by Jean Blackburn (Collections Coordinator, Vancouver Island University)
and Lisa Petrachenko (Associate University Librarian, Learning and Research Resources, University of Victoria Libraries)
Column Editors: John McDonald (EBSCO Information Services)
and Kathleen McEvoy (EBSCO Information Services)
As collections shift toward primarily electronic delivery, and pressures on library spaces increase, academic librarians must manage print collections to acknowledge both decreasing use and an enduring need to preserve the print scholarly record. How do we manage these competing demands? Increasingly, libraries are turning toward shared print archive initiatives to harness the power of group collaboration within our networks to achieve both objectives. The Council of Prairie and Pacific Libraries (COPPUL) is one such network. This regional Canadian consortium, formally established in 1991 and representing the four Western provinces (British Columbia, Alberta, Saskatchewan and Manitoba), has maintained an active shared print archive program since 2012: the COPPUL Shared Print Archive Network or SPAN (https://coppul.ca/programs/shared-print).
COPPUL’s core membership consists of 22 university libraries, ranging from very small teaching-focused institutions to large research universities, spread over a vast geographical expanse. The character and distribution of COPPUL’s membership has resulted in the evolution of a shared print archiving approach that is more distributed than centralized. Maintaining a single shared print repository is neither practical nor desirable in COPPUL’s context. Rather, sharing the responsibility for building and holding shared print archives across the network, according to each member institution’s capacity, has emerged as a more feasible approach in the Canadian West. COPPUL’s well-established resource sharing network has helped to ensure continued access to shared print archives within COPPUL and beyond.
After several phases of journal archiving, COPPUL’s SPAN launched a shared print monograph archiving project facilitated by Sustainable Collections Services (SCS), a small consulting company now owned by OCLC. Ten COPPUL institutions, representing mid-sized research universities and small teaching-focused institutions, signed on to participate. Project goals included identifying unique or scarcely-held titles for retention and preservation, contributing toward the “print safety net” within the COPPUL network, and facilitating the de-selection of print monographs with minimal impact on library users and partners within the network. The SPAN Monograph Project provided an opportunity for smaller libraries to meaningfully contribute to the network’s print safety net (in previous SPAN phases, Western Canada’s two largest research libraries shouldered most of the physical archiving burden on behalf of the network). Resource sharing relationships with other Canadian consortium partners also contributed to the retention model development, in which analysis of holdings of Canadian research libraries outside COPPUL was an important factor in retention decisions.
The project entailed designing shared retention scenarios based on group collections data from OCLC. A key premise of the SPAN Monograph Project was that holdings allocated for retention would remain circulating and able to be shared through interlibrary loan. Hence, the project’s work was conceived and carried out through two main perspectives or “lenses”: preservation and access. Because access and sharing were central to the project, monographs in non-circulating collections (e.g., special collections, reference) were considered out of scope. In scope were circulating monographs classed in LC or DDC, including juvenile materials and music scores.
Retention models, along with treatment of retained titles, were decided collectively with the intent that once retention commitments were allocated, participating libraries (and other COPPUL libraries) could move forward with local collection management projects (including weeding of non-retained titles). Representatives from participating libraries, with the support of COPPUL’s SPAN Coordinator, formed a committee to undertake consensus-building and decision-making throughout the project. COPPUL staff provided project support and liaison with SCS, and the SPAN Management Committee advised on policy matters and best practices. The project intended to complement and support other “last copy” shared print initiatives from Library and Archives Canada and other library consortia.
Group and comparator library data were loaded and available in SCS’s data modelling and visualization tool, GreenGlass, by May 2016. In addition to catalogue extracts from each participating library, SCS loaded OCLC holdings data for many other comparator library groups, e.g., all non-participating COPPUL libraries, COPPUL’s two R1 libraries (University of British Columbia and University of Alberta), other Canadian research libraries, etc. Further, project participants defined criteria to identify materials published in or about the COPPUL region, which were applied to participants’ holdings data in GreenGlass as a “COPPUL Canadiana” flag. With the resulting access to big data, and the means to manipulate and visualize the data in GreenGlass, the SPAN Monograph Project committee began exploring retention scenarios and building consensus around a preferred model. The group chose to focus on rarely-held materials (both within COPPUL and in other Canadian research libraries) and those of regional or local interest; the COPPUL model did not, in the end, factor in usage (circulation numbers and dates), publication years or acquisition dates. In September 2016, after much experimenting, the consensus retention model — actually a combination of two models, “Rarely-held” and “COPPUL Canadiana” — looked like this:
Retain 1 copy if:
UofA/UBC holdings equal 0 (same edition)
Other COPPUL holdings fewer than 3 (same edition)
CARL Libraries* fewer than 2 (same edition)
Not flagged as COPPUL Canadiana
COPPUL Canadiana Model:
Retain 2 copies if:
UofA/UBC holdings greater than 0 (same edition)
Flagged as COPPUL Canadiana
Retain 3 copies if:
UofA/UBC holdings equal 0 (same edition)
Flagged as COPPUL Canadiana
*Non-COPPUL CARL libraries
This model resulted in a 20 percent retention rate averaged across participating libraries — a fairly low retention rate compared with other shared print monograph projects such as Eastern Academic Scholars’ Trust (36 percent) and Washington Research Library Consortium (61 percent) — which SPAN Monograph Project members felt balanced the “print safety net” preservation imperative with local collection management and space reallocation goals. However, more consensus-building within the group became necessary when participants reviewed their retention allocations and experienced various degrees of “buyer’s remorse” at the thought of being compelled to retain items on the list that did not seem truly rare. For example, 7 Habits of Highly Effective People appeared on at least one list — not a rare title by any measure, but because the model specified “same edition” in comparator library holdings, the retentions included items for which a particular edition might be rarely held for a title otherwise considered widely-held. Guided by SCS staff, the committee adjusted the retention model by adding a further criterion to the “Rarely-held” model:
CARL Libraries* fewer than 5 (any edition)
This adjustment reduced the average retention commitment across participating libraries from 20 percent to 16 percent. The SPAN Monograph Project’s collective holdings data comprised 7,276,328 title holdings, of which 1,147,232 were allocated for retention. Research libraries within the group tended to have higher retention rates than those at teaching-focused, primarily undergraduate institutions. The bulk of our combined holdings (60 percent) are unique to our respective libraries, i.e. held in only one participating library. The holdings that are unique have a much higher percentage of zero-use titles — not really surprising, but since the SPAN Monograph Project is focusing on preserving rare materials, it also means that participants are committing to retain many items which have had very low use (i.e., items which may have been good candidates for weeding if not for the SPAN Monograph project).
