Additionally, never trust an index. Genealogical documentation has many layers of potential for human errors. There could be mistakes made when the record was originally created or it could just as easily have a error made when the record is transcribed into machine readable type. People have to type up the text of a document for a searchable index to be created. Wellllll....that is not entirely true. Now with artificial intelligence and its predecessor OCR (Optical Character Recognition), people don't always create the index. Although, people should be the final reviewers of those indexes.
For years and years I searched for my Grandpa Earle and his parents in the 1930 U.S. Federal Census with no success until I tried a search constructed with just the state, county, and household members first names - - no last name. And sure enough Abram, Ethel, Allen, and Edwin came up in New York, Nassau County, Town of Hempstead indexed as Carle, not Earle. And look at the image:
See the way that E is written? Totally looks like a C. And even soundexing searches wouldn't pick up that mis-transcription; if you could even call it a mis-transcription. Honestly, that is just shitty penmanship if you ask me.
Another example of an error in the index comes from working with another researcher. She knew where the family should be living at the time the 1940 census was taken. When we look at the index page on Ancestry if clearly indicated the family was black and my client insisted that it could not be her family even though all the names and address matched. "Really? You really don't think it's not them?"
Of course it is them. If you open that image of the actual record you can clearly see that the family is listed as "W" as in white.
No comments:
Post a Comment