Sunday, December 31, 2017

A Probate by any Other Name

Ancestry has apparently been selling their DNA tests beyond North America, as I've recently seen DNA matches who live in Sweden, Australia and the British Isles. This is a wonderful development for those of us with fairly recent ancestors who lived in other countries.

While messaging with a DNA match in Sweden, a gap in my tree jumped out at me. That gap reminded me of a series of posts where I described how a cousin paid for research in Sweden that was sloppily (or fraudulently) done. Four years ago, I recommended:
  • Never ignore evidence
  • Consider unusual sources for unusual problems
  • Even professionals get it wrong sometimes -- by accident or on purpose
  • When you pay for research, make sure to get lists and/or copies of sources

It was important to me to document the truth for cousins now and in the future who might see a copy of the bad research. This latest review leads me to a new recommendation for my cousins:
  • Throw away that research. It is good only for starting fires and lining bird cages.

One of the lines in this allied family ended at a woman named Brita Larsdotter who was born in 1772. Why did I not have a birth date, a birth parish or parents? The bad researcher had extended her line several generations. What was I missing?

Brita had married Olof Andersson, borne children and died at the young age of 39, all in the parish of Östra Vingåker in Södermanland. The parish records were a bit sparse and there was no indication that she had migrated in from a different parish. Hopefully she had been born in the parish. The first step was to research all girls named Brita who was born in that parish in 1772. If I couldn't identify her, I could search a couple of nearby parishes, but it would certainly not be practical to search the other 2500 parishes!

There were only two girls in the birth records who were named Brita and had a father named Lars. One had a middle name of Cathrina. The other had no middle name recorded. My Brita never appeared in any other record with a middle name. So I followed the other one -- the daughter of Lars Larsson.

Lars and his wife had several children. There was not much common geography between my Brita and the Larsson family. Nor did any of the Larsson children serve as godparents or witnesses at the christening of my Brita's children. When the Larsson family was visited by the minister in about 1790, Brita's name was crossed out with no indication why. Had she moved? Married? Died? If she was my Brita, the timing would have been right for her arrival at the place she met her husband.

I followed Lars Larsson until his death. The next step was to find Lars' probate, the bouppteckning, or estate inventory. It would list all his living children or other heirs. These records are available on ArkivDigital at this time. Unfortunately, finding the file within the bound books can take some time, as there are few indexes and the order is rather random. When I finally found the file, Brita was not listed. Since my Brita was alive at the time, this was the wrong family. The other Brita had indeed died.

It was time to start over and research Brita Cathrina. Her father, Lars Ekelman, was a discharged soldier who was 72 years old at the time of her birth. Her mother was a shocking 27. Lars died in 1776, a time for which few probate records survive. Brita's mother, Anna Ericsdotter, remarried in 1777, and had several children with her second husband. Those children did indeed serve as godparents or witnesses at the christening of my Brita's children. This was definitely the right Brita. Could the relationship be proven?

Returning to the bouppteckning records provided that proof. Brita's step-father's probate wasn't found, but Brita and her mother both had files. Brita died in 1812 and her file gave the name Brita Cajsa, a nickname for Cathrina. Her mother, Anna, died in 1815. Anna's probate mentioned four children whose rights would be protected by Olof Andersson, Brita's husband. That was concrete proof that Brita was the daughter of Anna Ericsdotter and Lars Ekelman.

Having found Brita's parents, they became the end-of-line ancestors, with the same challenges in the sparse records. However, the bouppteckning files had a very important clue. Both files mentioned a man named Anders Ericsson of Gribecken in the parish of Stora Malm.

Brita's mother being Anna Ericsdotter, it appeared that Anders was Anna's brother and Brita's uncle. Indeed that was the case. Following Anders to his birth parish of Björkvik was fruitful, resulting in finding yet another generation: Eric Ericsson and Kierstin Ericsdotter. The time and money spent to research the bouppteckning records were well worth it!

The paid researcher had followed and chosen Brita, the daughter of Lars Larsson. I doubt she considered Brita Cathrina, as she should have noticed the christening records supporting this choice. She obviously did not look at the bouppteckning files.

Be careful when hiring a researcher. Get the list of sources and review the work that was done. Don't ever accept a sloppy job.

