Jul 17

Peter Brantley

Peter Brantley

If Libraries had shareholders

In my day job as the Director of the Digital Library Federation, I represent a small number of very large research libraries. Given my constituency, I've often wondered what the real impact of networked electronic information resources is on the core traditional business of libraries - lending books - but I've never run across any statistics on this. I have wondered in other blogs (see: "Lost Cathedrals: Libraries and Steel") whether libraries might (to put it crassly) turn into acquisition agencies for licensed content, with small cafes on their ground floors or basements, existing in the physical realm primarily to serve as community centers for students. My conversations with university librarians (the library directors) have recurrently seemed paradoxically positive; I still hear comments like: "We think the Internet is probably increasing traffic, because people see information online, and then they come into the library and utilize our resources." They then go on to discuss the steady increase in numbers they see in "gate counts" as more and more people come into the library.

And then yesterday my friend Jerry McDonough of the University of Illinois' Graduate School of Library and Information Science forwarded me a talk that he gave recently at the British Library called, "We Are Not Alone: The Role of the Research Library in a Suddenly Crowded Information Universe." It contained some slides that made my eyes open very wide. His explanation of the slides is better than I could provide, so I've replicated the analyses on my own, uploaded them below, and with his permission, interleaved his narrative.

These statistics come from a larger collection of research libraries - the Association of Research Libraries (ARL) - than my own organization represents, many of whom are smaller; typically DLF institutions are more idiosyncratic in their profiles. The statistics that Jerry and I generated use median figures across all ARL member institutions for which data exist on the variables utilized (year endpoint = 2003). ARL and the University of Virginia have built a simple web interface to the main ARL statistics, so anyone can replicate or play with other analyses to the extent the data permit.

Over to Jerry -

This slide [below] shows the median ratio of reference queries to full-time students for all ARL libraries from 1995 through 2003. As you can see, right about 1997 (which would be about the time the Internet started seeing real wide-scale uptake in the U.S.), the number of reference questions we're getting relevant to our student body plummets like a rock. And in case you're wondering, no, this is not the result of a sudden vast increase in the number of full-time students enrolled in U.S. universities. Libraries simply aren't getting anything like the number of reference queries they did ten years ago.

Reference Queries / Fulltime Students

This slide [below] shows the median ratio of total circulation to full-time students. As you can see, it's the exact same story, although the decrease gets going a bit earlier. We simply aren't checking out anywhere near the number of books that we used to. Some of that can certainly be accounted for by the use of electronic materials that don't count in ARL libraries circulation figures. But notice that this decrease is already well underway by 1995, when electronic journals did not have anything like the degree of penetration into library collections that they have today.

Total Circulation / Fulltime Students

Jerry provides this closing analysis:

Undergraduates entering universities in the United States use the library as a study space, a socializing space, but to a shocking and frightening extent, they do not use library services or library materials. [...] We're losing clientele; students may come in the library to study, to socialize, to hit the newly installed cafe designed to lure them in, but they're not using library materials, or library services, at anything like the rate they did even ten years ago.

There are a variety of potential explanations I could put forward for this, but these would be guesses. However, while noting that correlation need not imply causality, this drop does coincide with the growth in Internet use in the U.S., and whether it has anything to do with the Internet or not, it shows a dramatic and negative change in our relationship with our patrons. We obviously do not occupy the same place in their scholarly or social world that we used to, and given University administrator's likely reaction to looking at graphs like these next to our escalating materials budgets, I think we need to spend some time figuring out why that is. We cannot afford to end up on our own in the universe of scholarly information.

I would be half-prone to thinking that these graphs had some peculiar resonance that should obviate or blunt their obvious interpretation. But last night, I chanced to read a not-yet-released document on another topic, and it held a chart that had me dancing back to these ARL graphs. The graph reproduced below is derived from publicly available data and it reflects the total number of physical visits to Library of Congress reading rooms. Lo and behold, over almost exactly the same period, we can see nearly the identical slope -- in recent years, monotonically heading downwards.

