Keywords

Demand-driven acquisitions; research; consortia; library collaboration; school libraries; special libraries; subject analysis; interlibrary loan

10.1 Specialized Libraries

One early example of research into demand-driven purchasing was conducted across three health sciences libraries and used previous interlibrary loan and acquisitions data to determine areas of collection need and increase funding for those subject areas. Request analysis revealed that over half of the items that had been borrowed via interlibrary loan represented current materials in the core disciplines of the institution. By charting the disciplines of purchase and interlibrary loan, they could identify areas that needed purchasing and areas that were being overpurchased with minimal effort (Byrd, Thomas, & Hughes, 1982). Another early example is the Boulder National Oceanic and Atmospheric Administration Environmental Research Laboratories (NOAA/ERL), which used deep discipline analysis to predict acquisitions strategy directions. The NOAA/ERL libraries, which serve over 500 scientists affiliated with NOAA, the National Bureau of Standards, and the Office of Telecommunications, began keeping computerized catalog and circulation records in 1976. They calculated inventories and circulations for particular disciplines within their subject areas and created an algorithm that determined whether discipline collections were sufficient or if the library should purchase more titles in particular disciplines (Wenger, Sweet, & Stiles, 1979). The strategy of discipline analysis works particularly well for specialized libraries because their greater focus allows for subsubject analysis which reveals intradisciplinary trends.

Many special libraries also serve diverse groups. The Fred Hutchinson Cancer Research Center Library, for example, is a medical library on a research institute, so their collection needs are very unusual. Their patrons include researchers, medical students, and patients. They have a core collection of medical titles that they purchase, but develop the rest of the collection based on patron suggestions via interlibrary loan or other methods (Clark, Levine, & Shane, 2013). DDA can also be used to balance competing collection management demands. The TU Delft Library in the Netherlands is a primary technical library that has been collecting digital books since 2003. Because they serve students’ immediate research needs, but also function as a leading technical library in the Netherlands, it is important for them to maintain a collection that serves both long-term preservation and short-term use. Their solution to this problem is a combination of catalog-integrated, mediated ebook DDA along with traditional collection development to make sure the collection is balanced (Brinkman Dzwig, 2013). Often specialized libraries serve diverse needs and groups and a hybrid strategy which incorporates a DDA program can be a good solution.

This type of strategy was also employed by the William Smith Morton Library at Union Presbyterian Seminary. They began a PDA project in 2010, but it has been so effective that the library has continually extended it. They define PDA as a point of need acquisitions strategy, in which all steps are selected by the library (vendor, book, etc.) but the final purchase decision is made by the patron. They loaded around 1000 records into the system initially, around 75% electronic and 25% print. The only issues they have experienced with this model is the lack of theological ebooks available and the fact that acquiring a more digital collection decreases physical use of the library (Deeds, 2014).

Evidence-based selection (EBS) can also be a good strategy for specialized libraries because it is easier for more focused collections to choose publishers and collections that are a good fit for their institutions. Librarians at Hong Kong University of Science and Technology Library used two levels of analysis to inform their purchasing. They began with a catalog-integrated ebook DDA project in 2014 with a 22,000 record discovery pool. They analyzed the data from this DDA pilot and determined that Wiley was the most frequently used publisher, so they began an EBS pilot using Wiley’s Usage Based Collection Management, which gave the University access to Wiley’s entire catalog of books with a set pledge for patron-initiated purchases. After the EBS pilot was over, the University purchased the most frequently used titles with the pledge funds. Based on the success of these programs, the University also embarked on a consortial DDA effort with JSTOR. Their deliberate, but quick expansion in ebook purchase options led to a sharp increase in their ebook purchasing. They now boast an ebook collection that represents 31% of their total monograph collection (Kwok, Chan, Cheung, & Wong, 2014).

When assessing DDA programs from specialized libraries, intradisciplinary analysis is likely going to be more important than interdisciplinary content. Determining who the stakeholders are for a collection and whether the goals of the collection focus on preservation, immediate research access, or a combination of the two is essential when creating new DDA or EBS strategies or assessing existing programs.

10.2 Consortia

The research on consortia reflect that scholarly publishers seem willing to work to develop DDA programs for the unique needs of different groups of institutions whether they represent separate university libraries within the same system, groups of universities sharing costs and materials, or state or regional institutions of many types in collaboration. These DDA agreements are on a larger scale, so often vendors will build in cost multipliers for consortial licensing, but this is subject to negotiations between the consortium and the publisher or vendor. Some programs even feature different licensing terms for consortial funds and individual library funds like the “Act Together Spend Separately” initiative in Massachusetts. In this model the consortium’s shared funds are used for short-term loans (STLs) across all institutions while the individual library funds are used only towards purchases (Machovec, 2013).

