ITHAKA Sustainable Scholarship 2013
I.JSTOR: Can Users Find Your Content Anymore?
Academic content is increasingly published online and then purchased or licensed by libraries. Still, the content is often less visible to patrons than it was in print. As users turn to commercial search engines and other web-scale discovery tools, it remains unclear if they are bettered served, and if the goals of libraries and publishers are being met. How do we consider this challenge together? What practical steps can we take? Learn about our research on JSTOR’s traffic sources, and how these trends have evolved over time. We will look at changing linking patterns related to discovery services, commercial search engines, and different device types. We’ll also preview a multi-institution research project, helping us focalize the impact of web-scale discovery services at a range of institutions.
- Michael Levine-Clark, Associate Dean for Scholarly Communication and Collections, University of Denver
- Bruce Heterick, Vice President, Outreach and Participation Services, ITHAKA
Speaker 1: Levine-Clark
A study to see if discovery tools increase use of e-content, specifically bundled e-journals
NB: Presenter noted that for Charleston Conference and UKSG, more in depth work will be done in this study.
This UKSG Invitation to Tender, “Assessing the Impact of Library Discovery Technology on Content Usage ” maybe what is behind the study. I don’t think Levine-Clark mentioned this as the motivation, but he did mention presenting for UKSG: http://www.uksg.org/sites/uksg.org/files/UKSG_RDS_Impact%20ITT.pdf
We make the assumption that if curriculum and the number of students stays the same, the number of searches will remain the same.
Some observations going into the study: Discovery search will draw students away from other searches; productivity will change because of this. This will also cause searchers to see different versions of articles.
Study examined 24 libraries in US, UK, NZ and Canada using Summon, Worldcat, Primo and Ebsco (six libraries for each tool); libraries of varying sizes with different sized collections; all implemented discovery between 2010 and 2012
Different tools vary with number of titles they offer; number of titles range between 5800 and 23,000+
Of the 4 tools, three showed increase in journal title use
Variations by publisher within each tool
Variations by institution within each tool
Some publishers are net winners, some are net losers
Some results flip (increase vs. decrease) depending on the tool, at different institutions
Speaker 2: Bruce Heterick
Most JSTOR use starts at JSTOR (33%)
Least use, 5%, comes from Serials Solutions; 11% from institutions, 11% from google scholar, 13% from google, 12% self referrer, 13% from link resolver
Google’s algorithm has changed; they now promote free content first
From 2011-2013, JSTOR use has dropped as much as 50% when comparing same month/different year
JSTOR went out to get customer lists from vendors to see if implementation of discovery from 4 major providers played a role
Wanted to see stats before and after implementation
Usage of JSTOR dropped for all 4 discovery tool using institutions – from 1.5-40%
How libraries set admin modules of the discovery tool may be causing this drop of JSTOR use
JSTOR hopes to come up with tip sheets to help libraries set things up (make link resolvers work, turn off newspapers, etc.) effectively
Some findings on what affects findability – subject metadata is important (JSTOR doesn’t have much subject metadata); libraries need to configure the discovery systems (70% leave system at default in the admin – involving the link resolver is really important); JSTOR needs to send its data out to discovery tools more often and discovery tool providers need to index more often. KBART files (which show what an institution has) need to be in good shape
Questions to ask
What are libraries trying to do by implementing discovery tools?
Are libraries doing enough to evaluate these tools?
Is there a good ROI by having these tools?
Is the problem of vendors not sharing metadata at the heart of the problem (e.g., EBSCO won’t give PRIMO data)?
What about other factors? Consider placement on web site, library instruction, etc.
Usefulness and use of discovery tools may vary by discipline
II. JSTOR: Experimental Purchasing Models for Journals
Right now, JSTOR provides access to the current issues of more than 200 journals. Though user demand continues for both new and existing titles, libraries and publishers are increasingly limited in their budgets for new content. How are both of these communities finding new ways to meet demand for journal access? In this session, we will explore the environments in which new experiments are arising, and will dive into two examples—one library using the Copyright Clearance Center “Get It Now” program, and another employing the article service ReadCube—to help us explore what these programs are, and how they function in libraries today.
