In the spring of 2016 I was enrolled in the University of Missouri’s School of Information Science & Learning Technology’s (SISLT) Online Education certificate program. I was only in the program for that one semester before taking a new job in a new state that made it financially impossible to continue with that certificate program since I was no longer getting the 75% employee discount on in-state tuition. One of the classes I was taking was called “Digital Humanities” (DH). This class introduced me to that field of study and had me thinking about switching enrollments into the DH certificate program they were just about to launch. That class required a final project and ultimately resulted in the largest and most professional product I ever created for SISLT, and this is after I’d graduated with their Master’s in Library Science degree in May 2015.
What follows is a version of that final project edited for this format.
While working in consortial lending for the University of Missouri St. Louis (UMSL) I was on the front line in the decline of monograph circulation in academic libraries. Not surprisingly, consortial usage of monographs have declined along with traditional monograph circulation. To what extent has monograph circulation declined in the consortial lending practices at UMSL? What might be the cause of this decline? What I propose to do in this project is show how drastically usage has declined over the years, even with the addition of consortium members and cooperative agreements with other consortia, using a selection of infographics and examine what may be the cause of this decline.
Key patterns in monograph usage can be reflected and culled by examining consortial usage statistics. This case study is based on the UMSL experience with the MOBIUS consortium and courier services. This is a snapshot of a larger and more complex system that interacts with other consortia.
Traditionally, lending has been a type of yardstick used by libraries to judge their effectiveness and relevance. Much like the number of volumes a library holds, this yardstick may no longer be effective or particularly meaningful in measuring a particular library’s value. The broader ramifications of such a decline in consortial lending may call into question the very need or purpose of library consortia.
This study is restricted to the perspective of UMSL and its relationship to the MOBIUS, Mid-America Library Alliance (MALA), Amigos, and Colorado Alliance (i.e. Alliance) consortia. However, the author feels that this perspective may be relevant to and representative of national trends in consortial lending broadly. This study is confined to the official statistics provided by the MOBIUS consortium and the data gathered by myself and my student assistants at UMSL. While total circulation numbers and the number of full-time equivalency (FTE) students or public library users at the various connected institutions are relevant to this problem, those factors will not be considered in this study.
Founded in 1998, MOBIUS is a 501 (c)(3) organization that serves seventy-five academic, public, and special libraries. MOBIUS contracts with Stat Courier to be transport requested items. The courier connects the MOBIUS system with non-MOBIUS member libraries and two other courier systems. Primarily, this arrangement allows MOBIUS libraries to easily transport our items to the Mid-America Library Alliance (MALA) libraries which give us access to most Missouri public libraries. Since summer 2014, they also have a working agreement among MOBIUS, our courier, the Amigos consortium, and the Colorado Alliance (i.e. Alliance) consortium.
When I took over the consortia lending duties for UMSL in 2010 there was no reliable way to track usage and courier shipments available to me. This forced me to devise my own system beginning in 2011. During the years 2011-2014 the data was maintained in a series of simple Microsoft Excel files containing fields for date, destination, bag number (i.e. container bar code), item barcode, and a notes field for issues of ambiguity. Frustrated that, given my level of familiarity with Excel, it was still more difficult to keep reliable statistics in this method than I would have liked. I decided to adapt the multiple Excel files into single Microsoft Access file to give me greater ability to track usage. This also allowed me to standardize nomenclature and formatting.
The Access database contains two tables. One is a master list of all libraries served by the system of couriers which the various consortia service. The second table is a recreation of the previously established Excel files. Data from 2011-2014 has been integrated into the Access database as new information is added daily as part of the outgoing shipment processing procedures.
Daily, the number of containers used per location is collected. Monthly, data is compiled on the total number of items sent through the courier service per location. Also monthly, data is collected on usage within the MOBIUS consortium, specifically. This data is compiled by the MOBIUS consortium and collected by the consortial lending supervisor for use in an annual report to library superiors. Those superiors are strictly concerned with the total MOBIUS usage numbers. The rest of the statistics are used to determine how many shipping labels to print for a given library and if a given library uses the system enough to justify their own sorting shelf in my work space. The data also serves as a check on processes in regards to patrons’ return claims, or in-transit-too-long issues between UMSL and its partner libraries. Since contracting with the Stat courier in 2016, they can also use the bag numbers to track specific shipments.
