Thursday, January 22, 2009

Why online databases are like pregnancy tests

So here's the thing. We don't really expect fallibility from computerized systems, and for scholars such systems increasingly mean online databases. I've grown to trust the MLA database, Google Scholar, the OED, etc. How else can we know what has been written and where we can join a scholarly conversation? But these databases are only as good as the data that is put into them, mostly by imperfect humans. That means that they tend to err in omission rather than in commission. If something appears in a database, it definitely exists. But if it doesn't appear in a database, it may still exist. This is a lot like pregnancy tests that have false negatives but never a false positive. I can use my own work as an example. My dissertation (UVa 1995) appears in the MLA database because it was indexed by Dissertation Abstracts International (DAI). It also appears in the UVa library catalog where the only paper copy (other than my own) resides. But it does NOT appear in a WorldCat Dissertation search, nor does it appear in a Google Scholar search (possibly because Google Scholar relies on WorldCat). Why? Who knows? Conversely, my very first published article, on early modern stigmatics and politics (Viator 2000) turns up in a Google Scholar search (use "MacInnes" and "Nun of Portugal") but only in a French database. It's not recorded in the MLA database even though the MLA does index Viator. For some reason, only one article from the whole 2000 edition made it into the database. Is this because some of the articles are not deemed to be literary? Who knows. What all this means is that it's really difficult to determine what has not been said on a given topic. One can try a variety of searches, but even if each one returns a negative result, it doesn't mean the information does not exist. Perhaps, given the amount of information that is now being put online, the newest kind of research will operate on the basis of finding and using "hidden" sources, the things that fell through the cracks in the digital world. They will be really hard to find!

1 comment:

  1. If it doesn't appear in a database, it may still exist. So true so true. My diss shows up in world cat but three of my articles, which were published in anthologies and a yearbook that were published in Germany don't show up in MLA OR Google scholar.

    ReplyDelete