By Donald T. Hawkins (Freelance Conference Blogger and Editor)
Note: A summary of this article appeared in the September 2017 issue of Against The Grain v 29 # 4 on Page 69.
The 2017 Electronic Resources & Libraries (ER&L) Conference drew 950 attendees to the AT&T Executive Education and Conference Center in Austin, TX on April 2-5. There were also 431 online attendees and 76 exhibitors.
The opening keynote address by Anna Lauren Hoffman, postdoctoral scholar at the School of Information, University of California-Berkeley, was on data violence, which occurs when digital versions of ourselves (Hoffman called them “data doubles”) conflict with our physical identities in ways that have unjust outcomes or damage on one’s dignity or self-respect. No system can possibly do everything; algorithms can discriminate among various systems, but we have not done a good job of capturing respect and dignity. Lives have been constrained and shaped by violence, which has had material consequences in people’s lives.
It is important to distinguish between our nominal and algorithmic identity versus our moral and political identities. Nominal identities are attributed to us from without (check boxes on a form, etc.) and are generally for administrative purposes; the moral identities are how we make sense of ourselves in the world. One example of data violence occurred in the late 1930s when the Netherlands developed sophisticated record-keeping systems on people. In 1940, the Nazis found the data, and it became a war target. The data became a cause of physical conflict, and Jews in the Netherlands suffered a higher death rate than any other country. Systems like this show us the immense power that data controllers hold. Context matters and we must think about what happens when it changes.
Today, categorization penetrates our lives and influences how we socialize with each other. For example, Facebook puts data about us into categories and applies it in decisions about what one sees on the system. The system can even block users or close their accounts. Some systems have changed how they make decisions, which is promising.
Linked Data in Academic Libraries
Andrew Nagy, Director, SaaS Innovation, EBSCO Information Services, opened this session by describing some new tools that EBSCO provides for managing collections and giving control of them to users. They have recently partnered with OpenAthens, a platform to help users progress from discovering all of a library’s content to linking with it and accessing it via a single login. It is important to make sure that users find the library’s website and then find the tools they need, so that the user acquisition process is optimized.
Scott Anderson, Associate Professor and Information Systems Librarian, Millersville University, said that the Millersville library is totally electronic; it purchased only 25 physical items this year. They now have more electronic content than physical—250,000 e-books and 170,000 e-journals and conference proceedings. About 1.5% of their content is unique and 2.7% consists of sparsely held materials; the remainder of their general collections consist of materials that are widely held. For local materials, they work with Atlas Systems for specialized content and request handling, EBSCO to maintain their catalog, and Zepheira to manage a linked data infrastructure to pull data into their databases.
Anderson said that we must figure out a way to pull openly accessible linked data into our platform and make it accessible to our users. Users have no need to know the mechanics of the platform; if they are looking at an item they should be able to get it locally or request it. From discovery, we need to flatten access and make our content accessible, not only to students but on the open web, so that we get more usage of our holdings.
Jeff Penka, VP, Product Management, Zepheira and the Library.Link Network, continued the theme of democratizing access and use. He said that the information industry has been behind for a long time, and silos need to be broken down. Linked data will not happen overnight, but it is moving into production and solving problems. The Library.Link Network was introduced two years ago so that people do not need to worry about machines, standards, and networks. We must present information to the user in an understandable way, which gives libraries opportunities to tell their stories and leverage the power of their community. Access across silos is the key, so that the best of libraries and the Web is woven together. Part of the conversation must involve telling vendors how users found their resources.
Considering Collections as a Service
Glenn Bunton, Director of Services at the University of South Carolina, said that we must transition from building collections to curating access. From their inception, libraries were built to organize and preserve collections of materials. For example:
- The Library of Ashurbanipal organized and applied security principles to its collection of cuneiform writings and tablets,
- The Royal Library of Alexandria was the largest collection of Greek literature in the world, and
- The University of South Carolina Library was the first free-standing library in America and houses a historical collection of materials on that state.
When libraries are widely dispersed geographically, it might make sense to collect everything, but in a city like Boston that has over 50 colleges and universities, should each institution try to collect and preserve as much as possible? ILL now allows a library to say it does not have to collect everything, and e-journals allow a single copy to be used by many people at the same time. E-readers and smartphones have impacted how we think about collections; how much local content do we need to be able to serve users’ needs?
Budgets and space are fundamental problems and govern what we can do with our resources. And users’ expectations of what the library can provide are different today than they were 10 years ago. If they must understand how the library works, it is not doing its job, and if they must know what we know about a system, the system is not a good one. When users come to the library, they only care about themselves and solving their problems. Think about the underlying philosophy that is driving what you are doing and ask if it is best for your users. If we did not have people coming through our doors and using our materials, most of us would not exist and would just be warehouses.