One first large consortia to set up an ebook DDA pilot was the Orbis Cascade Alliance (full-time equivalency (FTE): 235,000), a consortium of 37 academic libraries in the Pacific Northwest. The consortium developed a task force in 2009 to help establish a method for the purchase of ebooks that would be accessible to all member libraries. They began a DDA program that ensured that all the member libraries would contribute funds and all could access the resulting discovery records and purchased titles. Buy-in to the program was based solely on FTE for the pilot with the expectation that if this moved to a permanent program, other factors would be considered. They began with a small pool of 1700 discovery records because they could not find many publishers willing to work with them in 2011. Initially they had set up their program for 10 STLs before purchase, but this was lowered to five after the initial assessment revealed that no titles had been purchased in the first 3 months of the program (Doyle & Tucker, 2011).

Orbis did a lot of assessment and budget-based tweaking of their STLs over time. In 2012, as they reached their maximum budget, they increased the STLs from 5 to 8 and then again to 10. Another budget crisis in 2013 caused the consortium to raise the STLs to 15, then in 2014 they back-purchased materials that had 12 or more STLs for 6 member libraries (Anderson & Seely, 2014). The Orbis Cascade Alliance is a good example of a longstanding program that has dealt with the complexities of DDA in a consortial setting. Their continuing struggles with budget forecasting and attention to STLs as a budget management strategy are model practices for larger consortia hoping to begin DDA programs.

Other early consortia adopters of DDA programs include large university systems with many campuses that act together to provide DDA to students, even if they manage separate catalogs. The University of California system, which serves 23 distinct campuses and over 427,000 students, initiated a PDA program in 2011. Ebook sharing is often difficult in consortia as licensing rules usually restrict the sharing of materials between institutions. The University of California wanted to ensure that, even though they were operating as a single unit for purchasing, that they could assess usage by campus. They used a specialized EZProxy stanza to trigger a dropdown menu that users would see when they clicked links for books. This dropdown allowed users to identify their campus and would record the selection for later assessment (Shepherd & Langston, 2013). Another large university system with an early DDA trial was the Ontario Council of University Libraries, a system made up of 21 campuses that serve 420,000 faculty and students. One of the barriers that this consortium experienced was difficulty in communicating between librarians in order to solidify title lists and troubleshoot problems. They found many benefits to the program as well, including consistent participation from all their member libraries and volume purchase discounts through publishers (Davis, Lei, Neely, & Rykse, 2012).

Arizona State University, a six-campus system that serves an FTE of 128,000, developed a University-wide DDA program in 2012 in response to booming enrollment and declining print circulations. They also participate in the Arizona Universities Library Consortium’s DDA program. As a DDA participant in two very different consortia, they outlined some of the challenges institutions in these agreements face, including developing fair cost-sharing policies between institutions of different types and blurring lines between different modes of access (Richardson, 2013). The University of California system is also juggling different DDA programs, but this time they represent locally held DDA pilots in individual libraries and a collaborative DDA program that serves all 10 physical campuses and their digital library. They have found several benefits to the relationship between the two types of DDA in their system. The University of California’s primary concern in starting a consortial DDA program was to increase diversity by purchasing general interest titles across systems, so individual libraries could spend more money building specialized collections. Their buying power as a consortium has allowed the system to negotiate with vendors for their ebook purchasing principles, which include simultaneous unlimited users, interlibrary loan, and reasonable costs. Their individual programs benefit from this power, but also serve as scouts to help model budget expenditures and buying patterns for their much bulkier consortium program (Scott, Dooley, & Hruska, 2014). Small pilots within consortium or two consortial DDA programs can help institutions benchmark and experiment with different options for DDA.

The University of Colorado system also used a scout pilot to help lay the foundation for their DDA program, but they used discipline to create the subprofile. The University of Colorado system represents three different universities and four different campuses with five libraries that are all separately managed and serve different populations and have different staffing arrangements. They keep separate catalogs but also maintain a shared catalog system. The pilot began with five disciplines: religious studies, business, chemistry, women’s studies, and ethnic studies. They used this initial group of disciplines to outline practices and then added other disciplines as workflows became formalized (Wen-ying & Chambers, 2013).

The above examples represent catalog-integrated DDA programs, one of the most common styles for consortia, but some large university groups have also experimented with physical book DDA through interlibrary loan-to-purchase programs. Ten libraries in the State University of New York (SUNY) system began a coordinated effort to diversify collections using interlibrary loan-to-purchase DDA in 2010. The participating libraries were diverse, representing health sciences, research universities, comprehensive college libraries, and community college libraries. There was an existing consortial agreement that SUNY libraries borrow and lend to one another free of charge. This program formalized the process for when materials were requested via interlibrary loan and not owned in any SUNY library. The requesting institution would purchase the material from a collective fund that had been raised by all the participating institutions. Items had to be physical copies to facilitate lending, under $300, and published within the past 5 years. The program reduced duplication and increased the diversity of the shared collections. The program was considered a success, though some institutions had more active users so purchasing was slightly skewed between institutions (Booth & O’Brien, 2011). This is a good example of a consortium augmenting an established sharing agreement with a DDA component. Programs like this skirt some of the licensing and coordination challenges of digital DDA programs while providing a benefit to collection diversity.

