Have you learned thru the years that spelling matters when doing an Internet search? On the other hand, have you found that spelling of names varies — and thus a specific spelling doesn’t matter any more? That need to be able to search for various spellings of a name was behind the development of the Soundex code.
Soundex code was very valuable in pre-Internet days for locating census records. It can still be used today with searches of Ancestry’s databases. Unfortunately, this concept isn’t used when Ancestry’s computers compares the trees of people who have a DNA match to identify the common ancestor. Instead, the computer is looking for an exact match.
As I’ve started researching an ancestor that Ancestry identified as a potential match, I’m running into spelling issues.
This new ancestor is a revolutionary war veteran, Major Simon Van Arsdale. In addition to his revolutionary war service, Simon Van Arsdale was part of the Low Dutch Settlement that migrated to Kentucky.
The discovery of Simon Van Arsdale as a potential ancestor is opening up doors to other potential ancestors and a lot of interesting history. Unfortunately, the spelling of the Van Arsdale name is making it difficult to locate records and to identify DNA matches. So far, I’ve identified the following spellings for this surname:
- Van Arsdale
- Van Arsdalen
- Van Artsdalen
- Van Osdol
For the most part, clicking to also use Soundex when searching Ancestry databases will help me get around the many spellings of the name. However, that option isn’t available when working with DNA matches. I recently learned that I should use the ‘alternate name’ fact to add variations on the spelling of a name.
This morning as I was thinking about the need to add ‘alternate name’ facts for Simon Van Arsdale, I saw a Facebook post questioning why Ancestry’s computers can’t find common ancestors when both parties of a DNA match have large trees. I believe the same post also talked about how changing the spelling of a name (Fannie to Fanny) caused the number of matches on a ThruLines to drop. In the comments on the post was a suggestion to add an ‘alternate name’ fact for the different spelling of the name. (Unfortunately, I haven’t been able to find this post back. Thus, I can’t give credit to the parties who wrote the post and the comment.)
In thinking about this question as to why the computers aren’t finding the common ancestors, I realized that spelling of surnames and name variations could be a big issue with my tree. I have a lot of places in my tree where the name I have could be slightly different from the name another person might have in their tree. However, I have one surname where this could be a big issue: CURREY.
Over the years, I have found that when the name is spelled with the ‘e’, the record is usually for someone in my line. I have also found records using the CURRY spelling that are for individuals in my line. Thus, the name could be spelled CURREY or CURRY. Since I only have the CURREY spelling in my direct ancestral line, I’m going to experiment with adding CURRY as an alternate name to see what happens to my ThruLines.
Below are the number of ThruLines matches for each generation of my CURREY line:
- Hiram Currey – Dodge City – 1866-1943 —– 3 matches
- Hiram Currey – Leavenworth – 1835-1901 —– 7 matches
- Hiram Currey – Peoria – 1787 – ? —– 16 matches
- HIram Currey – Ohio –? – ? —– 22 matches
- Thomas Currey — Ohio — ? – ? —– 16 matches
I will add the ‘CURRY’ alternate name for each ancestor and their children. Then I will re-check my numbers in a few days. Hopefully, I will see the number of matches increase!