Evidence Based Library and Information Practice

Setting

The University of Memphis is a public university in an urban setting in the United States. The University Libraries use Innovative Interfaces, Inc.’s Sierra Integrated Library System (ILS). Using Sierra, a comprehensive inventory was conducted in one of the university’s branch libraries, the Music Library. The Music Library houses reference materials (2,803); non-circulating scores (6,189); circulating scores of different sizes—standard (19,503), oversize (307), and

miniature (2,115); desk items such as CDs and DVDs (10,263); and standard circulating monographic items (14,357) in a collection totaling around 56,000 items. The School of Music at the University of Memphis has roughly 400 undergraduate students, 100 graduate students, and 45 faculty members who comprise the main patron base of the Music Library. The Music Library nonetheless assists a variety of patrons, including undergraduate students, graduate students, faculty, and staff from all university programs, as well as community members who are not affiliated with the university. The personnel involved in this project included two faculty librarians, two full-time staff members, and one graduate assistant. It had been several years since the collection had been partially inventoried and the status of the physical collection was increasingly disordered. Although this workflow represents an example from a music library, the steps are applicable to most academic library settings because the branch includes a variety of item locations, types, and statuses.

Problem

Collections are not static and require ongoing analysis of their physical condition, availability, and the degree to which they meet the needs of their users. That an inventory of the Music Library collections was long overdue had become increasingly evident over the last few years. Library staff and student workers were spending more time with patrons in the stacks attempting to locate known items. Some items were merely shelved incorrectly, but other items were found to have call numbers, locations, or item statuses that differed from what was indicated in the catalogue. Music Library personnel kept a tally of items that required a search of the shelves and found that in roughly 25% of searched items there was an inconsistency between the information contained in the ILS and the physical item. This finding underscores a significant problem: if our catalogues do not accurately describe what is available to users, we will not meet their information needs. Consequently, we were interested in aligning the physical collections in the Music Library with the data in the ILS, and in using inventory data to support collections-related decisions.

Evidence

There is detailed literature on inventorying academic library collections, including studies that leverage the ILS to partially automate inventory processes (Ernick, 2005; Sung, Whisler, & Sung, 2009; Womack, 2010; Loesch, 2011; Greenwood, 2013). A local workflow for inventorying collections using Sierra has been recently established and is described in Barton & Scott (2020). Much of the literature on conducting inventories, however, is not focused on the particular needs of music library collections, which have relatively diverse formats and a host of complications not encountered with collections that are predominated by books. For example, music scores often contain multiple parts. The existence of multiple parts complicates inventory processes in that an item can appear to be on the shelf in the sense that its barcode is accounted for, but the item can still be missing parts, making it essentially useless. For this reason, we had to combine our local workflow for inventory processes with some music-specific processes, such as separately accounting for parts in multi-part items and assessing handwritten markings on notated music.

With this in mind, Music Library personnel scanned into .txt files the barcodes of all items available, which the ILS Librarian uploaded into the backend of Sierra. Using Sierra’s “Compare inventory to shelf list” program, the inventory date fields in item records were updated and inventory reports were generated. This data is our primary evidence; it reflected both the physical order of the scanned items and the representation of the scanned items in the ILS. The data also uncovered items that were not in the catalogue; that is, their barcodes were not associated with records in the ILS. Most importantly, the data pinpointed items that were shelved incorrectly. This was useful in two distinct ways. First, the reports identified items that were simply misshelved. Second, several items listed as misshelved in the reports were in fact in the correct place. Further investigation, however, revealed that the call numbers printed on spine labels did not always match the call numbers in the ILS. Examination of these items on a case-by-case basis allowed us to determine whether the spine label contained an error, or whether the data in the call number field in Sierra contained errors.

Implementation

During the inventory project, we implemented strategies to benefit our collection in a variety of ways. To do this, we conducted shelf-reading to count pieces in multi-part scores, put items in shelf order, and identify damaged materials. By physically handling every item, we were able to establish the physical condition and extent to which our collections matched data in the ILS. By verifying that all multi-part scores had all parts and were in usable condition, the project led directly to weeding and replacement decisions. Inspection of scores during the project also led to the identification of items that needed care, whether it was erasing markings or basic repair. Implementation falls into three categories: system updates, care or weeding, and acquisitions.

System Updates

Using the reports generated by the ILS, we identified items that were inaccurately listed in the ILS. We noted, for example, that some recital recordings and LPs that were listed as being available in the Music Library were in fact in the University Libraries’ storage facility. The item record location was accordingly updated globally via Sierra’s global update feature. Inventory scanning was done in sections based on item location, and doing so allowed us to globally update those items whose location was not accurately represented. For example, all music monuments have the item location mu51, but after scanning this section, we found items with the location for music reference or scores. We updated various item record fixed fields, including location, agency, and status, with each inventory scanning session.

