Chasing Our Long Tails

The Chron (subscription required, dang it!) reports briefly on research published in Science that, in spite of access to full text archives online, researchers are citing less, and including a smaller range of sources because they follow links from one paper to another. Of course, they’ve always done this – but perhaps are now more likely to cite those that are available in online archives and ignore those that are not. The reporter, Lila Guterman, had the good sense to check in with Carol Tenopir, who has been studying research behavior for years and finds these results quite the opposite of what her studies have shown.

My personal interest isn’t so much in scholar’s behavior as in that of undergraduates, because that’s the population I teach. And my own very limited research (conducting a handful of in-depth interviews in the early 1990s, repeating the study in the early 2000s) suggested that the problems students have doing research have nothing to do with changes in technology and everything to do with gaining enough knowledge to frame a good question within a matter of weeks, reading and writing in an unfamiliar form of discourse, and embracing their role as knowledge producers, not information reporters. Post-Internet students’ responses to my questions were totally consistent my findings in 1992, even though the pool of potential resources at our small library had turned into an ocean. Finding a focus, selecting and understanding sources, and developing their own voice remained their biggest challenge. Using the library and/or Internet to access materials was a doddle in comparison.

a Wordl picture of our article introduction

(This picture, by the by, was generated with Wordl.)

I’ve been mulling the enormous amount of information available to the average undergraduate since reading a piece of research in the Harvard Business Review. (Hold on – this is Steven B. territory, no? What’s an anarchist like me doing in a journal like that? Siva pointed it out.) Harvard Business School professor Anita Elberse challenges Chris Anderson’s Long Tail theory – that the digital world will not only enable discovery of niche products, but will change what we choose. Elberse says it’s the opposite – the digital environment actually amplifies the dominance of blockbusters. She also says that crowds, in their wisdom, gravitate toward blockbusters because they find them more satisfying than less-well-known items, and manufacturers and retailers should therefore put their money on known winners, not on promoting a longer tail. Naturally, there has been much debate about her methodology and conclusions, but it’s all very thought-provoking.

Naturally, my thoughts turn to undergraduates struggling to find what they need in an information-rich environment. Perhaps their experience with Wikipedia has been that it’s easy and it works better than more obscure alternatives. They have less trouble finding and deciphering the meaning of Wikipedia articles than they do making choices among thousands of scholarly articles and then having to figure out what an article means when it’s written for experts, which they are not. The blockbuster works. Except they don’t learn how to do the hard stuff or interpretation and building new meaning, which is why we torture them in the first place.

But what scaffolding helps them succeed at the hard stuff? And how, amidst the enormously long tail of information that students could use, do they find good sources – the kinds that can be used to build an original and compelling understanding of whatever it is they’re researching? We pay a lot of attention to exposing students to the abundance; not so much with the much harder job of making good choices. Wherever you fall on the Elberse / Anderson debate, we’re making a false assumption when we say more is always better.

Amy Fry, Julie Gilbert and I just published an article in portal (a self-archived copy is here) that had some surprising findings about the long tail in aggregated interdisciplinary databases: looking at use of one of the market leaders at 14 largely undergraduate institutions, 4% of titles accounted for half of downloads, and these were largely popular titles; articles in 40% of full text journals were not downloaded even once at all 14 institutions. We also found that, in aggregate, the number of articles downloaded fell from 2005 to 2006 by 10%, even though the database itself was growing. Curiously, a survey of librarians show they think these growing databases are about the right size and that more full text would be an improvement. Is more always a better investment? Really?

In the case of undergraduates – it’s complicated. They don’t think more is a problem with Google results. But that’s because they don’t have to sort through them; they look at the first page or two. Google’s ranking algorithms are far more sophisticated than those used in our databases. If they find what they’re looking for in the first two pages of a Google search, but do not in a library database, they get frustrated. (I know, they may be wrong – and they shouldn’t be lazy, etc. etc., but it works for them. And the databases don’t.)

Given undergraduates’ understandable difficulty in making sophisticated choices about sources in fields that they know little about, faculty sometimes try two unhelpful approaches. One is banning all Internet sources. Another is giving up on self-exploration altogether and relying entirely on assigned reading; that way the sophisticate – the teacher – eliminates the problem of making unsophisticated choices altogether.