- Rick Anderson, Associate Dean for Scholarly Resources & Collections, University of Utah
- Cyril Oberlander, Library Director, SUNY Geneseo
- Bruce Heterick, Vice President, Outreach and Participation Services, ITHAKA
40% of libraries spending 50% serials budget to pay for big deal pkgs from three publishers
Society and non-profit Publishers reaction – coming out of packages, increase prices
Speaker 1: Cyril Oberlander, SUNY Geneseo
10% of what researchers want is on open web
What happens when cost is too high? Researchers and librarians usually give up on getting the article
Get it Now, a service from the Copyright Clearance Center, is a way to get content your library does not own (http://www.copyright.com/content/dam/cc3/marketing/documents/pdfs/get-it-now-service-overview.pdf)
IDS partnered with Get it Now which offers two models: unmediated – end user can get needed articles from CCC for about $30; mediated by library – iiliad identifies if library owns the item or if “buy it option” should be offered (library makes the decision to purchase). Very speedy – 98% of articles from Get it Now received within 2 hours.
190 libraries participate in Get it Now.
<4% of requests from participating libraries are being filled this way
Libraries want more publishers to participate
Since ILL costs money (lending charges and royalty charges after 5th request), Get it Now can be cheaper
Users surveyed and have said if library has to pay more than $20 for an article, then forget it
Future – publishers and libraries should collaborate
Speaker 2: Rick Anderson – U of Utah
How do we bring articles into the PDA model?
Document delivery is an expensive model of article PDA
A journal subscription is like a “big deal” for getting articles
Utah doing pilot with ReadCube (http://www.readcube.com/)
ReadCube like Medeley, organizes users articles, and can gather articles; has an online version, Web Reader, in beta, ReadCube Access
At Utah in pilot all of Nature (including unsubscribed access – they subscribe to 78 of 105 titles) articles are available for rental (48 hour read only $3) or purchase (permanant access using ReadCube platform on a single device for $8)
Users of pilot like immediacy of delivery (compared to ILL), but found that device dedication of download was not good (this will change with web-based ReadCube Access). Search limits to Google Scholar and Pubmed a problem.
Usage was lower than expected 1320 invited and 102 registered, only 41 articles purchased, 2 articles rented
Nature Climate Change (subscription would have been $5000)
3 ILL requests cost $16; 12 RC requests $96
Another title (subs. to buy would have been about $4000)
30 ILL ($982)
25 RC ($195)
ReadCube doesn’t build a collection
Helps the individual but doesn’t create access for other users (article purchased is only available to the one user on one device)
Read-only access a problem
Undermines ILL rights
Will publishers consider disaggregating articles from a journal issue and let libraries buy articles? This might dilute the brand of the journal title.
III. JSTOR and Data Mining
It seems that every day a new digital humanities center opens, or a university posts new faculty or librarian positions to support this evolving area of study. Now and into the future, humanists must be able to work with text and data in ways that are constantly evolving. JSTOR’s Data for Research service is now seeing up to 700 datasets downloaded each month, and we work with increasing numbers of scholars on larger, more complex projects that involve the text in the JSTOR archive. Come and learn more about our service—why it is vital to humanists, and what challenges still exist.
- Bob Scott, Head, Digital Humanities Center, Columbia University Libraries
- Kristen Garlock, Associate Director of Outreach & Education, JSTOR
- Nancy Maron, Program Director, Sustainability and Scholarly Communications, Ithaka S+R
- Ronald Snyder, Director of Advanced Technology, JSTOR
JSTOR has a site called Data for Research (http://about.jstor.org/service/data-for-research)
They work directly with researchers (e.g. Mapping networks among authors)
You can request datasets on the fly
Accounts are free for all; affiliation required to see articles
DFR site mirrors regular JSTOR site
Caps at 1000 datasets
Works well as a discovery tool
Speaker 1: Bob Scott, Columbia U
Projects – charting cheers and booing at English parliament over years related to Ottoman Empire
Growing interest in news and press analysis projects
Network analysis projects
Declassified Documents project – What are Feds hiding?
Chartex – discovering spatial descriptions and relationships in medieval charters
Open Syllabus Project
Why do this in the library?
We give access to the resources
New gen librarians are researchers and can work with researchers
New vision of librarianship
Good place to remix
Advisor about resources
In the future
Clarity of data
Speaker 2: Nancy Maron, Ithaka
195 DH centers
See Melissa Terras’ Infographic “Quantifying Digital Humanities”
Many faculty engaged in DH, using collections and tools
They want support and resources and training
IV. What’s on the Horizon?
Over the past decade, Ithaka S+R has completed more than 70 research reports and has engaged with librarians, publishers, faculty, and administrators on a wide range of consulting projects to help shape the future direction of their institutions and products. Over the past year, we have expanded our range of services to include a new training program and several data gathering and analysis services for individual libraries and publishers. We invite you to learn about what is next on our near-term agenda, and to share your ideas about the research and services that would be most helpful to your organizations.
