Looking back over the last two weeks, I can’t believe I’ve only posted twice! That’s got to be some kind of record for me. But I had car issues, and then vacation (which was WONDERFUL and long), and then I spent all day yesterday trying to catch up from being away. I didn’t succeed, of course, but I made a healthy dent.
Anyway, I’ve become curious about link resolvers and how academic institutions set theirs up. We use SFX, and so far I’m absolutely thrilled with it. I think of it in terms of eras: I graduated from library school in 1B.SFX (one year Before SFX), and it is now the year 2A.SFX (two After SFX). My library implemented it the first term that I was working here, so I never had to teach the tedious methods of hunting through databases, using Ulrich’s, interlibrary loaning, having the loan request rejected because we have the article (somewhere), looking again, searching Google, and then resorting to cursing and pulling out hair when trying to find the full text of an article. Instead, all I had to teach was “click on this handy little button and it will search all our databases for you.” Who wouldn’t be thrilled?
But there were all sorts of decisions that went into setting up our instance of SFX, decisions I wasn’t a part of, but that I need to think about now as we go through the endless cycle of evaluating our service, tweaking it, and evaluating it again. At issue right now is the question of whether or not we made the correct decision when we chose to have an SFX menu pop up every time a user clicks on the Find It button, or whether we should switch to having it deliver full text (if available) from a preferred database.
For those of you who don’t have SFX, or who have it configured differently, here’s what I mean. Let’s say I search and find a citation for an article from the journal Sociological Perspectives. Click “Find It” and I’ll discover that we have two subscriptions from two different databases, and St. Olaf also has the print. Click on either of the database options and I’ll get the full text; click on the print option and you’ll be taken to the catalog record for location and holdings information.
This system has three distinct advantages. First, it helps students figure out which databases they should be searching other than the database they’re currently searching. I teach my students to keep notes on databases that come up often when they’re looking for full text because those databases obviously have a collection of journals on your topic. This technique is especially useful for upper class students when they’re beginning to do research for their comprehensive projects. But it can also work the other way. I’ve mentioned before that unlike my co-workers, I mainly teach freshmen. And consequently, I also teach the less specific databases more often than my co-workers do (my co-workers rarely have to deal with students who just have to write “a research paper” on any topic). So I tell my students to watch the Find It menu to see if there’s a more specific database on their topic.
Another reason for providing choice, beyond simply my love of letting people choose what works best for them, is that over time students develop familiarity with the databases they use most often, so they may prefer to click on a link to their favorite CSA database, Ebsco database, or ProQuest database when it appears in the list of choices. I know I default to certain databases over others because I know that I can manipulate some vendors’ databases more easily than others.
And finally, some poor, young students dealing with requirements for “print” sources handed down from professors who haven’t explained that you can get “print” sources online from these databases, would really just prefer to click on that “available in print” link and go photocopy the article for themselves.
On the other hand… there’s value in immediate delivery whenever possible. I’m torn. Is delivery more important than education and discovery? Possibly. But I’d hate to give up the information I’m given when I’m presented with choices.
Carleton and St. Olaf will be deliberating on this issue soon, and I’m comforted that we can’t make a “bad” decision. I just wish I knew which was the better decision.
tags: libraries, link resolvers, open URL
Have you thought of user testing? I might try and find a few students who haven’t used the SFX resolver before–if your outreach is *so successful* that you can’t find any, try students at another college or high school.
Come up with some examples that will give them a variety of experiences (multiple full text links, no full text link, etc.) and see how those students deal with the link resolver screen in its current setup. See how often they end up with the article.
Perhaps have other students, who have used the current version, try the new direct-to-fulltext method. See if they prefer it, or if they miss the extra information.
Perhaps those aren’t the best tasks, but I believe that seeing what the users think could be helpful. And remember that not every user will have the benefit of one of your instruction sessions. HTH.