This, of course, means students don’t get practice doing something they will need to do in future. They may have to find out whether that claim a presidential candidate is making about immigration has merit or not, even if they don’t have a degree in immigration studies. They may have to decide whether the treatment their doctor is recommending for their cancer warrants a second opinion. They may want to make a good case before their city council that the coal-fired power plant down the block is a health risk or that the effectiveness of the reading program being used in their fourth-grader’s class is not only confusing the heck out of their child, but isn’t supported by research. They may want to be able to think for themselves. At least, we hope they will. And for that, they need to be able to make sophisticated choices about things they don’t already know.

Relying on blockbusters – Wikipedia or Google or USA Today or the book / movie / person everyone is talking about – won’t cut it. But neither will simply assuming they’ll find it in the long tail. We need to think hard about not just increasing our resources and our training on how to use them, but helping faculty help students develop the ability to get to the good stuff. And not just to complete that paper, but to complete themselves as free and thoughtful human beings.

The Creative Library

It’s rare that I’ll write about one of my personal projects – maybe a casual link here and there – but today I want to share with you the link to a recent project that I’m particularly proud to bring to your attention. This past spring semester I engaged in a unique experience. For the first time in my career I served as the guest editor of a journal issue. A good friend and colleague, Lisa Finder, a librarian at Hunter College and current co-editor of Urban Library Journal, invited me to serve as the guest editor of the spring 2008 issue. When she said I could choose any theme I liked that sealed the deal. After some careful thought I decided to assemble a collection of articles that would showcase the creative abilities of librarians. We call this issue “The Creative Library“. Lauren Yannotta, also a librarian at Hunter College, is ULI’s other co-editor.

If you are new to Urban Library Journal you should know:

Urban Library Journal is an open access, refereed journal of research and discussion dealing with all aspects of urban libraries and librarianship, welcomes articles dealing with academic, research, public, school, and special libraries in an urban setting.

The editors and I were amazed at the number of quality manuscripts we received in response to our call for papers. Choosing those to include was quite difficult. I think you will find the articles in this issue offer great examples of creative librarians at their best. For an overview of what’s included take a look at my introduction to the issue. Here’s a snippet from that overview:

That’s why this special issue about creativity in libraries is just right for the times. First, it’s important to celebrate the many creative minds working in this profession. Libraries have traditionally worked with restrained resource pools. To have come so far with so many successes is owing to the high levels of creative thinking in our libraries. Second, as we find ourselves in times of rapid change our most valuable asset is our ability to master the art of adaptation. If one program fails, if users seem to be going elsewhere for their information, if user expectations shift unexpectedly, then library workers must use their creativity to quickly adapt. By understanding our user communities, we can create new programs that leverage our skill sets to deliver new services and new ideas that will continue to make the library a community destination, both physical and virtual. We have compiled here a set of dynamic articles that demonstrate that there is no lack of creativity in the world of librarianship. But you probably already knew that. Anyone who has worked in this field for any length of time knows there are many creative people attracted to the field of librarianship. Yet we rarely use our journal literature to promote the many acts of creativity happening at our libraries. This special issue of Urban Library Journal changes that.

Did I say that this is a free, open access journal. So it’s free. What are you waiting for?

How To Handle A Blog Attack

The more you blog and share your opinions when you blog, the greater the likelihood an anonymous party will disagree with you in a non-collegial manner. That’s a polite way of saying you’re going to get damn near slandered. If you want to disagree with something I write, and ACRLog readers do all the time, I respect your right to do so. Use the comment box to share your thoughts and challenge my ideas, but do keep it collegial and add to the discourse. When an anonymous blogger makes a personal attack or invents a few false accusations, that’s downright nasty. When it happened to me my initial reaction was to respond to the attack. It’s only natural to want to defend yourself. But I knew that would probably be a losing proposition so I decided to ignore it and move past the incident.

Based on something I just read in the July 2008 issue of WIRED I’m glad I took the high road. In this month’s Mr. Know-It All column (see letter #2) a restauranteur wanted to know how to handle a nasty, slanderous and anonymous review of his restaurant that was placed on Yelp, the review site. I thought that WIRED’s Mr. Know-It All gave some good advice on how to handle the urge to attack in return:

No one enjoys being raked over the coals by a pseudonymous commentator, especially when the attacker is motivated by hostility rather than honest disagreement. But don’t credit your detractor with too much influence. You need to trust in the sophistication of online-savvy consumers – specifically their ability to see the big picture and factor out aberrant comments.

So I will take this to heart. If one person wants to attack me anonymously, that’s fine. I can balance that with the dozens of well-meaning colleagues who e-mail me or talk to me at conferences to let me know that they derive value from my blogs, or they’ll mention a particular post that inspired them. I really appreciate that, and want to thank those of you who have taken the time to share your thoughts. It really means a great deal to any blogger to know that someone appreciates the time and effort he or she puts into their blogging.

