When DNA Leads You Astray

I’m currently going through what I refer to as “the great purge.”

This occurs when you can’t stand the accumulated piles and boxes of “stuff” and the file drawers are full, so you set about throwing away and giving away. (Yes, I know you just cringed. Me too.)

The great news is that I’ve run across so much old (as in decades old) genealogy from when I first began this journey. I used to make lists of questions and a research “to do” list. I was much more organized then, but there were also fewer “squirrel moments” available online to distract me with “look here, no, over here, no, wait….”

Most of those questions on my old genealogy research lists have (thankfully) since been answered, slowly, one tiny piece of evidence at a time. Believe me, that feeling is very rewarding and while on a daily basis we may not think we’re making much progress; in the big picture – we’re slaying that dragon!

However, genealogy is also fraught with landmines. If I had NOT found the documentation before the days of DNA testing, I could easily have been led astray.

“What?”, you ask, but “DNA doesn’t lie.” No, it doesn’t, but it will sure let you kid yourself about some things.

DNA is a joker and has no problem allowing you to fool yourself and by virtue of that, others as well.

Joke’s On Me

Decades ago, Aunt Margaret told me that her grandmother’s mother was “a Rosenbalm from up on the Lee County (VA) border.”

Now, at that time, I had absolutely NO reason to doubt what she said. After all, it’s her grandmother, Margaret Claxton/Clarkson who she knew personally, who didn’t pass away until my aunt was in her teens. Plenty close enough to know who Margaret Claxton’s mother was. Right?

DNA Astray Rosenbalm

Erroneous pedigree chart. Rebecca Rosenbalm is NOT the mother of Elizabeth Claxton/Clarkson.

I filled Rebecca Rosenbalm’s name into the appropriate space on my pedigree chart, was happy and smugly smiling like a Cheshire cat, right up until I accidentally discovered that the information was just plain wrong.

Uh oh….

Time Rolls On

As records became increasingly available, both in transcribed fashion and online, Hancock County, TN death certificates eventually could be obtained, one way or another. Being a dutiful genealogist, I collected all relevant documents for my ancestors, contentedly filing them in the “well that’s done” category – that is right up until Margaret Clarkson Bolton’s death certificate stopped me dead in my tracks.

margaret clarkson bolton death


Margaret’s mother wasn’t listed as Rebecca Rosenbalm, nor Rebecca anyone. She was listed as Betsy Speaks. Or was it Spears? In our family, Betsy is short for Elizabeth.

Who the heck was Elizabeth Speaks, or Spears. This was one fine monkey wrench!

A trip to Hancock County, Tennessee was in order.

I dug through dusty deed and court records, sifted through the archives in basements and the old jail building where I just KNEW my ancestors had inhabited cells at one time or another.

Yes, my ancestor’s records really were in jail!

Records revealed that the woman in question was Elizabeth Speaks, not Spears, although the Spears family did live in the area and had “married in” to many local families. Nothing is ever simple and our ancestors do have a perverse sense of humor.

Elizabeth Speak(s) was the daughter of Charles Speak, and the Speak family lived a few miles across the border into Lee County, Virginia. This high mountain land borders two states and three counties, so records are scattered among them – not to mention two fires in the Hancock County courthouse make research challenging.


I asked my Aunt Margaret who was still living at the time about this apparent discrepancy and she told me that the Rosenbalms “up in Rose Hill, Virginia” told her that her grandmother, Margaret Claxton/Clarkson was kin to them, so Margaret had assumed (there’s that word again) that Margaret Claxton’s mother was their Rebecca Rosenbalm.


The Kernel of Truth

Like so many family stories, there is a kernel of truth, surrounded by a multitude errors. Distilling the grain of truth is the challenge of course.

Margaret Claxton’s mother was Elizabeth (Betsy) Speak and her father was Charles Speak. Charles Speak’s sister, Rebecca married William Henderson Rosenbalm in 1854, had 4 children and died in February 1859. So there indeed was a woman named Rebecca (Speaks) Rosenbalm who had died young and wasn’t well known.