These are fascinating results. They do not reflect on the total value of libraries, and they surely do not pass judgment on the highly skilled information specialists that staff them. They do suggest that something momentous has changed in the fundamental environment that libraries operate within. And one has to think: if libraries had shareholders, would they, like newspapers, be in the midst of a gut-wrenching, brake-screeching exercise in redefinition?

tags: libraries  | comments: 30   | Sphere It

Previous  |  Next

0 TrackBacks

TrackBack URL for this entry: http://blogs.oreilly.com/cgi-bin/mt/mt-t.cgi/5693

Comments: 30

  Pilar [07.17.07 04:17 PM]

Yeah, I think so

  Margaret [07.17.07 04:50 PM]

With or without shareholders, libraries *are* in the midst of a "gut-wrenching, brake-screeching exercise in redefinition."

  Michael Jensen [07.17.07 07:59 PM]

So lots of traffic, using ("buying") fewer of the resources. Lots of traffic, but ever-less engagement with the fundamentals of what a library is.

If I were a shareholder, I'd be insisting that the marketing budget be increased, to "monetize" the traffic.

"Monetizing" in this case is getting ideas into brains, and librarian skills helping build good critical thinkers. Thus, not marketing for the library itself, but marketing for the information resources contained within the library (and other libraries).

Free! JSTOR, 100+ years of ideas! Thoughtful answers, nearly instantly! Expertise on demand! Just waiting for you: optimal interfaces for fiction from the last 300 years, with a FREE two-week trial!

  Thomas Lord [07.17.07 11:15 PM]

A triad:

1. The graphs are misleading and almost irrelevant. We want to know the value created by the physical operation of a library and numbers like "reference queries" and "circulation" don't tell us that.

2. Shareholders are irrelvant: extracted profit is inconsistent with the profession's mission.

3. Yes, of course the public face of libraries will trend towards "basement cafes". That's true of a lot of things simply because it is much, much cheaper to move bits than to move atoms. With any luck, this trend will result in more people being (for all practical purposes) closer to a library, on average, than they are today. In this scenario, the net result is an increase, not a decrease, in the amount of total real estate devoted to libraries.


  Allegra Searle-LeBel [07.17.07 11:29 PM]

Some entrepreneurial thoughts about libraries losing users...

User centric design must be included as part of the strategy. What do people want to do at libraries? Give them the tools to do those things. Do the people libraries want to be serving want to play video games, for example? I know many libraries let you check out video games. People get to read books at the library, let them play at the library. Develop the relationship.

Also, library professionals can apply the working knowledge they have about organizing ideas and enjoying information to the online resources that people are turning to now. Make them fun, interactive, meaningful, instead of the flat, dry search engine that many online libraries have now. Put the virtual reality hook in; make them look and navigate like libraries so that when people go into the actual one, they're already familiar with the setting and have performed the activities virtually. People learn how to interact from the activities in which they engage, so if people are going online, and libraries want them to actually come in, they have to teach them what coming in is like and then give them the option of checking out books online if they choose.

Also, libraries could make their members more like shareholders. Ask for information about which movies and books and games and consoles to get. Ask for funding for the gaming console. Provide opportunities for creating community - kids who don't have enough money to buy their own games, but who contribute $5 to a console will be much more likely to use it. Give people something in which to invest, with expected return (I put in $5, I get my friends to put in $5, we all get to play). They'll be there.

  Shahar Nechmad [07.18.07 01:36 AM]

I think the graphs are kind of obvious…
Hey, let’s face it - A lot of the times it’s much easier to find information in Google, than trying to crawl thousands of books in the library.
So what do the libraries need to do in order to save themselves from extension?

1. Go with the flow - It’s easier to deal with electronic data. You can copy paste it, embed it in your paper directly, etc. Instead of keep holding on the old concept of the book, jump into the new reality. Make more books visible online (under a closed network/site).

2. Add better ways to consume information - Have you heard the term Multimedia? Why not upload to the library site webcasts of professors and lessons from the university. Webcasts of students talking about their research, interviews with researches around the world, etc.

3. Make the search experience better - Searching something specific in Google is getting harder and harder. Develop a virtual representation of the library, where important information around the net is cataloged better (from the specific perspective of a student that wants to research university subjects).

4. Make it more social - Allow people to share ideas, comment on books, etc. The concepts of Web 2.0 will fit the library very well.