Don’t Ask Librarians For Help – Just ChunkIt!

As I wander through the massive ALA exhibit floor I’m always on the lookout for new and interesting vendors I’ve never heard of before. There’s usually a few, particularly those hawking some new technology. One that caught my eye in Anaheim was ChunkIt!. ChunkIt! is a search engine, but the idea is to use it to refine searches that you conduct on other search engines, like Google or Yahoo, or even library databases like EBSCO, ProQuest or LexisNexis. ChunkIt! basically lets you refine an initial search by adding an additional word or phrase, and then it presents the results in a split screen with the original results on the right and the ChunkIt! results on the left. Now this can actually provide some interesting results with search engine output.

So being an inquisitive librarian I went to the booth representative and got my brief demo and allowed myself to be signed up as a ChunkIt! beta tester. When I got back to work I had an email message waiting in my inbox telling me how to get started with ChunkIt!. Once I had installed ChunkIt! I experimented with some Google searches and it actually did some pretty cool stuff. For example, I searched my name in Google and since it is so common I get tons of results. Then I typed “keeping up” as a phrase in the ChunkIt! search box and I quickly narrowed the search results to just those items about myself and my keeping up resources (great vanity searching). At the ChunkIt! booth a primary selling point for librarians was the ability to use ChunkIt! on library databases. I wasn’t so sure about how well that would work or if it was even a good idea given the many search refinement features built into the typical library database.

I quickly received another email from my new friends at ChunkIt! telling me I could learn more about how to use it by examining some of their cool videos. So I did. Here’s an interesting one. It follows a student with a tight deadline to get a research paper done. Take a look.


Let’s breakdown this helpful video.

First of all, would any academic librarian recommend using LexisNexis Academic for research on the russian revolution? So that’s the first reason a student should ask a librarian for help. Using the wrong database for your research is usually the first step in an unpleasant research experience. Realistically though, raise your hand if you think the student would go right to Wikipedia.

The video continues to do a good job of demonstrating why students can really benefit from asking a librarian for help. Wouldn’t it be so much easier to just type “lenin” into the box right on the LexisNexis results page that says “search within results”. Why would you go to all the trouble to do the exact same thing with ChunkIt! when LexisNexis has already built that feature right into the database? But you knew that. The folks who made this video obviously didn’t talk to a librarian before writing the script. Furthermore you can do cool stuff in the LexisNexis “search within results” box like ATLEAST3 to quickly narrow to articles that mention Lenin three times (much more relevant than those that mention it one time) or LENGTH>500 (a longer article usually has more substance).

And another thing they didn’t realize is that LexisNexis is the original text chunker. We call that KWIC output. Yes, you don’t have to look at the results in list format. Just drop the “show” box and change the output format to KWIC and voila – you’ve got chunks of relevant text.

Click on the thumbnails below to see what I mean.
Example One:
lexisnexis screen shot

Example Two:
lexisnexis search example

So while ChunkIt! may have some useful applications for search engines that offer no useful features for search refinement, it may not work as well for library databases that do offer a nice range of methods for students that do need to quickly move from very broad to more narrow results. That’s the whole nature of search. The challenge for academic librarians is educating our user community to be aware of these features until such time as the technology and interfaces make them simpler to find and use. But in the meantime, I think there is a good use for this video. Why not show it in library instruction sessions as a good example of why it’s a bad idea to wait until two hours before your paper is due to start your research when you can ask a librarian for help the day before.

ACRL At ALA In Anaheim

The ALA Exhibits are getting more organized. Vendor booths are being organized into “pavilions” so that you can quickly check out all the competitors showing similar products and services. ACRL’s traditional booth was a part of the ALA Pavilion. Below is a photo I took at the ACRL booth.

acrl booth at ala 2008

The two academic librarians shown in the photo, staffing the booth when I stopped by, are Lori Mestre, Digital Learning Librarian, (on the right) and Merinda Kaye Hensley, Instructional Services Librarian, both of University of Illinois Urbana-Champaign.

I can’t speak for others, but this year’s conference exceeded my expectations – although I did set my expectations somewhat low in advance of the conference. The weather was great, it was fairly easy to get around, many programs were within walking distance, and while Anaheim isn’t exactly known for its cityscapes and cultural attractions, it did a good job of accommodating ALA. Everything considered I would call it a successful conference.

ala in anaheim 2008