Post
by morgano » Tue Dec 16, 2008 8:57 am
I use Ancestry quite a bit and shall continue to do so, but I have recently found a new (to me) hazard with Ancestry records, that I ought to share.
It involved the transcription of "ditto". In a Welsh census record, an address inhabited by family Rees was followed by one inhabited by family Evans, so that there was one "ditto" under the surname Rees and there were five dittos under Evans. The transcriber clearly misunderstood the concept of the "ditto", so translated all six of them as "Rees". If I hadn't already known enough about the family to recognise that it was the one for which I was looking, despite the error, that might have thrown me. In the event, I had searched on the one Evans family member (out of six) who was correctly indexed as such. If I had been searching on his wife, or their daughter (perfectly plausible, since they are my ancestors, too), I should not have found them in Ancestry, because they were both listed as Rees. (I have reported all the errors to Ancestry for correction.)
This could easily happen for Scottish census transcriptions in Ancestry, too, and might be harder to detect, since the actual census forms for Scotland are not visible via Ancestry, of course. Precisely the same kind of mistake is not likely to be made by SP, but SP does also make the odd error and I have needed Ancestry in the past, when a census record has been invisible in SP. May be just one to bear in mind.