5. Understand that the purpose of the old bricks library has changed. Its purpose is to help people learn better together and not to hold the vast information of the universe. Concentrate on the things that can help students when they need to meet, study, research and work together.

Trying to keep the library books and information out of the Internet in order to preserve the old concept of the library is like the useless fight the music industry tried to give to digital music. We all saw what happened. The only question is if you can wake up early enough to realize you need to transform yourself and get on with the new digital reality.

  Michael S. Hart [07.18.07 06:20 AM]

While I disagree with the more head in the sand type of responses that have been given, denial of fact in
cases where the fact is not understood nor proven...
I must still argue on both sides of this coin:

1. It is possible that the reason a fewer questions
total is listed in various results, is because those
questions are being answered better. i.e. I used to
wait until the shift change and ask again, hoping an
alternate set of answers or resources might appear--
but perhaps with the new tools at their disposal
the reference librarians are simply giving better
answers and the patrons don't need to ask again.

2. On the other hand, I find that the reference librarians I do ask are simply using the same tools I use already, and since I have studied how
to use search engines in an actual GSLIS course,
I usually know how to use them at least as well,
so I don't ask nearly as often, and probably the
students who grew up online feel the same about asking for help from someone who did not.

3. The idea of commercializing libraries to compete with video games, coffee shops, etc.,
seems to me to be on the same order as the way
they "commercialized" PBS, NPR, the BBC, etc.,
definitely a /BAD/ idea. . .just as a commercial
news program cannot serve both Truth and Mammon
at ths same time, a la Katie Couric as the new
Barbara Walters. . .I strongly suggest you watch
NETWORK [movie] which was inspired by the fact of
Barbara Walters' [the first female news anchor]
salary being split between The News Division and
The Entertainment/Programming Division, creating
what we now call today. . ."Infotainment."
I, personally, do not want to see libraries fall into the "Infotainment" catagory, even though I am
certainly one of those who check out fewer books and ask fewer questions, simply because I can "do it myself."

I am sure the same thing happened to libraries when The Gutenberg Press created more books in the first 50 years of its existence than were printed
in all previous history. . .you just didn't HAVE to go to the library as much because you could buy
your own books. Today there are millions of books
free for the downloading or reading on the Net and
that's just all the more reason that people aren't
going to the libraries or checking out books.

When you can literally buy a terabyte of space for
under $200, and fill it with millions of free books from Project Gutenberg and The Internet Archive, etc., there is obviously less need to go to any particular physical location to read them,
and, as was pointed out earlier, it is MUCH EASIER to search and quote eBooks than paper books.

Hmmm, my "preview" indicates that even if I use 3 hard returns, all my comments are jumbled together
in one paragraphs, please advise how to neaten up.


Michael S. Hart

BTW, take a look at http://www.worldebookfair.com
for a HUGE number of free eBooks available until August 4.

  Michael S. Hart [07.18.07 06:29 AM]

OK, I figured out that the "preview" is not real,
will try to do better next time. . . . Michael

  Nathan [07.18.07 07:08 AM]

My library use is way up now, BECAUSE of the internet. It is so convenient to look up a book on the regional library catalog, and have that book delivered to the nearest library for me to pick up.

I clued my mom into this and now she's a book-reserving fiend. Friends of mine get tons of books-on-cd for their long commute. Almost like Netflix! People have no idea.

WorldCat rocks too, and I use it to see if my hard-to-find book is somewhere reasonable for inter-library loan.

  M A Fields [07.18.07 07:09 AM]

Don't overlook changes in pedagogy that started in grade school and have worked their way up to higher ed. Many assignments used to be "exercises in research" that had the effect of forcing visits to the library and forcing use of certain resources. As teaching/learning change toward more self-directed and discovery learning we can expect a narrowing, but in some cases deepening, of the need for and use of library resources.

  A Librarian [07.18.07 07:53 AM]

In reponse to Shahar Nechmad’s comments:

1. Instead of keep holding on the old concept of the book, jump into the new reality.

>Today’s libraries not only offer numerous (upon numerous) specialty databases in which users can pull full-text articles… Think hidden web here… Most libraries do engage in elibraries with direct links to electronic books available from their online catalogs (yep, most of our catalogs are online).