- Rebecca Griffiths, Program Director, Online Learning, Ithaka S+R
- Deanna Marcum, Managing Director, Ithaka S+R
- Nancy Maron, Program Director, Sustainability and Scholarly Communications, Ithaka S+R
- Roger C. Schonfeld, Program Director for Libraries, Users, and Scholarly Practices, Ithaka S+R
This session was a combination of presentations and e-polling on the spot of audience
Audience questioned about do we see lots of changes in publishing coming down the line? Partnerships with libraries and presses? New types of DH projects taking off?
New surveys from Ithaka targeting students
Looking at user-centric design and needs assessment
Work on collections – what to retain, what to withdraw
The future of the monograph
What does it mean to have a research collection?
When queried separately, librarians were more likely than publishers to see a digital-only future for monographs but most see a mixed environment.
Ithaka interested in evaluating new methods, not in advocating a rush into these new technologies
Interested in consumption of new technology (not so much interested in creation of these (MOOCS, for example))
Big barrier to faculty adaptation they want their own mark to be on their courses
David Pakman – a former drummer; worked at Apple and founded Myplay (music in the cloud), worked with Eplay and now is at Venrock (a VC) and on Ithaka board
Where is tech today?
Pakman used to think computers were computing devices but they are really communication devices
Everything is social – brands have to adapt to social
Conversation with consumer must be two way
Apps vs Web – “phone is now the remote control for our lives”
80% of web is US/English in origin but 81% of users are outside of US and English is not the language of the majority of web users
3-D printers – soon will allow embedding electronics during manufacture
CNC machines make pre-fab parts for a house, etc.
Consumer participates in the manufacture (e.g., Nike lets you build your own sneaker and it costs them the same)
Sensor proliferation – devices figure out when to do whatever is needed
Examples of how some industries have adapted to shift
Music – unbundling led to shrinking of the industry; people are buying less, only the songs they want. Bundle puts an artificial value on things
Unbundling in newspapers and other industries
Bundles in higher education – hard to make an argument that these are needed to stay as bundles for any other reasons other than economic
Research and Teaching
Always a loser when things are unbundled.
They are about aggregation and distribution
In an unbundled world, distribution changes
Content discovery shifting to users/social
What is their value?
Do old stalwarts of industry attract the talent to move into the digital world?
Example of Apple’s thermostat the Nest (http://store.apple.com/us/product/HA895LL/A/nest-learning-thermostat-2nd-generation) took on Honeywell
Users confer it by spreading what they like
How to anticipate change
Watch user/consumer behavior
Incumbents start suing when upstarts are gaining ground
In higher ed
Some will unbundle
Physical attendance not required
Publishers will unbundle
Will libraries continue to be brokers of data?
Peer communities will bestow authority
HTML 5 may allow web pages to act like apps and re-open the web
Audience member suggests that the interest in convenience over quality. (Mp3 instead of LP) should not mean that we no longer offer the quality option
Audience member says education is a public good, music is not, so the analogy is not good. Convenient models (low quality) of education will harm society in the end
Day 2 Programs
I. Supporting Lifelong Learners
How can libraries and publishers support individuals who participate in online-learning courses, but who are not affiliated with an institution? Can libraries offer physical spaces where MOOC participants can meet and engage with online courses together? What responsibilities do institutions (as well as libraries and publishers) have with so-called “lifelong learners”?
