By Emily Campbell, Resource Sharing and Assessment Librarian, University of Michigan Library, and Ken Varnum, Senior Program Manager for Discovery, Delivery, and Library Analytics, Library Information Technology, University of Michigan Library

From our positions in the University of Michigan Library, with our responsibility for the library’s main discovery interface and the document fulfilment service, we have been part and parcel of the changes sweeping our institution. The way our scholars — not only established faculty and researchers, but nascent scholars at the beginning of their student careers — expect to identify and access the resources they need is constantly evolving. The way our community sees and understands information is influenced by a wide range of experiences, from burgeoning social media platforms, Google Scholar and copyright-conscious, discipline-focused open access archives, and thumbing-one’s-nose-at-copyright archives like SciHub and the humanities-focused Memory of the World. What has not changed in our service population is the need to identify and access the raw materials of scholarly research. What has changed is the scope of discovery and expected speed of access that equates to good service.

Approaches to Discovery

As noted in the ITHAKA report (figure 1), faculty choice of the library’s website or online catalog as the starting point for research has been more or less flat over the past 6 years, at around 20%, while use of Google Scholar has increased over the last 3 years, and use of scholarly databases has declined. While our overall licensed database numbers are not reported based on user role, our own database use figures show a decrease of about 10% from 2015 to 2018, although the trend is not very clear:

Figure 1: Database Regular Searches (COUNTER-compliant)

Interestingly, reporting from databases which are not COUNTER-compliant (this is most often STEM-related databases) shows a smaller percentage decrease over the same period, while still mirroring the pattern (a drop from 2015 to 2016, holding steady for two years, and a return to approximate 2015 levels in 2018) of COUNTER-compliant databases (figure 2).

Figure 2: Database Regular Search (non-COUNTER-compliant)

We do see trends similar to the national one in the way different types of scholars choose their starting points. We believe this is due in large part to the kinds of materials scholars from different disciplines. Scholars in the sciences and medical disciplines tend to focus on the “new,” most often article-centric, research to advance their research goals; scholars in the humanities and social sciences focus more on longer, slower-to-produce texts in more traditional forms.

In recognition of user- and discipline-specific needs, we last year launched a new search interface that allows scholars to start in the domain that most readily meets their needs, and to easily switch to a different approach when needed. This interface integrates what were five disconnected tools for discovery — traditionally-cataloged materials, licensed and open access materials, databases, journals, and information about the library and conducting research — into a single place. This structure provides needed flexibility when a researcher’s commonly-trod path to information is not sufficient, making it easy to take a different road when the need arises.

Approaches to Access

As the traditional means of discovery, such as the library catalog, have given way to the use of Google Scholar and the open web, our Interlibrary Loan requests have also evolved. Our faculty and graduate students are the primary users of our Interlibrary Loan services and while the number of requests is trending downward, the complexity of the requests has increased. While many of our requests come in through our catalog, the more complicated requests may come in with little to no information because the starting point was not the metadata rich catalog or article discovery tool, but rather a search engine or a referral from a colleague. These requests require more intensive staff mediation and there is frequently a need to conduct a reference-like interview to understand what the patron is looking for. At Michigan, we have been able to automate a significant amount of the behind-the-scenes work, such as call number lookup, which increases our accuracy and allows staff to handle the harder cases. This also increases the speed at which we can fill a patron request, which may not be immediate but can be within a day or two depending on the format. With the advent of real-world delivery services that promise immediate or almost immediate gratification (think Amazon or GrubHub), we are under pressure to provide the same level of service or risk losing the interest of the researcher.

It is true that immediacy is one of the driving forces for our patrons, but it isn’t the only one. There are still concerns or issues with various ebook platforms and the acceptance of ebooks can be discipline dependant  We have been seeing more patrons request a physical copy of a title, even if the library already holds the ebook.This has led us to supplement the catalog and allow patrons to request one or the other.  As the acceptance and desire for more Open Access material continues, we are looking at ways to integrate the searching for OA material into our behind the scenes processes, as well as in our patron facing interfaces. Our consortial work with the BTAA has provided us with a reliable and inexpensive way to quickly access the collections of our peers without a separate discovery interface because of our behind the scenes work that leverages technology to search the UBorrow catalog for us.

We see the core element of our service model, the thing that sets us apart from unmediated services, as service itself. While our actual collections are large, our service model enables us to provide both efficient, personal service to our campus, as well as draw on the immense scope of authoritative, vetted information resources our researchers need when “almost the peer reviewed version” is not sufficient. The change in access patterns is an opportunity to focus our resources where they have the most effect on the core missions of a research university.

Emily Campbell (ekflanag@umich.edu) is Resource Sharing and Assessment Librarian and Ken Varnum (varnum@umich.edu) is Senior Program Manager at the University of Michigan Library in Ann Arbor, Michigan.