The stories we can tell ebook usage in academic libraries

Editor's Notes

• #3: Image credit: http://www.worldatlas.com/webimage/countrys/namerica/province/onzna.gifWe’ll touch on:How we anticipate user interest or demandHow we make the best with our (limited) fundsHow we evaluate ebooks post-acquisitionHow we recreate and sustain success
• #4: We hope to touch upon our combined experience in all of these acquisition methods and How these acquisitions have been evaluated in our libraries
• #7: “Past usage is not a good predictor of future usage” (Bucknell, 2010)Nevertheless, usage is the low hanging fruit…Need to consider other data ILL statsPrint counterpart usage
• #9: Usage is a blunt instrument Usage suggests interest, popularity, patterns of use (frequency, trend, intensity)But we still lack a metric that actually measures *READING*
• #10: Consortium of 21 publicly funded universities in Ontario, Canada 21 university libraries. Some research intensive and some focused on teaching and learningA mix of arge and small institutions.
• #11: Acquired books are locally loadedMainly packages, direct from publisherAcquisition priorities and negotiations are guided by committeeModel license in placeDRM free locally loaded titles AND DRM restricted titles600,000+ titles so far
• #12: Challenging to evaluate because of multiple access points and lack of uniformity between our usage stats and the vendor’s Local load – preservation and perpetual accessAccess to logs and reliable usage dataComparison and benchmarking amongst member schools Acquired books are locally loadedMainly packages, direct from publisherAcquisition priorities and negotiations are guided by committeeModel license in placeDRM free locally loaded titles AND DRM restricted titles600,000+ titles so farUnique position of having a consortial platform where books are loaded in parallel to their availability on the vendor’s native platform. Allows libraries to have a reliable point of perpetual access as well as trustworthy usage statistics and comparison and benchmarking with comparator institutions in the province that subscribe to the same content
• #13: Challenging to evaluate because of multiple access points and lack of uniformity between our usage stats and the vendor’sChallenges – discoverability – targets have to be enabled in the OPAC, ordering of SP target vs. vendor platform target, delay in loading of books and records
• #14: Success depends on the intent with which the content was purchasedConsiderations –How much of the collection is being used? Long tail of unused titles or well spread?How intensely are titles being used? Are there a few key titles driving use?Is the package more popular with some schools? Might be a better buy for some schools and not othersCost per use/cost per title – was it a good buy?At the local level, schools might do further analysis against print equivalents, against ILL stats, etc.LimitationsDepends on the discoverability of the collection in that school’s OPAC.Are the records there? Which target is first in the list?
• #15: Context slide about the University of GuelphCARL = Canadian Association of Research Libraries
• #16: http://www.lib.uoguelph.ca/about/vision_&_values.cfmStewardship: One of the core values of the Library is our ethical responsibility for stewardship. Stewardship encompasses both our responsibility for the care and preservation of knowledge and information resources entrusted to us, and its transmission to future generations. Our stewardship reflects a bond of trust between the Library and our community as well as our mutual interest in and commitment to the cultural and historical legacy of the human record.Access: Ensuring equitable access to information and services is a core value of the Library. We provide impartial and non-judgmental access to resources and services. We strive to provide services, facilities, collections and systems that are accessible to all individualsThe library is central to campus and provides a strong collection and also 93% of the study space on campus. A member of 3 library consortia – provincial, national and local. The local consortia of 3 universities has a shared storage facility, fondly titled the ANNEX.