Rebecca’s sister Frances “Fanny” Speak also married that same William Henderson Rosenbalm in November 1859, a few months after Rebecca had died. Fannie also had 4 children, one of which was also named Rebecca Rosenbalm. Do you see a trend here?

So, indeed there were 7 living Rosenbalm children who were first cousins to Elizabeth Speak who married Samuel Claxton and lived a dozen miles away, over the mountains and across the Powell River. Now a dozen miles might not sound like much today, but in the mountains during horse and wagon days – 10 miles wasn’t trivial and required a multi-day commitment for a visit. In other words, the next generation of the family knew of their cousins but didn’t know them well.

The following generation included my Aunt Margaret who was told by those cousins that she was related to them through the Rosenbalm family. While, that was true for the Rosenbalm cousins, it was not true for Aunt Margaret who was related to the Rosenbalms through their common Speak ancestor.

Here’s what the family tree really looks like, only showing the lines under discussion.

DNA astray correct pedigree

You can see why Aunt Margaret might not know specifics. She was actually several generations removed from the common ancestor. She knew THAT they were related, but not HOW they were related and there were several Rebecca’s in several branches of the family.

Why Does This Matter?

You’ve probably guessed by now that someplace in here, there’s a moral to this story, so here it is!

You may have already surmised that I have autosomal DNA matches to cousins through the Rosenbalm/Speaks line.

DNA astray pedigree match

This is one example, but there are more, some being double cousins meaning two of Nicholas Speak’s 11 children’s descendants have intermarried. Life is a lot more complex in those hills and hollers than people think – and unraveling the relationships, both paper and genetic (which are sometimes two different things) is challenging.

DNA astray chromosome 10.png

I match this fourth cousin once removed (4C1R) on a healthy 18 cM segment on chromosome 10.

Wrong Conclusions

Now, think back to where I was originally in my research. I knew that Margaret Claxton/Clarkson was my aunt’s grandmother. I knew nothing at all about the Speak family and had never heard that surname.

Had I ONLY been looking to confirm the Rosenbalm connection, I certainly would have confirmed that I’m related to the Rosenbalm family descendants with this match. Except the conclusion that I descend from a Rosenbalm ancestor would have been WRONG. What we share are the Speak ancestors.

So really, the DNA didn’t lie, but unless I dissected what the DNA match was really telling me carefully and methodically with NO PRECONCEIVED NOTIONS, I would have “confirmed” erroneous information. Or, at least I would have thought that I confirmed it.

I would actually have been doing something worse meaning convincing myself of “facts” that weren’t accurate, which means I would have then been spreading around those cancerous bad trees. Guaranteed, I do NOT want to be that person.


I can tell you here and now that I have found several matches that were foolers because I share multiple ancestors with a person that I match, even if those multiple ancestors aren’t known to either or both of us. Every single DNA segment has its own unique history. I match one individual on two segments, one segment through my mom and one segment through my dad. Fortunately, we’ve identified both ancestors now, but imaging my initial surprise and confusion, especially given that my parents don’t share any common ancestors, communities or locations.

We have to evaluate all of the evidence to confirm that the conclusion being drawn in accurate.

DNA astray painting

One of the sanity checks I use, in addition to triangulation, is to paint my matches with known ancestors on my chromosomes using DNAPainter. Here’s the match to my cousin, and it overlaps with other people who share the same ancestor couple. Several matches are obscured behind the black box. If I discover someone that I supposedly match from a different ancestor couple sharing this segment of my father’s DNA, that’s a red neon flashing sign that something is wrong and I need to figure out what and why.

