New Post
Search
Help
ArchivesRe: Is Ancestry.com a reliable source?
in reply to a message by Mossy
ancestry.com (and other big genealogical sites) combines a lot of data sources of very different quality like census records, transcripts from church registers, and user submitted information.
All those sources aren't perfect, they contain errors of different types: Census takers may mishear and misspell names they note, transcribers may read the handwriting wrong or there are OCR errors, someone may inadvertedly or even deliberately enter wrong information.
So when you find something there, you still need to employ some judgement. I won't give much on a name form that is only recorded once in one census in the USA, it is probably wrong. When the same name form occurs in independent sources, things are better and more reliable.
All those sources aren't perfect, they contain errors of different types: Census takers may mishear and misspell names they note, transcribers may read the handwriting wrong or there are OCR errors, someone may inadvertedly or even deliberately enter wrong information.
So when you find something there, you still need to employ some judgement. I won't give much on a name form that is only recorded once in one census in the USA, it is probably wrong. When the same name form occurs in independent sources, things are better and more reliable.
3
Replies
And to expand on what's already been said:
You have to be very careful about just using data in the indexes in ancestry.com. Indexers often don't pay much attention to the characteristics of the handwriting of particular census takers and tend to guess that names which they aren't familiar with are more modern names. So you often need to check the actual census record, not just take the data presented in the index at face value.
For example, when I was doing research on the use of Pearl as a given name, I saw there were several supposed examples of women named Pearl who were born before 1850 in the ancestry.com indexes of the US census. When I checked to photocopies of the original census records, every one of these turned out to be a mistake -- most commonly the indexer read the now-rare New Testament name Persis as "Pearl". It turns out there really are no women named Pearl in the US census born before Nathaniel Hawthorne's novel "The Scarlet Letter" was published in 1850, but you would think there were if you just relied on the indexes.
You have to be very careful about just using data in the indexes in ancestry.com. Indexers often don't pay much attention to the characteristics of the handwriting of particular census takers and tend to guess that names which they aren't familiar with are more modern names. So you often need to check the actual census record, not just take the data presented in the index at face value.
For example, when I was doing research on the use of Pearl as a given name, I saw there were several supposed examples of women named Pearl who were born before 1850 in the ancestry.com indexes of the US census. When I checked to photocopies of the original census records, every one of these turned out to be a mistake -- most commonly the indexer read the now-rare New Testament name Persis as "Pearl". It turns out there really are no women named Pearl in the US census born before Nathaniel Hawthorne's novel "The Scarlet Letter" was published in 1850, but you would think there were if you just relied on the indexes.
4
What CKE says is so true. For instance, I have found seventeenth-century babies christened Tace given as Tracey by modern transcribers, and babies named Georgina in Victorian Britain transcribed as Georgia. The only babies in Victorian Britain actually named Georgia turned out to be the daughters of Greek parents. What did surprise me, though, was to find that eighteenth-century babies recorded by transcribers as Melody really were named Melody, a name I would have thought was coined in the 20th century.
1
Thanks! I'll keep that in mind.
1