Minutes of the January 24, 2008 Meeting

SML 409
2:30 – 4:00 p.m.

Present: Hannah Bennett, Gwyneth Crowley, Jo-Anne Giammattei, Graziano Krätli, Haruko Nakamura (Chair), Audrey Novak, Ann Okerson, Kimberly Parker, Susanne Roberts, Marcia Romanansky, Andrew Shimp, Paul Stuehrenberg, Dorothy Woodson.

Absent: Cynthia Crooker, George Miles, Martha Smalley, David Walls.

Guests: Nisa Bakkalbasi (Electronic Collections).

Recorder: Graziano Krätli.

I. Announcements, questions, future agenda items

Kimberly Parker (Head of Electronic Collections) announced that CoDGeR had recently approved an ongoing "blanket order" for Yale University Press books in online form via NetLibrary. As a result, the Library can now provide access to more than 750 titles, a number that will eventually increase to at least 2000, as more titles become available over a several year period.

II. Report on journal package usages (Nisa Bakkalbasi)

The Yale Library has been collecting usage statistics for subscribed electronic resources ever since these became available in the late 1990s. Nisa Bakkalbasi (Electronic Collections Librarian) showed a Usage Statistics Web page specifically set up and maintained by Electronic Collections for this purpose. It lists a few dozen "platforms" (from ABC-CLIO to Women Writers Project) and provides access to COUNTER-compliant data collected annually. (Nisa invited selectors to let her know if any publishers or vendors are missing from the list, while also pointing out that usage statistics reports are becoming increasingly available for electronic books.) Some reports are protected, which means that selectors and other interested staff need to obtain a password from Electronic Collections. The page was recently modified to reflect a decision caused by the growth of electronic acquisitions as well as of data made available by publishers; that is, the page is an archive raw data only without analysis.

Over the summer, Nisa performed statistical analyses on COUNTER usage reports for six journal packages to which the library subscribes: Blackwell, Cambridge, Elsevier, Nature, Project Muse, and Wiley. The results were summarized in An Analysis of Usage Statistics Reports for Six Journal Packages, which Nisa distributed and described at the meeting. The report is organized in two sections. The first compares usage statistics across the six packages, while the second presents detailed statistics (such as monthly and annual usage, top ten downloaded journals, etc.) about each package.

The six packages were chosen because they represent different combinations of journals as well as different types of publishers, both commercial and non-commercial (Project Muse), with some of them specialized in certain areas (e.g., Elsevier's strength in STM publications). The descriptive statistics for each package shows the total number of journals (both subscribed and non-subscribed), the average article download, the standard deviation and the coefficient of variation (the latter indicating the variation of title-to-title usage within each package).

The total monthly downloads for each journal include both articles that are simply "viewed" and articles that are actually downloaded.

Two significant trends were highlighted by this report. One, regarding the cumulative percentage of downloads, shows that a visible number of titles in each package are never used, while a fraction of is heavily used. For example, 90% of articles downloaded come from 20% of the journals in the Wiley package.

The other trend is the monthly usage pattern. For some packages (Blackwell, Cambridge and Project Muse), this follows the academic calendar, with higher usage during the spring and fall semesters and lower usage during the summer months. For others, instead, such as Elsevier and Nature, usage remains fairly uniform throughout the year. (In Wiley’s case, Nisa pointed out a strange and suspicious peak for the month of April, adding that further investigation is needed in order to confirm it.) In regard to the former case, Audrey Novak (Manager, Integrated Systems and Programming) pointed out similarities with Orbis usage statistics.

The discussion focused on the possible use of the type of information generated by such a report. Nisa, replying to a specific question from Marcia Romanansky (Chief Acquisitions Librarian), explained that it is still too early to say how and to what extent usage statistics could help with selection decisions, since we just started collecting and analyzing relevant data. Kimberly Parker, who is coordinating periodic reviews of vendor packages, remarked that individual package costs were intentionally left out of the report, as the latest figures are not available yet. The cost element, however, plays a significant role, given the fact that a small percentage of journals in each package are used heavily; therefore, the question arises about whether it would be more cost-effective to subscribe directly to these more popular journal titles, rather than to pay for the entire package. But, as it turns out, much of the time the latter solution remains the more convenient and cost-effective.