Ignoring this problem and hoping it will go away doesn’t work. I’ve tried😊

Three possible things can be wrong:

  1. The segment is identical by chance, not by descent. With a segment of 18 cM, that’s extremely unlikely. Triangulation with other people on this same segment on the same parent’s side should eliminate most false matches over 7cM. The larger the match, the more likely it is NOT identical by chance, meaning that it IS identical by descent or genealogically relevant.
  2. The segment is accurately matched but the genealogy is confused – such as my Rosenbalm example. This can happen with multiple ancestors, or descent from the same family but through an unknown connection. Looking for other connections to this family and sorting through matches’ trees often provides hints that resolve this situation. In my case, I might have noticed that I matched other people who descended from Nicholas Speak, which would not have been the case had I descended through the Rosenbalm family.
  3. The third scenarios is that the genealogy is plain flat out wrong. Yea, I know this one hurts. Get the saw ready.

The Devil in the Details

Always evaluate your matches in light of what you don’t know, not in order to confirm what you think you know. Play the devil’s advocate – all the time. After all, the devil really is in the details.



I receive a small contribution when you click on some (but not all) of the links to vendors in my articles. This does NOT increase the price you pay but helps me to keep the lights on and this informational blog free for everyone. Please click on the links in the articles or to the vendors below if you are purchasing products or DNA testing.

Thank you so much.

DNA Purchases and Free Transfers

Genealogy Services

Genealogy Research

22 thoughts on “When DNA Leads You Astray

  1. I have a whole bunch of Rosenbalm’s from Washington Co. VA. I have a copy of Rebecca Fox Rosenbalm’s Bible. My Grandmother was Sarah Frances Rosenbalm. She died in childbirth and her little girl died a month later named after her mother.

  2. This should be required reading for everyone who has an online tree, especially the comments about cancerous bad trees.
    Unfortunately, the testing companies are promoting copying unproven ancestors into trees!
    When MH tells me I can add 40 people to my tree by accepting a Smart Match,, I cringe! NO WAY do I want anything copied into my tree from Smart Matches.
    Ancestry is doing something just as bad now. They now show possible parents and put them in your tree! Without your permission! In pedigree view, you can’t remove them. I was nearly in tears trying to get rid of these! Finally, I went to profile view, and still having fits trying to change their version. But when I clicked on Facts it let me remove them. Took me ages to do this because every brick wall I look at has their bad version of possible parents added!

    How can we get these companies to quit being so pushy about adding people to our trees!!!

    I want truth, not a multitude of names!

    • I haven’t checked but they will probably put these incorrect possible (read impossible or unlikely) parents of my ancestors back in my tree next time I look at it!
      I wish there was a way to turn off their editing of my tree!

        • I had someone request permission at MyHeritage, then add people/records to my tree, before I had the chance to accept or deny their request. I then had to deny the request and remove their edits. The way MyHeritage promote these requests make you appear selfish for wanting full control of your own tree.

      • Ancestry’s “editing” of your tree is just a new way of displaying the same old tree hints. You have to turn it off separately in Site Preferences; it’s called “Potential Ancestor Hints”.

    • I haven’t looked at my Ancestry tree in a few days, but I’m offered parents for ancestors, but they have not been put there without me agreeing in the past. I wonder if that has changed. I surely hope not:(

      • I just checked my tree and am no longer having potential parents added to my tree. I have not changed any settings before or after. The potential parents were shown in green and represented what a popular version of the parent is from a compilation of trees. Maybe this was s short term experiment. Just glad it is no longer happening. It was even in my pedigree view which is what I use for screenshots to send family! Happily, I am no longer seeing potential parents added to my tree in bright green or otherwise.

        • Actually, I like the suggestions because I get to review that persons information. Mine were not automatically added though.

  3. Fascinating. I wish I had your indepth understanding of DNA and genealogy. Clearly, you have done your homework.

    Just five days ago, I discovered who my biological father was. This has been a mystery to me for most of my life. There were a lot of false rabbit trails in my search, too!