Other critical consensus decisions included settling on a retention period: fifteen years (with a review every five years), designated by a standard 583 field note for MaRC records (**** is a placeholder for the OCLC library symbol):
583 1#$aCommitted to retain$c20170101$d20321231$f
COPPUL SPAN Monograph$5****
It was understood within the group that participants may make local decisions to add public notes to bibliographic, holdings or item records. With respect to a central registry for up-to-date SPAN Monograph Project holdings, the group is using the new OCLC shared print registry service since COPPUL does not maintain a union catalogue for members, our access to collective data in GreenGlass ended on March 31, 2019, and 583 fields are not visible in Worldcat.
The SPAN Monograph Project committee also decided to observe a limited-time “rejection period” during which participants could review their adjusted retention commitment lists and identify items for removal according to criteria pre-determined by consensus. The Project committee agreed that rejection decisions should not be based on a library simply not wanting to keep certain titles; rather, participating libraries would need to balance the perception of local value with the regional “print safety net” goal. In the interests of furthering the collective “print safety net” goal, the group agreed retention rejection criteria as follows:
• Damaged items
• Outdated textbooks, study guides, and workbooks
• Out of scope materials captured in GreenGlass in error (e.g., non-circulating reference materials, serials)
The group felt strongly that a process to shelf-validate final retention allocations (adjusted for rejections) was critical for identifying missing items, particularly given that many participating libraries had not undergone an inventory process for many years. However, most participating libraries also felt that they could not spare the staff time necessary to validate every item. As a compromise, the Project committee agreed to adopt the sampling methodology developed by the EAST Shared Print Initiative — a group comprised of large and small university libraries much like SPAN Mono Project — which requires a randomized sample of 6000 items, generated by SCS from the project data in GreenGlass, to be verified per participating library (see https://eastlibraries.org/validation for more information). The group further agreed that item location in remote storage facilities was an acceptable proxy for shelf verification. The shelf validation process revealed an average missing rate of about five percent among reporting participants.
With respect to missing items, it was decided that within the SPAN Monograph Project purview, participating libraries would not be obliged to replace items discovered missing in the shelf verification process. However, it was acknowledged that libraries might make local decisions to replace missing items on a case by case basis, and agreed that project members would share information about missing items with one another. Data on missing items, replacements and item transfers between libraries are reported and shared via the COPPUL web site.
The Project committee debated whether or not to designate a preservation copy for each retained COPPUL Canadiana title. Given the SPAN Monograph Project’s access and sharing goals, and since the “Rarely-held” model only retained one copy within the group, the designation of preservation copies was not possible under that model. It was possible, however, under the “COPPUL Canadiana” model which retained two or three copies depending on the criteria. The University of Calgary — the only library in the SPAN Monograph Project group to have preservation storage capacity — carefully considered the possibility of becoming an Archive Holder for preservation copies, but ultimately determined that it was not in a position to offer the necessary storage. As no other participating libraries had storage capacity, the group confirmed that preservation copies would remain out of scope for this particular project.
Related concerns have emerged from scholars at participating institutions about that ensuring adequate interlibrary loan periods are in place for shared print monographs. This has resulted in a national conversation, begun by COPPUL, toward reducing access barriers within Canadian resource sharing networks by increasing interlibrary loan periods to six weeks with the possibility of renewal, and ceasing service fees for interlibrary lending.
Since settling on a shared retention model and retention period, participating SPAN Monograph project members have used project data within the GreenGlass tool to model scenarios for de-selecting low use, widely held materials and advance other local collection management goals in responsible, sustainable, evidence-informed ways. Several members have contracted with SCS separately to load more up-to-date holdings extracts in GreenGlass for continued collection assessment purposes.
The COPPUL SPAN Monograph Project demonstrates a distributed model of shared print archiving wherein smaller institutions, without explicit preservation mandates or extensive storage facilities, can effectively contribute to solving the problem of preserving the print scholarly record into the future. All academic libraries have rare or unique materials within their collections; if shared print archives can be seen as a “print safety net,” ensuring reliable access to the print record into the future, then the distributed shared print archive model allows all libraries the opportunity to form part of the fabric. As a related benefit, participating libraries can move forward with local collection management decisions with increased confidence, knowing that rare and regional-interest materials have been identified and retained within the consortium.