III. Borrow Direct and cooperative collection development (Susanne Roberts)

Sue Roberts (Librarian for European History and Coordinator of Humanities Collections) reported on a meeting she attended at the recent American Library Association's midwinter conference in Philadelphia. The topic of using Borrow Direct for cooperative collection development initiatives was proposed by Stephen Lehmann, Librarian and Coordinator of Humanities Collections at the University of Pennsylvania, and discussed by Martha Brogan, Penn's Director of Collection Management, at a meeting the day before (Yale librarians were unable to attend because of schedule conflicts).

In anticipation of this meeting, Mr. Lehmann had conducted a simple survey on 29 French books on environmental history, 12 held by five or more libraries, 17 by four or more, and 24 by three or more. He had found out that 13 of these books had never circulated, 4 had circulated once and 2 had circulated twice. These findings – and the likelihood of similar results for a significant amount of Borrow Direct titles – raised the question about whether it would be possible to distribute collection development strategies and responsibilities for less important titles among Borrow Direct member libraries. The Philadelphia meeting provided a good opportunity to review past attempts in this direction, assess their outcomes and shortcomings, and possibly find a more effective approach to cooperative collection development.

Part of the discussion focused on how and why past cooperative collecting initiatives have failed. The main reason seems to be that, in spite of cooperative agreements, member institutions continue to buy individually, and information about who has (and buys) does not circulate as quickly and efficiently as it should to coordinate acquisitions and prevent overlaps. Cooperative agreements and their effectiveness are further affected by institutional changes (curricula, student body, etc.) and external, industry-related events (e.g., RLG and OCLC merge in 2006). The object of the Midwinter meeting was, therefore, to explore new approaches to cooperative collection development, such as using Borrow Direct as a tool. On this point, Sue acknowledged that Borrow Direct data can be misleading, as Yale patrons are known to request books that are checked out rather than books that are not in the collection. On the other hand, Audrey Novak remarked that, technically speaking, the Borrow Direct concept could be easily applied to see who collects what.

A few ideas were tossed around, such as looking at certain publishers (2 nd tier and up), Area Studies, and the OCLC WorldCat Selection service, and eventually participants agreed to focus on a combination of the first two. As a first step, Sue promised to explore the catalog of the German publisher LIT Verlag, which focuses on geography, history, and the social sciences, while her colleague at Cornell agreed to look at series in South East Asian studies. (One of the problems, in dealing with Area Studies, is that they are organized differently in each library, and this of course affects the way materials are collected. For example, one Africana curator may collect materials on Africa, no matter where they are published, while another collects materials published in Africa only).

Questions focused on the possible ways a cooperative collection development option, such as the one discussed in Philadelphia, could actually be implemented. Andrew Shimp (Engineering Librarian) wondered whether the solution was as simple as drawing a list of titles that nobody had, dividing them among member libraries, and deciding who would buy what. Sue acknowledged that there was more at stake, and things were not that easy. A few years back, Harrassowitz had offered German Studies librarians a list of approval-level titles that nobody had bought. However, they first had problems creating the list and making it available online, and then, when the list was finally available, no one of the recipients knew for sure what to do with it. Another attempt was made to acquire Italian local government publications through Casalini, but it did not work very well and lasted no more than a couple of years. Ann pointed out that similar efforts in Area Studies are more successful when a strong organization is involved (such as those under the aegis of the Center for Research Libraries). Haruko mentioned that, in the case of Japanese Studies, a group of Ivy League libraries agreed to collect local history region by region, each institution focusing on a specific region. The system works, although a combination of competitiveness and institutional pride will occasionally tempt some member institutions to overstep the agreed-upon boundaries and try to collect materials from all regions.