• #17: Our collecting has always been closely tied to the curriculum and research needs of the university. In this we are like so many other research libraries, creating a carefully curated collection of books for our users to choose from. We share our catalogue & discovery system with two nearby universities and collection building is done with this in mind; eBooks throw a real wrench into this paradigm with respect to interlibrary loan.Library culture – in 2009 reorganized from a primarily liaison model to a team based model.Collections Team: resource selection done by team of collections librarians - values rooted strongly in creating a curated collection that aligns with the curricular and research programs of the university. Decisions about ebooks are made both at the individual librarian and team levels. Working in a small focused group is advantageous when dealing with the complex ecosystem of ebooksD&A: Works to optimize findability, accessibility, and usability of resources and seamlessly connect users with resources and services. This team is charged with continually improving the user's experience by designing and delivering dynamic user services grounded in a deep understanding of user behaviours, expectations and needs. User services include online and in-person services and point-of-need assistance. The team participates in the creation of an adaptable information network and architecture and provides tools to enable discovery and enhance user productivity.For ebooks these two teams must work together since we are no longer just buying content…http://www.lib.uoguelph.ca/about/team_based_service_model.cfm
• #18: - We are buying content, we are “we buying content, software, licenses, DRM and an ongoing relationship with a vendor.” – Sue Polanka (http://ebookchallenge.org.uk/wp-content/uploads/2012/09/Curation_ebooks_Polanka_Sept2012_for_JISC.pdf)Created an ebook Strategy Working Group …. from selection to UX
• #19: We select ebooks instead of print books for a variety of reasonsPackages – via consortia or on our own, frontlist or backlistE-preferred approvals – in some call number ranges onlyTitle by title selection – usually via book vendor, but sometimes publisher directStanding orders – convert to ebooksDemand driven and patron driven models – we are very interested but most of these are only available via aggregators which is not our preferred purchase model for reasons already stated.
• #21: Prefer purchase over subscriptions – we are still building a collectionPreference is DRM free (multiple user), perpetual access – including local load on Scholars PortalFor restricted titles, prefer multiple user access
• #22: Range of platforms:Publisher direct examples – Wiley, Springer, Elsevier, RSC, - this is where we purchase backlisted titlesAggregators – general, subject based, university presses – e.g. Project Muse Specialized collections – by topic or type of resourcePrimary source databases – ECCO, Alexander Street PressOpen access – including Hathi Trust, Project Gutenburg, Internet ArchiveEbooks on reserve for Fall 2013 ~250 ebooks or ebook chapters on 31 different platforms including open access ebooks on Hathi Trust, Project Gutenburg, Internet Archive & elsewhere
• #23: - Usage data – not all BR2 Counter reports are alike or are available – what is counted varies even within standardized Counter reports e.g. BR2; Title by title reports not always available- Monitoring turnaways – look at turnaways for user limits, but also some publishers provide for non-purchased titles – see what our users are trying to access – items in shared catalogue held by one of the other two universities; - Overlap between platforms – local load pretty much ensures this, but it also happens elsewhere- Usage can only happen if titles can be discovered – delays in access mean less (or no) usePoor MARC records can also inhibit discovery – no subject headings / no chapter titles or summariesUX study – some of our users would rather choose another book or even change their essay topic rather than jump through DRM-related hoops like having to sign in to download a book
• #24: Purchased front list ebooks from major publisher and blocked this publisher on approval plan. Special offer, OTO funding. Not a consortial purchase.AdvantagesOne invoice – much less time required by acquisitions staff in ordering ebooksCost effectiveness – good deal; value for moneySimplified collection – one decision buys a lot of contentPublisher platform free of Digital Rights Management restrictionsNo limit on the number of simultaneous usersLong term preservation via local loading on Scholars Portal – this is very important in terms of actually owning and hosting the content; recognize that this is usually not an option
• #25: What didn’t work:Collection development- turns out it doesn’t include all of the books, just the ones available on the publisher’s platform which exclude books expected to be ‘course adoption’ books; in some areas it was unclear whether or not certain books would fall into this area; figuring all this out tended to cancel out a lot of the original efficiencies for ebook selectionProblems with quality of MARC records – a lot of records so required a lot of time to fix themWorkflow – too many MARC records to handle with current staffing, access delayed by over a yearUsage data – low, due to delay in discoveryUser experience – DRM free, pdf based is what our users like
• #26: What we are doing differentlyNot buying the front list packageRemove publisher from blocked list in approval plan and instead set to slips (we have one big approval plan but only some call number ranges are set to e-preferred)Select as slips for access via the publisher’s platformPurchase ebook cataloguing records from book vendorWhat changed to make this possible? Can now buy titles available on publisher platform via our book vendor.Advantages are we know what we own, we still get the DRM free platform and local load but the titles are discoverable much sooner so we hope that use will increase.
