12 March 2014

Stevens Finds Stephens in 1810, but not the other way around...apparently

[note: this post has been modified since it was originally posted as the "problem" has been discovered--it was in a setting on my search box. The modifications to the original post are in red.]

Hopefully someone will correct me if I'm making a mistake. Russ W. tried to validate these results for me and was unable to do so. When I went back an hour later and searched again the results were correct if the "exact" button is left unchecked on the search entry page. 

I searched for Rufus Stevens in the 1810 census index at Ancestry.com using the slider to perform a search on both the first and last names as:
"Exact, sounds like, and similar"
Three results were obtained:

  • Rufus Stevens in Vermont
  • Rufus Stevens in Ontario County, New York
  • Rufus Stephens in Lewis County, New York


Makes perfect sense.

But...

I also searched for Rufus Stephens in the 1810 census at Ancestry.com using the slider to perform a search for both the first and last names as:
"Exact, sounds like, and similar"

I only obtained one: the Rufus Stephens in Lewis County, New York. Note: at 11:15 am on 12 March 2014, I obtained correct results for this search--the same three entries as above--but to obtain the correct entries the "exact" search button on the search page must be left unchecked. 


It would appear to me that the results should be the same and they now are--after all Stephens and Stevens are soundex equivalents and they certainly sound alike (at least in English).

And, if a search for Stevens using "exact, sounds like, and similar" locates Stephens entries, shouldn't a Stephens search using "exact, sounds like, and similar" locate Stevens entries?

Or am I doing something wrong? If I am, I'd be happy to post a correction.

I'm still not certain what I was doing wrong. And really wish I knew. 

No comments: