Here I am at the reference desk for one of my Sunday shifts this term.* Sitting here, trying not to think about the tornado watch we’re under, or about cook-outs, or about holidays in general, I started thinking about JSTOR and what a perfect example it is of a whole cluster of things I’m sure my students rarely think about. Namely, the role that collection development choices make on the kinds of information to which they’ll have easy access.

JSTOR, like any other collection, was developed and continues to evolve based on choices people make about what to include and what to exclude. I’m not privy to the inner workings of JSTOR, but I don’t have to know those details in order to know that decisions must have been made to exclude some journals from the collection and to seek out other journals for inclusion. There’s no way around it.

Students, though, assume that it’s comprehensive. If anything, they may know that it doesn’t have much recent content, but they often figure that if JSTOR doesn’t have what they need they’re sunk because probably what they’re looking for doesn’t exist. I’ve seen this assumption loom it’s frustrating head in classes, at the reference desk, in my office, and now again in a set of focus groups we had done. Truth be told, it’s not just the students that think of JSTOR as The access point for scholarly articles in the library… we’ve heard the same assumptions bubbling up from some faculty.

But if you only search JSTOR, how might this affect your views of the scholarly discussion on a particular topic? A PoliSci professor on campus loves to point out that the journals in JSTOR for his discipline come, by and large, from one particular theoretical framework within the world of Political Science. I’d never thought about this until a couple of years ago (when I heard him say that for the first time), so I thought I’d take a closer look at the Language and Literature journals there. And sure enough, there are a couple of interesting trends in evidence.

First of all, there’s a distinct preference for journals published in the United Kingdom or the United States. I know other English-speaking countries publish high-quality, peer reviewed journals on language and literature, but vanishingly few show up in JSTOR.

Then there’s the question of dominant theoretical frameworks. Well, after admitting in the comments on my last post that distinguishing theories of reading is difficult, I’m sure you’ll forgive me if I pain in broad strokes here. That said… all the ways of reading a text (which I define very loosely, but which is easier to type than “mode of expression”) choose some method by which expression and meaning interact and dub it the most important or interesting site of meaning to explore. Some look to the text itself and exclude other contexts while others look to the reader’s interaction with the text or to the cultural context surrounding the text’s creation. JSTOR journals tend heavily toward the cultural-context option and exclude, almost entirely, the text-by-itself option. (Interestingly, the curriculum in the English department here has historically tended toward the text-by-itself option.) So if, as many undergraduates are, you’re looking for examples of ways that scholars have interpreted a given text, JSTOR will only give you a piece of that story because it seems to have chosen to develop its strength in contextually based criticism.

Of course, none of this is meant to imply that JSTOR is trying to pull one over on its subscribers, or that it’s not valuable, or anything like that. They can’t digitize everything, and they do have many of the absolute core journals in my disciplines. My point is just that students think it’s comprehensive, but it’s not. This is why we subscribe to more than just JSTOR. Each collection, physical or virtual, has it’s strengths and weaknesses, and remembering that this is true is a key part of becoming a savvy researcher.

* Life at this college only gives passing glances to holidays as it plunges headlong through 10-week terms. In the case of Memorial Day, this glance takes the form of an email message from the college president urging us to “take a moment” to honor those who died defending our nation, a vigil in the campus chapel, and probably sidewalk art. We enjoy sidewalk art here.

I wish it were possible for our bibliographic databases to index works according to the methodology or theoretical approach of the author. I know it’s not possible in many cases, but man oh man I wish it were. For a lot of undergraduates, finding examples of scholars employing a methodology or approach is high up there on the list of information needs, and it’s just not something that search can really help with.

So what do you do when a student comes and needs an example of a Marxist reading of The Wasteland or a post-structuralist reading of The Between? What about students who want to see examples of formal lab write-ups, or those formal research papers that actually use the prescribed headings Introduction… Methodology… etc? Part of the learning process is seeing accomplished researchers employing the forms that are taught in the undergraduate classroom, so for these students methodology and approach is a very real form of content to be sought, found, and analyzed.

Personally, I’ve developed some inefficient work-arounds for some of the questions I get most often. Try searching for research articles that have “study of” in the title… people likely to put “study of” in their research reports are also likely to use the traditional subheadings. And for the theoretical-approach question, I keep an anthology of the major literary theorists in my office and then do this complicated thing where I do a cited reference search in Web of Knowledge for an appropriate theorist, then I do a search for the topic or author or work we’re trying to interpret, and then I combine the two searches. Sometimes that works fabulously… sometimes it flops. But at least it’s something. (Of course, this doesn’t work very well at all if you’re trying to find an example of close reading… since close readers often only cite the work they’re reading, er, closely.)

‘Tis a puzzle. It’s one of those examples of a very common kind of research task (for my population, at least) that cannot be solved without quite a bit of prior disciplinary knowledge. Search, just by itself, will almost always fail.