2. Have you heard the term Multimedia?

>Have you visited your local public or academic library lately?

Besides podcasts and vlogs about library services, story time, author visits, local events, etc. Most libraries today are engaging in some sort of multimedia experience including research tutorials.

3. Make the search experience better - Searching something specific in Google is getting harder and harder.

>Again, most libraries subscribe to numerous specialty databases in which searching is, for the most part, easy. If a student or patron needs medical research, they only need to visit a database.

Also, the majority of libraries offer their databases online meaning users can log into them from home, work, Starbuck’s, etc. Today's libraries have no boundaries.

To add to the pot... Most librarians today engage in instruction. Visit your nearest academic library and you'll find instruction librarians giving classes on how to search not only in their libraries' stacks, databases but the Web (yes, we like Google too).

4. Make it more social - Allow people to share ideas, comment on books, etc. The concepts of Web 2.0 will fit the library very well.

>Oh dear, you really haven’t stepped foot in your local library in ages…

Most libraries and librarians are Web 2.0 crazy.

We’ve got MySpace, Facebook profiles; we’re blogging; we’re making videos for YouTube; we’re creating wikis for interoffice communication; we’re even Ninging-- http://library20.ning.com/

We’ve embraced Web 2.0 wholeheartedly.

5. Understand that the purpose of the old bricks library has changed. Its purpose is to help people learn better together and not to hold the vast information of the universe. Concentrate on the things that can help students when they need to meet, study, research and work together.

>May I suggest going to your local library!
You’ll notice we’re NOT just books, books, and books… Nowadays, we have coffee shops, meeting rooms, study rooms, computer labs, even art galleries!

Most public libraries host community events, gaming competitions…

Most libraries, including my own, offer students the software and the equipment to create and edit their own videos, record music…

We’ve moved (or are moving) past the concept of the “old brick library”.

Go to your local library Shahar.

  Cindy Fuerst [07.18.07 08:13 AM]

I work in a public library, and our usage has gone up over the past ten years. The number of items has only increased slightly, but the gate count, computer usage, and program attendance is three and four times what it was ten years ago. Some of this may be because of a new facility, but some of this is because we have changed our thinking about how to best serve our community. I think all libraries - regardless of type - need to make sure that their services have have real value to the people they are serving.

  Alan Swarm [07.18.07 09:35 AM]

I think a better question would be:
Why Aren't Libraries shareholders in Book and Journal Publishing?
We all know the prices go up 10-15% a year on journals. What better investment return could there be?

  Josh [07.18.07 09:40 AM]

Is there a way to measure the quality of the requests? It seems like all the reference requests that could be handled by wikipedia or google would move there. Also all the book requests would move toward in depth research instead of basic materials.

If so, libraries should serve more depth and less breadth. Moving to "Mysql Performance" and away from "Computer Encyclopedia".

I am a little scared that libraries will move too far in the meeting space direction and actually cull the books that are rarely referenced but have high value for the user. So change the goal to maximize value to users instead of visits or circulation.

Also search is a disaster in libraries, even leaving aside the great utility of full text search. I use Amazon to search the San Francisco public library. Greping through a text file would be more fun than using the interfaces now available. Libraries should release their data into the wild and let people come up with a better way to use it. If current contracts don't allow that, they should get together and fund something that does. Maybe Open Library can help with this.

  Joshua Franklin [07.18.07 09:44 AM]

I am curious what the pre-1995 graphs would look like. I have a hard time believing that very many students were going online (or using DIALOG) for research in the mid-1990s, when the options were things like Mosaic or Netscape 1.0 and not even every university had a website. I would guess that the downward trend is much longer, perhaps coinciding with the entertainment-first MTV Generation rather than technological change.

An unrelated note: publications in many fields moved away from the scholarly monograph a long time prior to 1995.

Oh, and naturally it would be fascinating to see public library numbers since circulation there fills a different role.

  Jonathan [07.18.07 01:20 PM]

My hometown library, the Gail Borden Public Library in Elgin, Illinois, is experiencing a lot of growth in traffic and usage. It's hard for me to believe that libraries are going the way of the dodo.