  4. A few years ago I was looking for my great grandfather’s death certificate. His name was John Thomas Warren, often known as Jack. I couldn’t find anything with that name. Fortunately, I knew the date and place of death and was able to search online records using that information. The certificate I found had the name Thomas Jackson Warren. That information had been given by one of his son’s! You would think a son would know his father’s name, but other sources show that he did not. Lesson learned

    • My great-grandmother’s birth surname was wrong too. But the person giving the information was not in the same state and who knows who gave it.

  5. Such a good article for me! I’ll admit I look at clues from other online trees. Especially when I’m trying to figure out how a match is kin to me. And I know not to blindly accept the first possible conclusion (in big part thanks to your earlier articles). But my version of your Aunt Margaret’s family story of Aunt Margaret’s Rosenbalm being her grandmother’s mother is my own paternal family stories that my paternal grandfather’s birth father was named Jones. And also that the man was a businessman from town, with a wife and family. But maybe it’s just an illusion I’ve clung to. On some level I’ve known it from the beginning. Nevertheless, I keep spending hours and hours trying to bolster up my illusion with “evidence.” I’ve knocked this paternal brick wall down so many times, only to have to build it back up, again and again. How do I know that the “truths” I’m finding are really true? Yikes! :O

  6. My Uncle Curtis is named as the informant on my paternal grandmother’s death certiificate. He got her mother’s name wrong. My grandmother’s mother was Margaret Campbell. They had written Martha (her sisterr’s name!) but then looks like they marked it out and didn’t know what name to change it to. My grandmother also had a sister Martha whom my Uncle would have known before she moved while he was a child. His grandmother died before he was born. So he must have got the aunt and the grandmother confused. Since his mother had just died and he was grieving maybe that contributed to his confusion.

    But obviously records are not always correct even when information is given by close relatives. He signed as informant but anyone in the room may have answered the questions.

    Incidentally, the big thick index of Arkansas death records my library has cannot be trusted. The transcriber must have got off on the wrong line sometimes? Or the typist? They have this grandmother as B and M a black male. Her death certificate has her as white female which is correct. The index is wrong.

    • My mother’s obituary had to be republished twice because someplace between us and the publication, it was wrong.

  7. Awesome article! One needs to be always careful when drawing conclusions from DNA and from family hearsay.

    A great grandson of my 4th Great Grandmother Rebecca Hall said that her parents were Samuel and Rebecca Hall. It did make sense circumstantially. After all, you have the name Rebecca and the family did live close to where Rebecca Hall’s husband Calvary Chancey lived in the same county. DNA has shown otherwise. Rebecca Hall’s parents were really Elijah and Permelia (Schoolcraft) Hall. Circumstantial evidence supports it. Permelia (Schoolcraft) Hall lived next door to Calvary Chancey’s brother James Chancey in the same county where Calvary and Rebecca were married. The 1840 census with Permeiia Hall shows an unaccounted female of the right age to be my Rebecca Hall. This illustrates why one can never be too careful.

  8. Plenty of people fail to consider #3 which I think happens more often than we would hope, particularly on distant matches. Especially if the match had been locked away initially as coming from another line and may have crept into the #2 category. We really need to be open to the possibility that sawing needs to happen!!

  9. I’ve been trying for ages to track down the biological grandfather of a family member. A moderately close DNA match popped up a couple of years ago, but I didn’t pay a lot of attention because they also matched a 1st cousin on the maternal side, and so clearly it was a maternal match and thus not immediately interesting … Until I started allocating matches to different grandparents on DNAPainter and it became clear that this new person only matched other people suspected of being on the unknown grandfather’s line. And also, lo and behold, the two 1st cousins are related on 2 separate lines on opposite sides of their families. Looks like this new match is in fact pretty closely related to the mystery biological grandfather. I love DNAPainter.

    I know I fell into the trap of making unwarranted assumptions, but it just seemed such a very reasonable assumption to make at the time – I’m not sure whether the DNA led me astray, or set me straight.

    Now if I could only get that DNA match to communicate. Sigh.

Leave a Reply