Tuesday, December 26, 2017

Hidden DNA at Ancestry

A couple of months ago Ancestry made another change to their policies for managing DNA test results. It is a mix of both good and bad, depending on your perspective. You can choose to hide your test from your DNA matches. If you make that choice, you also can’t see your matches. If you have extreme privacy concerns, this change could be a blessing. However, you could also skip the test or choose to delete the test from Ancestry. So this change feels a bit ridiculous.

It seems a lot of people merely want their ethnicity results, but not matches. If they choose to hide their results, they will disappear from our match lists. That would help reduce the wasteland of useless results that we researchers must wade through.

However, sometimes even a match without a tree can help with a breakthrough. I’ve assisted a couple of adoptees who relate to tests I administer, with skeleton trees that are private. I recently received a nice thank you from such a researcher who identified her birth grandfather after I shared just a little information.

I also had a personal victory. Over a year ago Roberta asked on the DNAExplained blog about converting a New Ancestor Discovery into an actual ancestor. I was able to do that recently with the help of a treeless match. It did take a lot of lucky breaks.

I administer a test for a college student that I’ll call Missy. Her parents divorced when she was young and her father poisoned the relationship by casting doubt on Missy’s parentage. His French-Canadian Catholic family was not pleased with Missy's mother. Conversely, her mother’s family, including me, was incensed at his behavior.

I tested Missy simply as part of testing many family members. Ironically, her DNA is more heavily French-Canadian than her older sibling’s. She has 12 new ancestor discoveries — more than anyone else among my 11 family tests — and most of those discoveries are French-Canadian.

For several years I have been blocked at Missy’s living paternal grandmother. Let’s call her Grandma Case (as in case study). I knew her birth date and her maiden and married names, but could not identify her parents. I also needed to do this work without reaching out to other Case researchers.

One of Missy’s New Ancestor Discoveries has the Case surname and was born in the Montreal area in about 1799. This looked like a good hint, but I was not interested in spending the time to do a descendants study.

Missy has a close DNA match with no tree, but with 329 centimorgans shared across 14 DNA segments. Lucky break number one.

The man we’ll call Eddy Case used his real name when he registered his DNA test. Lucky break number two. I waited for a year to see if Eddy would provide a tree, but it didn’t appear. One day, when reviewing the new ancestor discoveries, I decided to see if it was possible to break through the brick wall with just the information I had.

A Google search on Eddy’s name turned up only four matches in the entire US. Lucky break number three. Too many matches would have put a quick stop to the research.

One of the four men lived in the right area of Michigan. He had been interviewed in a newspaper article, giving his age. Lucky break number four.

Eddy was born before the 1940 census. Lucky break number five.

Starting from the 1940 census, I quickly ran up his tree and arrived at the new ancestor discovery couple. All the work to this point would have been done for me if Eddy had posted a tree. So this part of the journey is not a show-stopper when working with a promising NAND.

The next step was to determine possible relationships between Grandma Case and Eddy Case. One of my considerations needed to be the fact that the French-Canadian community is endogamous, which can make the match stronger than it might otherwise be.

Blaine Bettinger at The Genetic Genealogist has researched how DNA match strength corresponds to relationships. He has posted a PDF with several handy charts at his website and he updates it periodically. He has clustered relationships by strength and provides tips about how to determine probable relationships. He also tracks endogamy in his collected information.

From the charts, I could see that the match between Eddy and Missy falls into clusters 4 or 5. Grandma Case could be a younger half-sister to Eddy, a first cousin, a niece, or some other relative within two generations.


The next step was to document Eddy’s siblings and cousins. Missy had no matches to anyone in Eddy’s mothers family,  but had a number of matches related via Eddy’s grandmother’s family. I decided to ignore the possibility that Eddy was a half-sibling to Grandma Case and focus instead on Eddy's first cousins.

Tracing Eddy’s aunts and uncles, I ran into a number of roadblocks but was able to eliminate most branches through obituaries or early deaths. Finally, when ready to give up with three open branches, I found an obituary on an obscure website that listed Grandma Case as a daughter of the deceased. Lucky break number six.

Grandma Case was indeed a younger first cousin to Eddy, making Missy a 1C2R to Eddy, which is a cluster 5 match.

Would I have eventually made the find without Eddy’s DNA match? The close match with a clear name and the wonderful relationship chart led me to the right branch. If Eddy chooses now to hide his DNA tests from his matches, someone else may not have the hint they need to make a discovery.

The entire project was completed in one weekend. That is fast in genealogy time! So thank you, Eddy Case, for your contribution.