  Reb Yudel [07.18.07 05:14 PM]

I think there may be two diverging trends. My suburban public library is also circulating more and more books (and videos) each year. I'm sure the Internet is a factor -- I can hear about a book online, place a request online, and then pick it up when the library receives it for only a 25 cent fee. My wife and I will stumble across a reference to an interesting historical character, and then request three biographies. This is a level of recreational research that a suburban library couldn't provide 15 years ago.

The question I would ask the research librarians is: How many of your queries or books circulated were in response to real research curiosity -- and how many to enable students to write a required paper for a required course?

  Ron Peterson [07.19.07 07:19 AM]

I agree with an earlier comment that these statistics really aren't telling us much about the big picture. For ARL libraries anyway, we need to be more focused on outcomes, such as the quality of research and the ability of graduates to use information effectively.

As for the decline in reference requests, while it probably doesn't explain all of the decrease, I think there has been a shift in the definition of "reference question". The reference desk manual at a previous employer defined a "reference question" as any time you needed to "refer" to a source. So essentially anything you couldn't answer off the top of your head. What we were told to record were "questions that took more than a few minutes." If people had recorded questions based on a the manual, they would have higher counts than we did - the statistics did, in fact, show a decline.

I think this is partly influenced by the internet. Walking over to the Readers' Guide seems like more work than pulling up EBSCO. I also recall someone telling me that they didn't count a question as a reference question if they answered it using a website, it was only a reference question if you used licensed library information.

The counts are also influenced by people self-reporting. If people are looking to cut reference staff or if the reference librarians want fewer desk hours, then it is in their interest to show that there is less use.

  Matt [07.19.07 10:13 AM]

I couldn't help but notice that a lot of the comments here are missing the context of the data in the graphs - these trends are taken from *research* libraries (i.e. libraries serving very large research-oriented universities), not public libraries. There is a world of difference in the purpose and audience that research libraries and public libraries serve.

  Renee Reed [07.19.07 01:38 PM]

As a university reference librarian who has been working for 20 years, I seem to remember that the bulk of the reference questions we were asked pre-Google were short answer questions; often requests for addresses and telephone numbers, word definitions--easily accessible for anyone from internet sources now. We don't even purchase many of the directories (in print or online version) that used to be high priority reference titles. I think I see a similar number of questions today as in the past from students who need some direction in beginning their research or in locating a print or online source. We do have a somewhat smaller staff (same number of librarians, no non-librarina staff or student workers)in reference than we did a few years ago as a result of having fewer easy questions to answer. Books may be a differenct story in research libraries--I see many students spending huge blocks of time using e-mail, engaging in cell phone conversations, and playing games or fantasy sports--free time that previously might have been used sometimes for reading books. Students also seem to spend more of their time working part-time and volunteering than students did a few years ago. I agree that when I visit my own public library I see much more variety of activity, including people browsing books. I suspect that the percentage of people who actually read books voluntarily hasn't changed that much over the years, but a larger percentage of the population does attend college now, and as in the past, many will read as little as possible to get by. I just finished looking at a short piece in the July 2007 Wired (p. 58) that presents some data from a Pew Research study that seems to show that Americans on the average are less able to answer questions about current events now than they were in 1989. They also chart percentages of correct answers by the respondents' declared primary news source (major newspapers online, NPR, news blogs, Fox News, etc.). When I look at online news (and TV news), there seems to be much less content there than I would have expected in news sources in the past. Unfortunately, some newspapers seem to be following that trend now as well. Mabye university book collections will decline and shrink also, but I hope not.

  Edward [07.20.07 04:11 AM]

Another possibility to account for changes in the number of reference questions is the increase of bibliographic instruction sessions. While I haven't worked in an ARL library, in the academic libraries I have worked in are doing a greater amount of uinstruction. This leads to less "basic" reference questions about which resource should be used, and how to use them.

  Anonymous [07.20.07 12:54 PM]

I'd rather buy a book off Amazon than go borrow one from the library. We need telecommuting libraries.

  Katherine Kott [07.20.07 01:15 PM]

Some people think gate counts are up while circulation is down because the library is the best place to find and use traditional print material that is still important in some fields synchronously with digital content. The hypothesis would be that as libraries have allocated space for information commons by adding computers and enabling wireless access, people have become more likely to capture snippets from print books and journals in the library than to haul the tome across campus to a dorm room that is probably being used for other purposes than studying.