- Leah Clapman, Managing Editor, Education, PBS News Hours
- David Ferriero, 10th Archivist of the United States (Pamela Wright spoke instead)
- Deborah Howes, Director of Digital Learning, Department of Education, MoMA
- Philipp Schmidt, Executive Director, P2PU and at MIT
Speaker 1: Philipp Schmidt
German educated; he feels government in Germany is more responsive to users
Electro-thermal graph of cognitive activity shows students are least engaged when sitting in class, and more engaged when learning online
Learning Creative Learning (http://learn.media.mit.edu/) is a course about how students learn
Speaker 2: Leah Clapman
Newshour has been unbundled
Community input high
Creating curriculum support
Public “listen to me” videos created based on issues and meant to inform candidates
Speaker 3: Deborah Howes
Educating HS students and seniors on Google-Hangout
Online courses – enrollees from 61 countries; some courses present material that would be hard to show in the physical galleries (e.g., pouring water to music of Flow artists)
Alumni have formed online communities
Invited by Coursera to work with them in teacher training
Speaker 4: Pamela Wright
Citizen archivist initiative (http://www.archives.gov/citizen-archivist/)
Their web site provides a dashboard
Tagging and transcribing projects
They use a wikipedian
Encourage user to uoad photos
Q & A
Is online delivery increasing access?
No, according to Schmidt, MOOC stats show most users are educated white males
But online communities lead to users forming real world bonds and meeting off-line
PBS worked with libraries offering badges for after school programs that have students looking at their content; google hangout connects students across different public libraries
II. Course Redesign
This session explores the implications of course redesign for faculty, students, libraries, publishers, and institutions. How are faculty members redesigning courses for the digital environment? What institutional and network-level resources are needed to improve course redesign planning? How can libraries and publishers support faculty members as they adapt their courses for use online? What impact does course redesign have on students?
- Barbara Anne Bichelmeyer,Interim Chancellor, Indiana University Southeast (also has a background in instructional design
- M.J. Bishop, Director, Center for Learning and Excellence in Learning and Teaching, University System of Maryland
- Marguerite Weber, Director, Student Academic Affairs and Academic Initiatives, University of Baltimore
Speaker 1: Barbara Anne Bichekmeyer
History of learning – it has always been just in time, just in place, just as needed (e.g., caves, Socrates, guilds)
Higher ed was originally a conservatory model and came along with a huge infrastructure
Now everyone is interested in accountability
Impact of internet
conservatory is now in the cloud
Education the ability to do something you haven’t done before; lecture format doesn’t necessarily do this; interactivity is the core to learning. Net encourages interactivity.
New definitions of expertise. Customization of programs.
More competition among programs
Internet is disaggregating teaching from certification
MOOC is textbook of 21st century
Education should be building of bridges not being gate keepers
Speaker 2: MJ Bishop
The Faculty Experience
Faculty feel pressured and that they may be replaced
Mentions E.M. Rogers, 2013 diffusion of innovations theory
Five Factors that determine adaption (does the new method offer an advantage, is it compatible, is it complex or simple, is it trialable, observable?)
Episodic vs. Sustainable change
3 dimensions of Sustainability
Look at economic, social and the environmental factors
In education, these are productivity, learning experience and learning community
Speaker 3: Marguerite Weber
Academic transformation from Student perspective
Design thinking – viability (business needs) desirability (human factors), feasibility (technical issues) – where these overlap is innovation
Course-centered academic innovations
Face to Face – learn with students, team-based learning
Baltimore online learning communities for low-achieving students; students have 3 online courses from UB and then a hard MOOC from Duke on argumentation; students get coaching in how to learn with a MOOC
See chart (slide 9) on what makes innovation worth it; compares cost of each method
III. Digital Humanities in Higher Education
How has the creation of large-scale digital humanities projects changed the way we do our work? This panel brings together those who have created, managed, and used digital resources to share their experiences, and discuss the implications for teaching, research, and librarianship.
Karen Calhoun (moderator), AUL Organizational Development and Assessment, University of Pittsburgh
Mark LeBlanc, Professor of Computer Science, Meneely Chair, Wheaton College
Josh Sosin, Associate Professor of Classical Studies and History, Duke University
Jennifer Vinopal, Librarian for Digital Scholarship Initiatives, New York University
Panelists gave brief intro of themselves and then answered questions from moderator
Sosin, a classicist, particularly interested in inscribed stone; runs center for digital computing; “doing very new things with new tools on very old things”
Leblanc – developed course in computing for poets which is paired with a course in anglo-saxon lit; students learn text mining. Tools are going non-English centric; has students use MOOC to learn coding (http://wheatoncollege.edu/lexomics/)
Vinopal – various projects; digital library tries to involve more users as producers
Digital scholar services – assess project needs, select new services, advice on project planning and management; avoid offering too much customization; discern whom you can support, you can’t support them all
New way of looking at familiar things. Libraries uniquely have expertise to work with these scholars.
