AK'S Genealogy Research: Comparing the new 23andMe Ethnicity/ A New 2nd Cousin Match

In October 2020 new ethnicity estimates were provided to people who tested at 23andMe with the V5 chip. At the time some thought you could only get the update if you tested with the newer V5 chip. My family tested with the V4 chip and we didn't get the update when it first came out. Checking my ethnicity results at 23andMe recently I discovered updated results.

A previous update to my results, the one before the current estimates, was disappointing. 23andMe appeared to provide the most accurate estimates for my family before that update. The new update seems to correct some of the issues I had with the previous estimates. I believe the estimates around 2018 were the best.

Comparing my estimates 2015 to the present 2021

My European ethnicity estimate has been pretty consistent in the 90% range over the years. I was surprised when it dropped to 88% with the previous update, but it's back up to 90%.

My British and Irish started out at 29.5%, went up to 32.6% with an update, went down to 30% with another update. The current update sent that estimate way up to 44.4%. I believe the previous estimates are more accurate.

French and German estimates started out at 5.6%. It went way up to 13.9% with an update. It likely went way up after my Mother tested and her results were phased with mine. 23andMe phases ethnicity results if parents test which often changes estimates. Another update estimated I had 14.3% French German ethnicity. The current estimate has taken my French German down to 6.3%, which would appear to be too low because one of my Grandfathers was born in Austria and I have French Canadian roots.

Iberian estimate started out very low at 6.0%. I believe that was way too low considering one of my Grandmother's was Nicaraguan. When my mother tested and her results were phased with mine the Iberian estimate went way up to 18.4%. It went up again with another update to 19.9%. The current estimate is 12.3%.

Eastern European started out very low at 1.7% and went up to 4.9% after my results were phased with my Mother's. It went down slightly with the last update to 4.4%. It's gone up to 7.4% with the current. That estimate should be higher since paternal Grandfather's family was from an area on the Austrian Hungarian border.

Native American estimates haven't seen any wild swings. The first estimate was 6.7%, an update raised that to 7.3%. The last update took that estimate down to 6.9%. The current estimate took it back up a tiny bit to 7.1%,

Sub-Saharan African from 2015 to the present this estimate has never changed and stayed at 1.3%.

Ashkenazi has been added to my estimates with the current estimate at a trace amount of 0.1%.

My Mother's European estimate changed quite a bit from 75.6% European to 81.8% European. I believe around 80% is fairly accurate.

Conclusions

For some reason the previous update seemed to underestimate European admixture? It's now returned to the previous level. The current update raised my British and Irish from around 30% to 44% which is a substantial difference. My French and German estimates seemed to change for the worse while my Eastern European changed for the better. Looking at the updates from 2015 to the present they have become much more specific as you can see in the growing list of places. Personally I know where most my ancestors came from so the ethnicity results are not that important to me. Many Americans don't know where their ancestors came from so accurate results are important to them.

Update: When I added up my known British and Irish heritage it came out to 38%.

A new 2nd cousin match

Reviewing my 23andMe matches I also recently found a new 2nd cousin match. I was able to narrow down which line this match came from by looking at shared matches. The match is on my paternal Grandmother's side. Only one of her full siblings had children and I assumed this person was his descendant. Recently I had discovered my paternal Grandmother also had two half siblings, but I assumed they would have shared less DNA with me.

I contacted this match and she was nice enough to respond to my message. She indeed is a Granddaughter of my Grand-Uncle Frank Mason.

I've added the new segments to my Genome Mate Pro chromosome chart. Many of the segments overlapped with previous Mason cousins, but a few were new. One segment was extended as you can see looking at the red segments in the charts below. The one on the left shows all my collected segments before the new ones were added, and one on the right shows the Mason segments after I added the new segments.

I only met Frank once when I was a child and only remembered him mentioning one daughter. With this new match and two other 2nd cousin matches descended from Frank Mason I realized he had more than one daughter.

Doing research to find out exactly how many children he had I discovered a passenger list for Uncle Frank Mason and his first wife and daughter. According to this passenger list they were being deported from Cuba in 1939. I knew Frank spent time in Cuba, but hadn't heard he was deported from there. I have no idea why he was deported, but thought it was an interesting fact I never knew before.

Frank had been in Cuba even earlier than that visit. He returned from Cuba in 1934 at the age of 20. Evidently he used his birth certificate to prove he was a citizen.

I'm not sure why Grand-Uncle Frank spent time in Cuba? Maybe some of our Mullen cousins relocated there during prohibition? Or he was working as a painter there? He must have stayed long enough to learn to speak Spanish because my mother said his Spanish was very good.