ARL libraries don't count or report in-house use of traditional library materials (e.g. scanning bar codes for material that's been left by the photocopier or on study tables) so the hypothesis would be hard to test against the ARL circulation statistics that are gathered. It would be interesting to look at the circulation statistics juxtaposed with information about numbers of computers and wireless access points but ARL statistics have only tracked expenditure levels for hardware and software since 2004. We could use some additional facts to get a clearer picture of what is happening now to be thoughtful about the contributions libraries have to make in academic environments. Public libraries, as others have said are another story.

  Another Librarian [07.20.07 01:23 PM]

I'm entering the discussion late, but feel compelled to comment on the questions that ARL asks. I became a librarian in 1996 and worked in a very large academic research library until June 1 when I moved to accept a new position. During that time, I watched a thriving reference desk downsize considerably because of the number and type of questions that were being asked. We could see the change. As a business subject specialist, I knew that I had to find new ways to reach faculty and students. I developed a long range plan to market my services. Over the last 6 years, the number of reference questions that I answered increased 500% as did the number of instructional classes and students in attendance. The majority of the questions were via email, but I also met with many more students in my office. Over 25 hours a week were devoted to answering student and faculty questions. I marketed my services, attempting to insert myself in all undergraduate business courses with research components, several graduate courses and all orientation sessions. The result was a substantial increase in library visits, through email, web site and course page hits, IM, etc. I consistently surveyed faculty and students asking about their needs. ARL conttinues to be more interested in reference inquiries at staffed physical library reference points, gate counts and physical book checkouts than email and IM questions, one-on-one and group consults. We know through numerous LibQual reports that faculty and students want to be self-sufficient, but if they know that help is a friendly click away, they'll ask for help. I can guarantee that over the last 10 years I saw an exponential increase in "library" use, even though that library may have been virtual. ARL needs to be asking the right kinds of questions if they want the whole story.

  Leo Klein [07.21.07 10:41 PM]

"Some of that can certainly be accounted for by the use of electronic materials that don't count in ARL libraries circulation figures"

Well, duh! If you're not including material accessed online through research databases, you're liable to under-count the circ figures in a phenomenally significant way.

I think one of the major problems in assessing where we stand is making sure our metrics are actually up to the task.

  Amos Lakos [07.23.07 10:33 AM]

These trends have been knows for some time and libraries are trying to "go with the flow" - however, the flow is much faster than the going.

As far as research library (ARL counts) are concerned, I agree with Leo - it took us years to do come up with some measurable measures - but the ability of many libraries to count them and analyze them is still fragmented.

The "real' issue for me is that libraries need to think beyond the local - both in terms of collections and service delivery. When we do that, efficiencies of scale will kick in and the cost benefit per user may be more encouraging.

  Steve Cisler [07.23.07 05:56 PM]

Recently, SOLINET, a library networking service, solicited input on three scenarios for the future of libraries. You can read the summary here:


I think non-librarians and librarians will find it quite interesting.

  Douglas Galbi [07.23.07 05:59 PM]

For public libraries, book circulation per user is down about 50% from 1977, but audiovisuals circulation now accounts for about 25% of total circulation. Including audiovisuals, total circulation per user is now on the low side historically, but there's been little over-all trend since the mid-nineteenth century. See
purplemotes.net .

  DaveShumaker [07.28.07 01:35 PM]


These libraries may be providing "embedded" instructional and informational services. If so, they may find that when adopting this new service model, the number of beans (i.e. reference questions) to count drops, but the richness of collaboration among librarians, faculty, and students goes up and the value of librarians' contributions goes up. "Another Librarian" seems to be describing something similar above.

  Larry Schwartz [02.22.08 11:10 AM]

I found this article by trawling through my inbox archive, and here's my question: can someone plug in "average class size" as a variable? did the number of students that professors are expected to teach in a class session increase and does that increase correlate to a decline in reference questions and library visits?

Post A Comment:

 (please be patient, comments may take awhile to post)

Type the characters you see in the picture above.