Digital Humanities can be very disruptive.
Sustainability and scalable –
We have to anticipate a DH landscape with not much D and lots of H
Solitary scholar model is not workable in DH. Scholarship done mostly in “teams” not individuals.
Class section length is not appropriate for what needs to be taught
A problem – libraries are not funded to do research
IV. Libraries as Publisher
As libraries begin to take on the work of publishers, many new questions arise about the relationship between these two industries. Are academic publishers being absorbed into libraries? How are the two groups working together? What value does each group add in respect to scholarly communication? Are libraries focusing on publishing special collections material?
- Rebecca Kennison (moderator), Director, Center for Digital Research and Scholarship
- James Mullins, Dean of Libraries and Esther Ellis Norton Professor, Purdue University
- David Seaman, Associate Librarian for Information Management, Dartmouth College
- Susan Skomal, President and Chief Executive Officer, BioOne
- Charles Watkinson, Director, Purdue University Press, Purdue University
Two examples where a pre-existing press joined with a campus library (one was affiliated with the campus, the other was not). For examples of libraries that are publishers as well, see: http://librarypublishing.org
2008 PU Press a financial mess; Charles Watkinson brought in, moves press into library
PU has two platforms, PUP – formal publications aligned with mission of host institutions; SPS – scholarly publishing services – conference proceedings, OA journals, repository, lots of tech reports from local institution
Benefit of alliance with library – more financial security allows for experimentation, better digital capacity, closer connection to campus
Challenges – maintain a business-like perspective, avoiding over commitment to new opportunities, retaining editorial independence
Challenges for library – expenses/gains
Editorial decisions – what areas to cover
A little tricky – as a library pricing books that libraries will buy
Elemental – (elementascience.org) partnership between BioOne and Dartmouth Library. Aims to maximize access to critical research (OA author fee journal); publish research articles, commentaries and datasets in a rapid and continuous publication model. Publish in many formats: xml, html 5, json, pdf, epub3, kindle. Launch Nov 2013. First epub journal to be archived in Portica. Covers 6 subject areas
Benefits of being with library – lends credibility with scholars, bring digital library and preservation expertise. Looking for opportunities in new open-source technical infrastructure
Cultural differences between business and scholarly views of the world
Keeping author fees low; big investment in marketing; constant investment in development needed
Uses PLOS platform Ambra
If library goes after a press, it should not be a power grab
V. Learning Analytics
How are people using learning analytics to improve pedagogical techniques? Whose responsibility is it to collect and analyze these data? What role can the non-profit sector have in encouraging the use of learning analytics?
Ryan Baker, Visiting Associate Professor, Teachers College, Columbia University
Alfred Essa, Vice President of Research & Development and Analytics, Digital Platform Group, McGraw-Hill Education
Philip Long, Chief Information Officer (retired), Yale University
Wendy Pradt Lougee, University Librarian and McKnight Presidential Professor, University of Minnesota-
Speaker 1: Wendy Pradt Lougee
Did library workflow analysis
Looked at content, access and services to see group and indiv data
Use the data on library users to push out recommended resources based on peers
Examined GPA and student library use
(13 types of library use)
APLUS System captures info on about 1000 students a day
Speaker 2: Alfred Essa
Looks at economic cost to public and the individual of 50% of college students dropping out
Using analytics to identify at-risk students
Poor and minority students tend to either not go to college or if they go to college they tend to be over-qualified for where they do go
Speaker 3: Ryan Baker
Learning analytics for predictions
Games, etc. Collect data on students
Information about student behavior while learning online can be mined to determine these factors and make adjustments to improve the learning mechanism: gaming system, off-task behavior, carelessness, inexplicable WTF behavior, boredom, frustration
Things can be predicted based on student behavior data
Q & A
Privacy – disclosure needed; students may be ok with data collection if it will be used to help them (intervention); private info must be stripped if data is published
most CMS’s are set up to collect data
Blackboard includes a learning module. Data on success of the course.
A lot of concern from audience about what is done with the data. Does university give it to someone else? What if outsiders are gathering data about our students? If we don’t gather data, others will. Coursera is growing fast because they collect lots of data.
VII. Closing Remarks
We are going from paper to people (e.g. MOOCs are videos of people; people sharing and getting interpersonal in the new media).
There shouldn’t be a sense that the new is in competition with the old
Leaders in libraries need to work with the new stuff the right way