What?!? I’m not reaching all the students?!?!? :)
Very good points, as always. And I like your idea of testing the experience, somehow. But then there’s the dark side of me that wonders if one or the other is “better” regardless of the students’ preferences… Of course, I’ll deny I ever wrote that sentence if you ask me. (Sure… somebody hacked into my account and commented… it happens all the time… just ask Google.)
Well, you can try and do user testing that doesn’t really ask for their preferences, but instead tests how successful they are in using the software.
If they can find the full text reasonably quickly without help in either instance, then you can just pick the “better” one. But if people get confused with one or another of the interfaces, I’m sure you will do the right thing. :)
True. Very true. And another thing I just though of: we’d have to be VERY rigorous in making sure that all our information was absolutely current for our top-choice databases if we stop providing choices because there’d be no way to get around a dead link.
Here you’re getting a peek into my thought-process:
1) Cool!
2) Wait, why did we decide to go the way we did the first time?
3) What could go wrong if we change?
4) Would it be so terribly bad if that did go wrong?
5) Still… it is cool.
6) Test stuff.
7) What do the test results REALLY tell us?
8) Throw all answers and thoughts from steps 1-7 into a hat.
9) Disregard the items in the hat.
10) Go with overall gut feeling…
11) Implement decision.
12) Evaluate the implementation.
13) Repeat 1-12 as necessary, always being open to saying, “Wow, we didn’t realize THAT would happen. What were we thinking?”.
Ironic that there are 14 steps. Maybe I should think up some more so as not to incur the wrath of the luck gods.
We provide the best of both worlds at the Tri-Colleges. We have our DirectLink option configured to, for those items that provide full text, rank them on reliability and send users to the most reliable. If there’s no full-text we provide the menu with options.
For those items where the user sees the full text, we’ve done custom programming to ‘frame’ the content with a link back to the full menu. It work’s great for us.
Ooooh, I like the idea of framing the content. Actually, I hate framed content, but I love the idea of having a way to link back to the full menu. Do you have any statistics on how often that link gets clicked?
Hi Iris. I was trying to frame a response, but couldn’t find words. Then I came across this post about Agents and Gatekeepers on Library Garden.
I think the “these are your alternatives” is a “gatekeeper” way of looking at it, where “here’s the full text” is an agent way of looking at it.
I know I prefer to meet agents, not gatekeepers when I’m receiving service, but I’m still not convinced that libraries (who have responsibility to ALL users, not just the one they are serving now) should drop the “gatekeeper” role completely.
So there you are..none the wiser, but two more buzzwords to play with :)
While I agree that linking directly to full text is completely an incontrovertibly acting as an agent, I’m not sure that providing options erects barriers for my users. Or rather, I’m not sure that the barrier of one extra click outweighs the barriers of not discovering fruitful areas for research.
I’m completely playing devil’s advocate, here, because I’m really still trying to sort out what I think about the issue. But it seems that in all the discussions about being “agents,” we lose a little bit of perspective. We always assume that “agents” help me get what I need AT THIS MOMENT but forget that they can also enact assistance that goes beyond the present. (Remembering, of course, that I’m speaking from an academic library that has been working to integrate information literacy into the thought processes of our faculty and students).
So the contrarian in me, the devil’s advocate, protests that I am training students to think about research, information gathering, information production and dissemination, and all the other facets of information literacy. What I want to do is resolve this contrarian in me with the other side of me that is trained to deliver information when asked.
I’m so confused. And I’m not at all wise. But I’ll have fun throwing new buzzwords around at our next departmental meeting. :)
Adam and Kathryn have interesting ideas. I, too, like the idea of getting back to the link resolver screen, but am not a fan of frames. But no solution is perfect.
As for agents and gatekeepers, I’d say (he said, climbing back on the hobbyhorse) that a quick user test will tell you if the intermediary screen is really much of a barrier. When a student encounters it for the first time, do they say “WTF? What should I click?” Or do they say “Hmm, ‘choose a source for full text?’ OK, I choose this one!”
Good point, Steve. I’ll be sure to advocate for at least a small usability study when we meet in January.