a full-sibling connection for your father cannot necessarily be discounted.

And as for all paternal matches being half-relatives definitely not. A 2C will be a 2C for every situation where the grandparents were full siblings and the same for 3Cs where the great grandparents are full siblings and so on for 4Cs and beyond.

 

[Helen7]

I also regularly look for new 'Common ancestor' matches, of which I seem to have a lot, and thoroughly scrutinise the supposed connection, especially where a common ancestor match is based on extrapolating the match's tree using other Ancestry trees. So far I have found that most of my common ancestor matches can be confirmed with a bit of research, while a few can be definitely discounted. Others seem, from shared matches etc, to be connected in the same general area that is suggested, but clearly not via the route suggested, and need further investigation.

I too have found 'Common ancestor' matches useful, though in the past, I've been guilty of dismissing some of these matches too readily, where the suggested connection is extrapolated from other trees and appears wrong. However, following your suggestion to scrutinise these and look for other routes in the same general area, I recently looked at some of these again, and in one case I was able to discover how I was related to one cousin, having previously dismissed it as a mismatch.

In this case, I didn't recognise any of the surnames in my DNA match's tree, and Ancestry suggested that he was descended from Hannah M, supposedly a younger sister of my ancestor Mary M (Hannah being 20 years Mary's junior). The evidence for this was just one other tree which had Mary and Hannah as siblings (with other errors in the tree which didn't fill me with confidence). Now I was pretty certain that Mary didn't have a sister Hannah, so I dismissed this link. But looking up the birth registration of this Hannah I found the mother's maiden name was blank, and the baptism record on FMP told me Hannah was illegitimate and her mother's name was Martha. Now my ancestor Mary did have a sister Martha, just 2 years her junior. So Hannah was Mary's niece, not her sister. I then found Hannah married and had a daughter who was the great-grandmother of my DNA match, so it all fitted nicely into place. And it enabled me to add another (large) family of cousins in the 1881 census into my 'My Ancestors' page at Lost Cousins!