Performing a literature review for this project has been rather difficult. In the resources available to me I can find very little literature to support the idea that monograph circulation has been declining over recent years, much less the causes for it. O’Neil’s and Gammon’s “Consortial Book Circulation Patterns: The OCLC-OhioLINK Study” (2014) had a promising title, but the study was of monograph usage over a single calendar year and measured frequency of use, duplication levels, obsolescence rates, etc., but did not address monograph circulation over a number of years. Cheung’s and Chung’s “Monograph Circulation Over a 15-Year Period in a Liberal Arts University” (2011) again held a promising title, but this article concentrated on usage from a collection development standpoint instead of access services, interlibrary loan, or cosortial lending. They conclude, among other things, that not all books acquired by a library circulate within the first fifteen years of acquisition and that books as young as five years may be considered for removal to a remote storage site. Table 4 in this article does seem to reinforce the notion of declining circulation however showing the decline of volumes with checkouts ranging generally downward between 1995 (21,338 volumes) and 2009 (12,366 volumes), but volumes with checkouts is not the same as total checkouts over several years. In 2016, Rose-Wiles and Irwin take as a matter of course a decade of declining monograph circulation. This study, again with a collection development bent, attempts to revive the practice of measuring in-house usage of books to determine their value to the collection, as well as their circulation. The same Rose-Wiles wrote in 2013 that Seton Hall University’s circulation rate dropped 23% between 2005 and 2009. This is in line with what my data shows and the anecdotal feeling around academic libraries. Rodriguez-Bravo and Rodriguez-Sedano (2016) looks at what library materials are being used and by whom over the course of two academic years, but does not address total circulation numbers of monographs. The single best source I’ve been able to find in the published literature is Martell’s “The Absent User: Physical Use of Academic Library Collections and Services Continues to Decline 1995-2006” (2008). Here, Martell sites other sources that show circulation at ARL and Ivy League libraries, ranging from an increase of 2% to a decrease of 58% depending on the type of library (Martell, Tables 1-2). His data comes from a variety of earlier sources, however and I have not been able to locate a similar study with more recent numbers. Moving from a database search to a web search, in 2012 the ACRL Tech Connect blog published “The End of Academic Library Circulation?” (Kurt) which posits that “by 2020 university libraries will no longer have circulation desks.” Kurt describes the drop in circulation as a function of the rise of e-books, yes, but more so in the change of library user behavior. Primarily in a graph titled “Circulation/User – PhD Granting Universities” which shows trend lines for average number of books checked out by patrons declining between 1995-2020 to hit zero on, or near, the year 2020. Library Journal published “Print on the Margins: Circulation Trends in Major Research Libraries” in June, 2011 (Anderson). Here, they stress the need to take into account FTE enrollment changes as one looks at total circulation rates. This typically means that the decline in circulation is even worse than a simple line graph can demonstrate. This paper will not consider FTE enrollments in its data.
Clearly, there is a significant hole in the research on these matters. None of the research I collected considered interlibrary or consortial lending practices as major features of their research. Universities were either measured individually, or in larger groups, but no serious consideration of either interlibrary or consortial lending usage trends were found.
From January 2011 through December 2014 all data was collected into a series of Microsoft Excel files, each with a separate sheet for every month. Prior to that time, the only tracking that the University of Missouri St. Louis’ Thomas Jefferson Library performed consisted of filling out by hand a paper table on which was written the date, the destination, the bag number, and the number of items contained in the bag. This placed us in a position in which we could show that a number of items were sent to a particular location on a particular day, but we had no way to determine which items were sent. The January 2011 creation of the Excel files created a simple, efficient, and persistent way for us to keep better track of what we were shipping, and create better accountability for all parties involved.
Over the years this system served us well enough, however given my degree of familiarity with Excel, I was frustrated that the only way to cull statistics from this method was little better than hand counting every entry. Try as I might, I would still make errors in judging how much we sent to any given location at the end of a month or a year. Ultimately, I decided that I had to come up with a better and more reliable solution to this problem if I was going to be able to accurately account for MOBIUS and courier usage for Thomas Jefferson Library.