Sue confirmed, adding that cooperative collection development agreements works well with close-knit associations and consortia (Research Triangle Park, BayNet, etc.), while they prove very difficult for an institution such as Yale that has (a) and encyclopedic collecting mandate, and (b) no comprehensive consortial relationships.

Marcia added that the demand time for a large percentage of scholarly books is delayed until they are cited in other works, often years after they are published. She then asked how approval plan vendors could support such a cooperative agreement. According to Sue, this would not be difficult technically; the real problem would have to do with the scope of individual profiles and the extent to which a library relies upon its approval plans.

Ann mentioned a 2003 meeting at Columbia University, which she attended with other Yale librarians (Andy Shimp, Tobin Nelhaus for Drama, Film, Theater Studies, and Simon Samoeil as Near East Curator). The purpose was to explore cooperative collection development opportunities, but the group did not succeed at crafting an agreement on how to go beyond individual participation. Since then, six out of the seven libraries represented at that meeting have changed their Associate University Librarian in charge of collection development: another example of how institutional changes may hamper the progress of such initiatives.

IV. Report on new CDC Education Committee (Haruko Nakamura)

Haruko Nakamura (Librarian for Japanese Collection) updated the Council on the revamping of its Education Committee, now with a membership including: Anne Oechtering (Librarian for Philosophy and German Humanities), Emily Horning (Librarian for Religious Studies, and Anthropology, and Coordinator of Instruction), Joan Emmet (who will continue to maintain the Web site), and Haruko herself.

V. Report from the NERL consortium meeting - Use of subscription agents for purchasing electronic packages from large publishers (Ann Okerson and Kimberly Parker)

Ann reported on a January 11, 2008, all-day NERL (NorthEast Research Libraries) meeting held in conjunction with the American Library Association's midwinter conference in Philadelphia. The heavyweight topic was subscription agents and packages, and two representatives from each of the three largest subscription agents (EBSCO, Harrassowitz and Swets) were invited to join NERL participants at the meeting.

Like other members, Yale subscribes to a number of journal packages through NERL. Some of these packages include hundreds or thousands of e-journals, although in such cases no consortial member had ever subscribed to the entire print collection (whereas the package gives access to each and every e-journal title in the package). Subscription agreements are negotiated at Yale by Ann and Joan Emmet (NERL Program Support Librarian), who try to obtain the best prices and usage terms and conditions. Vendors either bill NERL, which then rebills individual libraries, or send their invoices directly to member institutions. The second approach is preferred, as it requires less work on NERL's (i.e., Joan’s) part.

However, in the last couple of years, a growing number of NERL member libraries started to question the effectiveness of receiving one invoice per package. For budgetary and statistical reasons, they need their invoices to show a title-by-title breakdown, with detailed information attached to each title. Kim pointed out that the Yale Library only pays attention to detailed, title-by-title information in a cycle (every few years), while in the meantime YUL can address the proportional cost of the entire package. She feels nervous about subscription agents handling everything, since they have a very limited knowledge of the large picture.

In anticipation of the Philadelphia meeting, Ann sent a brief survey to NERL members, asking them to indicate their e-package service needs. She received a variety of responses, and when some of the needs were discussed in Philadelphia, participants realized that a number of services requested by libraries are brand new; agents expected to provide them should be compensated for the time and resources invested. On their part, subscription agents are optimistic that they can deliver whatever services libraries may require, while the publishers that Ann met at ALA were far less confident. At the same time, agents think that the whole concept of package pricing is doomed to evaporate, to be replaced by some kind of agreement based of user demographics and other defining elements such as usage.

Some next steps discussed at the NERL meeting were to (1) conduct a needs assessment; (2) discuss these needs in a meeting with agents; and (3) document the workflow required to fulfill such needs.

On March 12th, Ann, Joan and a couple of e-resource librarians are scheduled to meet with Elsevier representatives and try to figure out the kind of workflow required in a 4-5 ways relationship, which presents a more complex and challenging scenario.

The meeting was adjourned at 4:00 p.m.

Next meeting: February 21, 2008.