• #27: Disadvantages:- Requires more time for selectors to choose titles and acquisitions staff to order titlesHigher cost - not only for on a cost/title basis but also we are now purchasing the records from our book vendor
• #28: Prefer purchase over subscriptions – we are still building a collectionPreference is DRM free (multiple user), perpetual access – including local load on Scholars PortalFor restricted titles, prefer multiple user access
• #29: Time lags caused by problems of entitlement confusion and poor quality MARC are unacceptable.User experience – some users would rather choose another title or even change their topic than jump through DRM-related hoopsIn both models cost sustainability is an issue, but it is easier to adjust mid-fiscal when buying title by title. Some of our preferences (good quality MARC records) are expensive – will this be sustainable for us?
• #30: Ryerson is opting out of ebook packages in favour DDA via aggregator on a large scale. Our budget can no longer sustain the increased costs behind ebook packages and we are weary of creating another Big Deal monster.
• #31: University status granted in 1993, previously a polytechnic institutionUrban commuter campus 38,950 students, including Masters, PhD and large continuing education cohortFocused on innovation and career-focused education
• #32: Use drives both our selection and our de-selection decisions. We are an urban campus with a finite amount of space that will not see growth and we do not have offsite storage.We emphasize access over ownership. Our polytechnic roots are likely why we approach our collection this way. Collectively, we have not attempted to provide in-depth collections in any one particular subject area for the circulating collection. Aside from a small number of areas that are notorious for copious publishing such as English, History and Philosophy, the vast majority of our collections are automated with limited firm ordering. DDA was welcomed with open arms by our liaison librarians who often had a difficult time prioritizing firm orders with limited budgets. We are driven by efficiency and take out extraneous steps where possible. For example, our DDA is unmediated and we do not reject approval books
• #33: This graph shows our history after having implemented approval plans when we saw a massive budget increase to support new grad programs. After 2008, our budget began to sustain year over year cuts, first controlled by swift cuts to the approval plan, followed by an increase in DDA.
• #34: With a couple of exceptions (e.g. EMP) we are publisher neutral – the plan is subject drivenWe are e-preferred in all areas except for Design related (Architecture, Interior Design, Fashion and Image Arts) due to ephemeral nature of contentWe have signed on with all aggregator platforms in order to be inclusive on content. Excluding content based solely on whether we had a license for a particular platform made no sense, so we took the plunge and signed with all (except publisher platforms with fees)Our auto-shipped titles are a very exclusive bunch and have steadily declined over time due to budget reductions (easily reduced target) – only those with Research Essential or Basic Essential tag, all others are slipped/moved into DDA pool.Increasingly we are moving away from firm ordering. STL threshold is low – only one STL and subsequent use triggers purchase. Ebrary is auto-purchase (SUPO) based on the assumption that in absence of demonstrated demand (i.e. a course reserve), a single user license is sufficient.
• #35: One source for cataloguing, able to integrate into existing workflow – one step EDI loading for bib/order/item via a load table that automatically populates the relevant fixed fields.One source for purchase means greater confidence in duplication control, particularly by opting out of front list dealsRecord quality from the large packages has been inconsistent at best. Our cataloguing department spends a tremendous amount of time enhancing them with subject analysis and LC classification. Records are also available in a dramatically shortened turnaround time.Invoices are simply incorporated into existing workflow, less staff time at all levels (i.e. having to get approvals from higher levels of admin due to large amounts)We are able to track price data at the item level to support subject analysis, whereas this is impossible with package deals
• #36: This graph pretty much speaks for itself. We’re looking forward to continuing our data analysis for a longitudinal study, hoping that we see more uptake of the approval and firm books, but in the short term, as long as use is our main driver in decision making, the evidence indicates that the titles we are letting our users select are getting more bang for our buck.
• #37: Of course, I would be remiss if I did not acknowledge that this is not all sunshine and rainbows. There are a couple of noteworthy drawbacks to mention.You may have noticed that I haven’t included much in the way of expenditure. This is not an omission. The bottom line is that as it stands, this model is unsustainable for publishers. It may seem odd for me to be playing the role of apologist, but our financial gain is the publisher’s loss to their bottom line and we need to find a way to make this model sustainable for all.Those conversations are actually taking place at this conference, so I won’t belabor the point.
• #38: And of course, the elephant in the room is that by choosing content via DDA, we are deliberately choosing “here today gone tomorrow” aggregator content, and saddling our users with cumbersome DRM.
• #39: So, to recap, Ryerson is basically taking the opposite approach to Guelph, as is starkly demonstrated by our preferred platform as our vendors see it.