In the summer of 2014 I took it upon myself to learn Microsoft Access and created a database file to solve all of my needs. The database is made up of two tables. The first, entitled “Libraries Master List,” is a compiled list of every MOBIUS member library and every other library that our courier (then 1st Choice, now Stat) serves, plus the libraries served by the Trans-Amigos Express courier which services Amigos Libraries, and the COKAMO courier for the Colorado Alliance consortium. This table consists of fields for the library name, city, state, OCLC symbol, consortium symbol, Stat courier symbol, consortium name, and cluster name. A helpful byproduct of this table is that now there is a clear list of locations for which our traditional interlibrary loan services can utilize to determine if an item needs to be mailed or simply added to the courier pick-up which saves us a great deal in postage. As of May 2017, this table had 846 records to it. Of those 846 records, 258 locations had been utilized since January 2011.
Once this table was completed, I was able to create the second table, called simply “Outgoing,” which would contain the actual tracking data. This table mirrored the original Excel files providing fields for packing date, destination symbol (i.e. consortium symbol), bag number, barcode number, and a notes field to clear up any ambiguities. Of these fields only the notes field does not require data. The Symbol field in this table was tied to the Consortium Symbol field in the Libraries Master List table creating a standardized abbreviation for each location.
Beginning in January 2015, this database became the standard for tracking outgoing courier shipments for the library. It was not until the Spring of 2017 that I decided to back fill the database with the 2011-2014 data. After several weeks of laboriously copying and pasting data, and correcting formatting errors I finally completed the conversion in early April 2017. The database now contains more than six years of shipment data in a stable and sortable database for the culling of quantifications resulting in 99,005 records, to date of this writing. For the purposes of this project we will only consider the calendar years of 2011-2016 as they are the completed datasets equaling 96,213 entries.
In terms of creating the visuals to illustrate my data, I have tried a few different options Microsoft Excel provides chart and graph capabilities, which are functional, but don’t necessarily make for engaging presentation. I use those in the paper below. I made an effort to use Tableau Public, which, while provided better aesthetics than Excel, did not allow for the flexibility I needed in its free version. Ultimately, I’ve settled on Vengage.com for creating my visuals. This product offers the creative and technical features I require in a free platform. Vengage.com’s biggest handicap, however, is that in the free version the graphics are not downloadable, but are available to the public on the web.
Starting from the widest possible view, we’ll begin by looking at the total number of items shipped from UMSL to our lending partners via our courier system, called C-Total. Right away we see a dramatic drop off in usage (Figure 1). In 2011, we shipped 22,018 items, versus in 2016 wherein we shipped 11,219 items. This is a reduction of 50.95%, or 10,799 individual items. In this time period the greatest reduction of usage happened between 2012 and 2013 which saw a reduction of 16.54%, or 3,277 items. The smallest reduction happened between 2015 and 2016 which saw a reduction of 8.51%, or 1,043 items.
Those numbers include all Mobius requests, along with all courier service to MALA, Amigos, and Alliance destinations. But what if we narrowed our focus to just the Mobius member libraries (M-Total)? Not surprisingly, this data (Figure 2) approximates the C-Total data. The difference between 2011 and 2016 is 9,730, or a loss of 52.47%. The largest difference between them is again between 2012 and 2013 showing a drop off of 15.17%, or 2,809 items. The period with the lowest loss of usage this time between 2011 and 2012 with only a 9.57% loss, or 1,959 items.
Because this dataset comes directly from the Mobius Consortium Office, and they keep records differently than I do at UMSL, we can also look at the total numbers for borrowing vs. lending. The first thing that jumps out at anyone that sees this data is that UMSL has always been a net-lender. We always lend more items than our patrons request. Applying the same measures, we see that the difference in total number of items lent from UMSL to our Mobius partners between 2011 and 2016 is 4,850 items, or 56.84%. The greatest period of loss in lending request was between 2015 and 2016, with a loss of 14.84%, or 1,113 items. The least amount of loss was between 2011 and 2012 with a loss of 6.10%, or 686 items. In the borrowing statistics, the difference between 2011 and 2016 is 4,880, or 47.14%. The greatest loss was between 2012 and 2013 in which we lost 20.64% in borrowing requests, or 1,643 requests. The period of least loss was between 2013 and 2014 in which we lost only 9.31%, or 588 requests.