DDA is also a beneficial model for groups of universities and consortia that are made up of many different types of libraries. Florida State University and the University of Florida collaborated on a DDA program aimed specifically at graduate student researchers. The law libraries on both campuses also got in on the action, each contributing $20,000 and setting up profiles that looked only at K class items. There were struggles with licensing and duplication in the catalog and eventually they found that centralized record loading was necessary to keep materials organized across libraries (Carrico, Shelton, & Ziegler, 2013). The Tri-College Consortium (Bryn Mawr, Haverford, and Swathmore Colleges) has shared resources and a union catalog since 1986, so the transition to shared DDA was smooth. They embarked on a DDA program which relied on five STLs before purchase was triggered for budget maintenance (Hartman-Caverly, McColl, Medeiros, & Persick, 2014).

STLs were also influential in vendor selection for ConnectNY, an 18-member resource sharing consortium in New York. The consortium began their ebook DDA with a contribution from each institution of 1% of their annual materials budget (a total of $248,000). This let institutions of many budgets and sizes participate fully in the shared resources. Projected return on investment for this project was calculated inclusively, considering factors like the benefit of introducing a new workflow to the member institutions as well as access benefits. To extend access to all members of the consortia, a multiplier was used to calculate the cost and the consortium was able to negotiate the multiplier from eight “copies” to three. They found that different institutions used the program at different rates. They calculated the value for each institution based on FTE and usage. Some of the members had a value below their contribution, but when they broke down individual institution usage by book variety, they found that all but one of the institutions had a much higher usage value than they would have been able to purchase in print or individually. One of the major benefits to this model is that the committee ironed out the workflow with the vendors and passed that knowledge around to all the individual institution libraries rather than each library figuring this out for themselves (Machovec, Harloe, Hults, & Traub, 2015). This workflow saves the time and headache for individual consortia members who are interested in setting up DDA programs but might not have the staff to be able to do this on their own.

Southern Illinois University Carbondale represents the perspective of a consortium member library that participated in a DDA program through Greater Western Library Alliance. One of the major benefits for Southern Illinois University Carbondale is that they saw increased circulations and a preference for ebook materials when they analyzed their usage in this program. Some of the drawbacks they suggested were discoverability, particularly making titles appear in WorldCat. They also had problems with duplication in their print workflow, selectors were not always made aware of the discovery records in the system and sometimes duplicated the titles in print (Nabe, Imre, & Mann, 2011). Embarking on a first DDA or ebook program as part of a consortium can help ease this transition without the stakes of a budget provided solely by the institution and the burden of negotiating for fair prices and beneficial policies. Consortia negotiations can have a big effect on vendor policies and often member libraries are the beneficiaries of these agreements. There are challenges to setting up consortial workflows, making sure member libraries feel that their stake in the endeavor is fair, and ensuring that materials are equally accessible between institutions, but for many consortia the benefits far outweigh these drawbacks.

10.3 School Libraries

Ebook programs are slowly growing in school libraries, School Library Journal’s 2014 Ebook Usage in U.S. School (K-12) Libraries survey found that 92% of programs relied on school-owned devices to read ebooks but one-to-one technology programs bumped up the percentage of students that were reading books on their own tablet or dedicated e-reader (http://www.slj.com/2014/11/industry-news/ebooks-take-hold-slowly/). Caltech demoed a Kindle-based DDA program that let users purchase materials on demand through an institutional Amazon account (McCaslin, 2013) and a modified version of this might be a good option for school library DDA, especially if the library has dedicated tablets or readers available for checkout and the librarian can mediate the purchasing process. Another option is an external service like Brain Hive (http://www.brainhive.com). Brain Hive is a pay-per-use ebook platform for school libraries that also offers a perpetual purchase option and catalog integration. School libraries assessing their pay-per-view programs could set up a manual short-term loan-to-purchase strategy in which they could seek out purchases for heavily checked-out materials. This would be a little less convenient than a fully automated DDA program, but still serve the same purpose in a school setting and with an appropriate selection of titles.

One of the primary concerns of school librarians offering ebooks to their patrons is the potential effects on comprehension for students still growing as readers. The literature on this subject is conflicting, with some well-cited studies reporting that comprehension is better on paper than computers (Mangen, Walgermo, & Brønnick, 2013), while others find that there is no difference between the two formats (Porion, Aparicio, Megalakaki, Robert, & Baccino, 2016). It is clear from the research on this subject that many factors including the purpose of the reading, the format of the device that is used to access the material, and the arrangement of text on the screen all influence readers’ speed and comprehension. It is probably true that some ebook vendors produce content that has a similar effect as reading on paper, that some content (like graphic novels) work just as well in digital format as physical, and that students used to reading on screens have fewer problems switching between physical books and ebooks.

School librarians should use appropriate caution introducing any technology to their students, but learning to use both ebooks and physical books in a library setting might best prepare students for higher education libraries. A study of undergraduate students found that searching strategies were very different between physical books and ebooks. Students used a linear approach to find specific information in physical books using the table of contents and index to investigate potential areas of interest while ebook users preferred to use keyword searching to find relevant content (Berg, Hoffmann, & Dawson, 2010). Though both ebook penetration and DDA-enabled strategies have been slow to develop in school libraries, there’s a lot of potential for student learning and collection benefit from assessing school library collections.