ERIC

I felt that the Education Resources Information Center (ERIC) database would be the better choice for this search, following through on the search I performed last week.  To see the difference between a natural vocabulary search and a controlled vocabulary search, I started with a few terms that worked on the last search, except that I replaced “genealogy” with “family history,” because to me that seems more generic, and more like what a patron would use. 

37 hits.  Although the first one was close, I didn’t really get anything that I could use for my first three results: 

So I checked the thesaurus to see if there were any similar vocabulary terms that would improve my results list.  The first search term came up immediately: 

“Teach” didn’t come up as a term, either, but it provided an alternative: 

History was a valid term, so I kept that one.

My new search looked like this: 

And my first three results out of 46: 

Still not really what I was looking for.  Interestingly enough, I did find a few good results at the bottom of my search list, and a few that redirected me to JSTOR (my last database), and so I’d already seen them.  I wondered what would happen if I removed “instruction” from my search terms, but that didn’t work well, either.  I removed “instruction” and changed “history” to “history instruction” and got what I was looking for.  I had fewer hits, but they were more relevant to my searching: 

The third article is exactly the sort of information that I was aiming for in this search: a way to use genealogy and local history to teach students that history isn’t a dry, soulless list of dates and names.  It was lived by our ancestors, and without it happening, we wouldn’t be here today. 

I like the idea of controlled terms in theory, but I don't think that they are as easy to use as natural language, and I think it stumps a number of people who try to use the databases we offer at the library.  Patrons are used to natural language searching like on Google, and when databases don't bring up what they need after a natural language search, they just assume the databases don't have the information and go back to Google.