We can break this data out into a monthly representation where we see the trends from year to year (Figure 3). This chart (M-Monthly) represents the monthly usage of strictly Mobius libraries over the six years. We can see that our highest usage came in January 2012 with 2,309 fulfilled requests, whereas our lowest usage was December 2016 with only 585 fulfilled requests. Over the course of the six years, the mean number of fulfilled requests is 1,272, with a median number of 1,169. In 2011, the highest usage was in March, with 2,173 fulfilled requests. Lowest was in December, with 951 requests. In 2012, the month with highest usage was January, with 2,309 fulfilled requests. The lowest was December with 784. In 2013, the month with highest usage was April, with 1,702 fulfilled requests. The lowest was December with 805. In 2014, the month with highest usage was January, with 1,469 fulfilled requests. The lowest was July with 793. In 2015, the month with highest usage was September, with 1,301 fulfilled requests. The lowest was December with 671. Finally, in 2016, the month with highest usage was February, with 1,174 fulfilled requests. The lowest was December with 585. Through the measured time period (Figure 4) the month with the largest average usage is January, with 1,673 fulfilled requests, while December is typically the least used month with only 767 fulfilled requests.
Going back to the larger set of total courier usage we can get a picture of who has used the system the most over time (Figure 5). Of the total 96,213 items in our sample size it is unsurprising that our home Mobius consortium makes up the bulk of the usage with 92,176 total items, or 95.80% of the usage. Rounding out the totals, MALA used the system 3.45%. Amigos used the system 0.39%, and Alliance used it 0.36% of the total usage. One explanation for the scarcity of Amigos and Alliance usages is that neither system was partnered with Mobius or our couriers for the full six years. Amigos was first integrated in July 2014, while Alliance was integrated in August 2014.
We can break this up further accounting for the various clusters (Figure 6) that make up the Mobius consortium. A cluster is a smaller service area within Mobius that is allowed to set cluster-specific rules within the Mobius organizational framework. These clusters tend to represent universities with similar organization structures or patron bases. For instance, UMSL is a member of the Merlin Cluster, which is made up of the libraries across the four University of Missouri campuses. The Archway cluster is a group of primarily community colleges that serve the St. Louis area, etc.
You can see here that the Merlin cluster makes up more than half of the usage across the time frame, with 12.41% of usage made up of clusters that make up less than 2% of the total courier traffic moving through UMSL. If I were to break the Merlin cluster down within this chart you would see that the University of Missouri – Columbia (i.e. Mizzou or MU) provides approximately 50% of that 54.39%.
These numbers are all well and good, but what do they tell us about the drop off in courier usage over the years? To test this question I created an Excel table that displayed the relative percentage of cluster usage in percentages of the year’s total across all six years (Table 1). I then took a mean of each of those percentages for each cluster and performed a standard deviation of the percentages. The result was that there is almost no measureable deviation in the percentage of the total usage from year to year. This tells me that no one cluster or location is responsible for the loss of usage, but that the drop off is steady across the board.
Analysis & Conclusion
The data suggests that there is an irreversible decline in monograph usage through consortial lending. One of the few aspects of my findings that is reassuring is that the curve in the decline is shallowing (Figure 1). Usage is declining less over time. In part, this is due to the ever-increasing cooperative agreements and membership which the Mobius consortium has negotiated since 2014. Comparing the relative usage numbers between 2011 and 2016 the greatest area of stability providing growth is the inclusion of more public libraries, specifically Springfield-Greene County [Missouri] Library (SGCL), Tulsa [Oklahoma] City-County Library (TCCL), and St. Charles City-County [Missouri] Library (SCCL). However, this cannot provide sufficient usage to stem the ebbing tide in overall usage.
Looking at the M-Monthly graph I also find something reassuring in the flattening of the yearly curves. It tells me that while usage will continue to decline that there is a basement to the decline somewhere above zero. I don’t have the statistical skills to predict this basement, but I am hopeful that we are nearing it.
If I may speculate, and given the lack of published material on the matter I must, I see the decline in usage to have a three-fold cause. One is that circulation desks are seeing dwindling usage numbers across academia. Two is that the rise of e-book licenses purchased by universities and consortia are increasing exponentially, making the traffic of physical monographs less and less necessary. And thirdly, that the ever-growing number of libraries available in the Mobius consortium are spreading out the availability of titles. More items in more libraries would naturally create a circumstance in which any particular library’s individual holdings are less important than before.