Saturday, December 23, 2017

Dear Santa -- My Ancestry Christmas List

Dear Ancestry

Dear Santa,

I really need some things from Ancestry. Would you help them out, please.

Give back the number of pages of DNA matches that I have. I'm not a fan of slot machines! It's just not fair to the newbies! I know I have a bazillion matches, but they have no clue.  Show Ancestry that a heart is as important as a wallet. I know, Santa, that's a tough one. Maybe their CEO needs a visit from a Christmas ghost.

Make the data content, data indexes and search parameters behave consistently. 5 record hits should not become zero or 5,000. Help Ancestry realize they are not the only game in town. I haven't seen those bugs at their competitors' sites.

Add a source quality feature and get real with it. In fact, stop counting a source of someone else's tree as a valid source. My Chancery Court case file is a far better source than their error-filled census, but Ancestry doesn't honor that official file because it isn't in their record sets. Help Ancestry see that they are not the only source of data -- merely a tool. Sourcing from John Doe's tree isn't a quality source and Ancestry needs to treat it with no more respect than it deserves.

Return the DNA matches that were lost to Timber. I want back my matches from the Church of the Brethren endogamous group. There were several good matches that are now lost.

A three-way privacy option on trees would be fabulous. It would be private, public, or shared only with DNA matches.

There are so many other ways that  you could help Ancestry improve, Santa. But here's the last and biggest wish -- a chromosome browser. Today I found a DNA match to three of my family tests that makes no sense. Those three areas don't intersect that I know of. It may forever remain a mystery. You know the answer, Santa, because you know us all. But we humans really need a chromosome browser.

Thank you, Santa. I'm counting on you!

Monday, December 4, 2017

Adoptions and Roots Magic Genealogy Reports

An orphan named Jacob sent me into the weeds this weekend while accepting a challenge on genealogy errors. Read a bit of Jacob's story as told to me by his son and think about where you would place him in the family tree.

My father was born about 1887 or 1888. He was orphaned at a very early age and existed as a waif on the streets of Dallas, Texas. Sometime in the early 1890's a rancher by the name of Lucas took him off the streets back to his ranch to raise.  It would have been Andrew Jackson Lucas ... Whether he was officially adopted or just brought in as a foster child I don't know. My father didn't know his name so he took the name Theodore Lucas.

He lived with the Lucases until their daughter [Nora] married Frank Allee [in 1896] and he then went to live at the Allee place. He stayed with them until in his mid to late teens, when he left... He also said he was called Jakey. At one time while he was living with the Allees he considered changing his name to Jake Allee.
- Fayette B. Lucas, 2000

Was Jacob a foster child of Andrew Jackson Lucas and Mary Elizabeth Houston or was he a foster child of Nora Bell Lucas and Abraham Frank Allee? There is no right answer. The choice I made was flagged as an error in Roots Magic when I took Randy Seaver's challenge for Saturday night genealogy fun: checking for errors by running a Genealogy Database Problem Report.

To respond to Randy's challenge, I imported my Family Tree Maker database into Roots Magic, as I highly value the Roots Magic reporting capabilities. I started with the same values he used, but changed them a bit as I had a number of girls who married as early as 13. Jacob was flagged as being born when his mother (Nora) was 12 and his father (Frank) was 14. I understand that this could be an error, even though he was a foster child. The idea is to look at the dates to be sure there isn't a typo or research error.

After marking some items as not an error, including Jacob's birth, there were 18 errors remaining to be reviewed for 6,299 people and 10,688 events. That's an error rate of 0.29% on people and 0.17% on events.

However, checking on Jacob's error did show me a gap in Roots Magic. In the two versions of Family Tree Maker that I use, there are options to create a report of people who are not biological children of one or both parents. Roots Magic does not seem to have any way to create such a report. I also could not find a way to make these relationship indicators print on a family group sheet in Roots Magic, while they can be selected in FTM. I spent hours looking for how to report on relationships, as my intent is to entirely switch to Roots Magic and away from the now-defunct FTM.

That missing information is a problem, as adoptions are scattered throughout my tree and step-parents abound. With the new focus on DNA, it's an important piece of information. The primary way to see the relationship is on the specific family page. The relationship for Jacob shows as foster, while the other children have a relationship of birth, meaning biological children.

When the child has a different relationship with each of the two parents, there are two words in the relationship column, one for father and one for mother.

The reporting features of Roots Magic still make it my favorite reporting program. There is a new version in the works and I have high hopes it will include options to include relationships on reports and forms.