We also ran reports and confirmed that several items marked as lost or missing had been scanned during the inventory process. These items could then be updated globally to reflect that they were available in the Music Library. Relatively few items were identified that had been listed as being available in the ILS, but had not been inventoried. After searching to confirm that these were not missed or skipped somehow, these items could be updated with an unavailable status in the ILS and suppressed from public view. OCLC holdings were also removed from these titles, as we would be unable to fulfill interlibrary loan requests for them.

Care & Weeding

Before we started scanning the collection for inventory, each of the multi-part scores was inspected to ensure that all of the pieces were in the appropriate folder and that they were in good physical condition. Scores that were missing parts were considered for weeding, or if previous use merited such, replacement. Several items had extensive markings and were carefully erased when possible. If the markings had rendered the work unreadable or had otherwise damaged the score, it was considered for weeding or replacement, depending on usage and other holdings. We added prompts to remind employees to count the number of parts for those multi-part scores that did not already include a pop-up message indicating the number of pieces. After inventory scanning, the music librarian noted some instances in which several copies of lower-quality editions were available and decided to weed some of these.

Acquisitions

We weeded damaged materials first. Then we identified titles that were missing and no longer on the shelf. With this information, we determined titles that needed to be replaced. We ultimately did not need to purchase a great deal of content to fill these gaps. With a large-scale space reclamation project in another of the university’s branch libraries, we were able to compare their music holdings to our recently-inventoried collection and use some of their materials, which were to be weeded, to plug holes in our music collection.

Outcome

The implemented changes ultimately made the representation of our physical collection more accurate. Specifically, over 2,500 records were updated. Some of these were items that were able to be updated en masse, such as 1,713 recital recordings that had been moved to storage years ago without the location having been changed. Another 243 items were found to be missing and were deleted as a result. In addition, 16 items were found to have been sitting on the shelf with barcodes and call numbers, yet without records in the ILS; these were catalogued and added to the library’s collection in the ILS. Around 100 monuments previously had an item status indicating that they were to be used in the library only; these were updated to show a status of available. The remainder of the items were updated with respect to their call numbers, status, or location.

As a result of this project, not only are the shelves in better physical order, but call number discrepancies have been fixed in the ILS. In some cases, the call number that was in the system had not matched the call number on the spine of the book for decades, making the task of finding that particular book nearly impossible. Similarly, item locations for hundreds of items, particularly for reference materials that had previously been listed in the regular stacks, were updated. There is now a sense of confidence that when an item is listed as “available,” it is on the shelf. Prior to conducting inventory, when walking back into the stacks to assist a patron, there was a lack of certainty as to whether the item would be there or not.

Reflection

This process was straightforward, and we encountered no major obstacles. By physically handling every item in the collection, we gained great insight to its physical condition and availability. Updating our ILS by running a systematic progression of reports and global updates is also a process with which we were familiar. By successfully executing this project in a relatively short time period of four months, we realized that it is indeed doable and should be prioritized on a regular and ongoing basis. Because there are so many aspects to a project of this nature—erasing, scanning, reading reports, processing bulk updates to bibliographic records, and others—there are opportunities for all music library employees, including student workers, graduate assistants, library assistants, and librarians, to contribute. Consequently, this large-scale project ultimately proved to be a valuable team-building venture involving close collaboration between personnel in the Music Library, Systems, and Collection Development departments.

References

Barton, G. P., & Scott, R. E. (2020). Relocating, downsizing, and merging: Inventory projects to manage change in a digital environment in S. Marien (Ed.), Library technical services: Adapting to a changing environment. West Lafayette, IN: Purdue University Press.

Ernick, L. (2005). Floating bibs and orphan bar codes: Benefits of an inventory at a small college. Library Resources & Technical Services, 49(3), 210–216. https://doi.org/10.5860/lrts.49n3.210

Greenwood, J. T. (2013). Taking it to the stacks: An inventory project at the University of Mississippi Libraries. Journal of Access Services, 10(2), 77–89.

Loesch, M. F. (2011). Inventory redux: A twenty-first century adaptation. Technical Services Quarterly, 28(3), 301–311. https://doi.org/10.1080/07317131.2011.571636

Sung, J. S., Whisler, J. A., & Sung, N. (2009). A cost-benefit analysis of a collections inventory project: A statistical analysis of inventory data from a medium-sized academic library. The Journal of Academic Librarianship, 35(4), 314-323. https://doi.org/10.1016/j.acalib.2009.04.002

Womack, J. (2010). Inventory or stockcheck? The Christian Librarian, 53(3), 111–113. Retrieved from https://digitalcommons.georgefox.edu/tcl/vol53/iss3/4