Let it not be said that consortial lending, or even monograph lending more broadly, is dead. There is an unarguable reduction of monograph circulation, certainly. However, I find it hard to believe that Kurt’s prediction that by 2020 no academic library will not need a circulation desk. This sounds to me like the 1990’s prediction that in the future offices will not use paper. It is now 2017 and I still print pages daily, and I bet you do, too. Predictions like this I believe are short-sighted and inevitably erroneous. All doomsday predictions have one thing in common – they’ve all been wrong. The end of the world has been predicted many times and we’re all still here. The light bulb did not make fire obsolete, but simply made it a niche tool for specific purposes. This is the future I see for monographs. There will always be a need or desire for printed material through which one can flip and peruse. However, they will become specialized items for specific user groups or keepsake nostalgic items. The future of a position like mine is in much doubt, though. In the very near future no library will need a person dedicated to nothing more than consortial lending. In fact, at this moment, I am the only person I know of in the Mobius system whose job is solely to manage consortial lending for any institution. Ideally this job duty should be folded in with ILL-lending procedures. When I have moved on to the next stage of my career there will not be another person hired to fulfill my duties, but those duties will be spread out over the rest of my department, and that is the right thing to do. Consortial lending is very important to the life of the library at UMSL and Mobius. However, in the perspective of total monograph circulation trends it is declining at a similar rate. The practice must be maintained for as long as there are cooperative agreements between libraries and patrons who want distant books. But it is foolish optimism to expect that current rates will rise or flatten any time soon.
Anderson, R. (2011). Print on the margins: circulation trends in major research libraries. Library Journal. Retrieved from http://lj.libraryjournal.com/2011/06/academic-libraries/print-on-the-margins-circulation-trends-in-major-research-libraries/
Cheung, S. and Chung, T. (2011) Monograph circulation over a 15-year period in a liberal arts university. Library Management 32(6/7), 419-434/
Kurt, W. (2012) The end of academic library circulation? ACRL Tech Connect. Retrieved from http://acrl.ala.org/techconnect/post/the-end-of-academic-library-circulation
Martell, C. (2008). The absent user: physical use of academic library collections and services continues to decline 1995-2006. Journal of Academic Librarianship, 34 (5), 400-407.
MOBIUS. (2011-2012). MOBIUS lending and borrowing statistics [FY 2011-2012 Monthly & Year-to-Date Statistics Final]. Retrieved from https://mobiusconsortium.org/mobius-lending-borrowing
MOBIUS. (2012-2013). MOBIUS lending and borrowing statistics [FY 2012-2013 Monthly & Year-to-Date Statistics Final]. Retrieved from https://mobiusconsortium.org/mobius-lending-borrowing
MOBIUS. (2013-2014). MOBIUS lending and borrowing statistics [FY 2013-2014 Monthly & Year-to-Date Statistics Final]. Retrieved from https://mobiusconsortium.org/mobius-lending-borrowing
MOBIUS. (2014-2015). MOBIUS lending and borrowing statistics [FY 2014-2015 Monthly & Year-to-Date Statistics Final]. Retrieved from https://mobiusconsortium.org/mobius-lending-borrowing
MOBIUS. (2015-2016). MOBIUS lending and borrowing statistics [FY 2015-2016 Monthly & Year-to-Date Statistics Final]. Retrieved from https://mobiusconsortium.org/mobius-lending-borrowing
MOBIUS. (2016-2017). MOBIUS lending and borrowing statistics [FY 2016-2017 Monthly & Year-to-Date Statistics – Ongoing]. Retrieved from https://mobiusconsortium.org/mobius-lending-borrowing
MOBIUS. (2017). Mission & vision. Retrieved from https://mobiusconsortium.org/about-mobius
O’Neill, E. T. and Gammon, J. A. (2014) Consortial book circulation patterns: the OCLC-ohioLINK study. College & Research Libraries 75(6), 791-807.
Rodriguez-Bravo, B. and Rodriguez-Sedano, F. (2016) Trends in library collection circulation in spanish universities. Library Resources & Technical Services 60(4), 248-258.
Rose-Wiles, L. M. (2013) Are print books dead? an investigation of book circulation at a mid-sized academic library. Technical Services Quarterly 30(2), 129-152.
Rose-Wiles, L. M. and Irwin, J. P. (2016) An old horse revived?: in-house use of print books at seton hall university. Journal of Academic Librarianship 42(3), 207-214.