Read my lips: no more surveys!
The latest issue of the Journal of Digital Information is all about institutional repositories. A lot of it is garbage. Some is reports on initiatives that may well be worthwhile but won’t bear fruit for years (”asset actions” are definitely the right idea, but my head hurts at the mere idea of implementing them in something other than Fedora). Some is more vague business-speak happytalk prescriptions, and nobody needs that any more.
(Repository-rat to Thibodeau: It takes two to tango. The repository can do its level effing best to court allies, but if the platform is crap and possible allies just plain aren’t playing, what next? And does all the fault lie with the repository and its managers, as your article implies? If I sound frustrated, it’s only because I am. No fun being everybody’s whipping-boy.)
Now to my thesis: No more surveys. No. More. Surveys. Damn. It. They produce no insight, no practical suggestions, and no comfort. They’re useless. Stop sending them out.
Their counts are highly suspect, especially when they try to interrogate growth over time. This is partly because item counts are themselves substantially useless, and partly because I do not waste my precious work time obsessively keeping track of how much stuff I have at any given moment. I don’t know how fast the repository I run has grown—but I do know that some of the apparent growth is spurious, because I imported a large image collection recently, one item per image. Item counts are useless.
Since these surveys are usually not designed by repository-rats, the questions are often poorly-phrased or unanswerable. I do not know who does submissions in a lot of the repository I run. The software is designed such that I don’t have to know that, and frankly I don’t want to! I do not know, except in vague and impressionistic (and likely wrong) terms, what proportion of the materials in the repository are peer-reviewed versus student research versus ETDs versus interesting tidbits like the snowflake collection and the audio-lecture series I’m currently working on. I do not know which disciplines contribute what, and frankly I think it’s a useless question.
DSpace statistics suck. Find out how badly they suck before you ask me statistics questions (downloads, most popular authors, most popular collections, etc.) that I can’t bloody answer.
Moreover, I cannot accurately estimate the success rate of any of my outreach efforts, most crucially because results are rarely immediate. Sometimes something I do that I think is a complete dud bears fruit months or years later. Sometimes success is completely random. Sometimes interventions that I have good reason to believe would be successful are completely out of my reach. That’s life in big bad Repository City, and if it makes your damnable surveys unreliable, stop doing them.
I’ve bailed on a few repository surveys recently, they’ve made me so mad with their genial cluelessness and the number of questions on them that I can’t realistically answer. Keep that in mind when you read the next bog-standard research article based on a survey.
Qualitative research (which in my opinion is the way to go right now) hasn’t been done well yet. I talked to somebody a while ago who was very enthusiastic and fun to interact with, but obviously hadn’t done her homework in the library literature and didn’t know nearly enough about how academia and academic libraries function. I don’t have high hopes for what gets distilled out of her interviews; I’d rather see her publish transcripts, frankly, because those might be good.
Right now I’ve got some researcher wanting to talk to me about the planning process for the repository I run. Hello? I wasn’t there. Hiring the repository-rat is what happens after the planning process is done. This basic failure of clue makes me want to say extremely impolite things. I won’t, except here; I’ll be good and let myself be interviewed.
But I ask you, what does it take to find some researchers in possession of a clue, not to mention an actual desire to help people like me instead of treating us like lab-rats?