Bunton applied Ranganathan’s Laws of Library Science to collections:
- Collections are for use; the value of a library’s collection is directly related to the degree it is used. Instead of tracking how many materials we have, we should be tracking how much they are being used.
- Collections must reflect their users. Determine your users’ needs and make them your driving force.
- Collections should save the time of their users. More is not better and not efficient. A student cannot work with over 500 databases!
- Collections must evolve as the library evolves. The sustainability of a collection is directly related to the degree it reflects the organization’s evolution. Many libraries weed to make space, but we need to think of weeding more as the natural evolution of a collection over time. Your user population will change; has your library followed suit?
If we take these Laws at some value, we will reach these conclusions:
- For most of us, our collections should be smaller than they are.
- Collections and collection development should be linked to public services and users, not technical services.
- Collection budgets should not be sacrosanct. It is hard to argue that it is reasonable to spend money on many things that are rarely used.
Libraries should take the following actions:
- Move from the collecting mindset to curating—selecting things specifically to meet the needs of our users.
- Keep urging vendors to unbundle their products so we do not need to buy things we do not need.
- Be places that exchange knowledge. (Note that in the mission statement of the MIT libraries the word “collection” does not appear.)
- Distinguish your library by its special collections and the services provided. All collections should be special collections.
Why Don’t I Have Access?
According to Jessie Copeland and Chris Palazzolo from the Emory University library, many users expect continued access to electronic resources after they leave the university, especially if they go to smaller institutions that do not have many resources and wish to continue their professional development. At Emory, retired and emeritus faculty members may keep their university email address, but that does not include access to resources.
Some vendors allow emeritus access and directly incorporate alumni into their license terms; others require further registration. Most vendor licenses to universities specify that usage of their content is for educational non-commercial uses. The geographic dispersion of alumni after they leave the campus is a significant challenge; it is hard to estimate the numbers of alumni or retirees and keep up with changes. To help solve this problem, the Emory Alumni Office has produced a list of databases accessible by alumni through a special portal separate from the general library portal.
Perpetual access to e-journals is a similar issue. Here is an edited question and answer session featuring speakers representing IGI Global (a small publisher) and EBSCO:
How does your company provide access for post-cancellation policies?
If a journal package is being cancelled or added, IGI writes an addendum to the lease saying they have perpetual access.
How does your company track perpetual/post-cancellation entitlements of your customers and ensure that access remains available?
IGI verifies with the libraries and adds an addendum to the license. Persistent URLs can be downloaded. EBSCO has a title information and publisher relations department, and is continually contacting publishers to get updated license data in customers’ files.
How many years of customer orders and/or invoices do you keep for their e-journal subscriptions and how can they be accessed by the customer? Can more years be made available?
IGI uses a system called Acumen that maintains data going back to 2006. Titles are provided on invoices in an Excel file. EBSCO has 5 years available in its subscription tool.
What systems does your company have to track movement of e-journals between publishers and platforms?
IGI has struggled with this question and tries to make the information clear on its website. EBSCO’s e-resources department tracks this information.
Do you indicate past and succeeding publisher and platform information? How do you share this information with libraries?
IGI has an e-journal alert that provides this information.
How does your company track and present information about title changes?
IGI has a column on its spreadsheet that lists all former titles of the journal. EBSCO has alerting that tracks the title changes and produces several reports.
Doing More with Your Data: How to Use Statistics
Rebecca Boughan, Electronic Resources Librarian, and J. Curtis Thacker, Director of Discovery Systems, both at Brigham Young University, discussed the use of statistics to improve library services, enhance collections, and impress your boss. They noted that we cannot always track the data that will give us the information we want; sometimes we get to the point where we have too much data. Large amounts of data can be difficult to query, and the basic tools like Excel and SQL cannot handle them. Visualizing the data is a solution to this problem because you can show much more information on a single visualization than in a table. For example, the data in these visualizations show that mobile usage has increased dramatically in the past five years, and it also increases at the beginning and end of a semester as papers become due and final exams approach. Conversions (the proportion of people who click to access something) are also higher at the end of the semester.
Photos Courtesy of J. Curtis Thacker
Using data such as this, one can track where conversions are occurring and they can be predicted. When the internet access on campus was down, it cost the library $800 a minute in lost journal access, and for every second longer a search took, 7% fewer conversions occurred. This data was used to make a case for the purchase of faster servers.
COUNTER statistics are useful for cancellation decisions, but care must be exercised in drawing conclusions from them. Custom reports showing the librarian’s name are useful in showing the data that applies to their searches: does a database need more promotion or should access to it be cancelled? Librarians need the big picture to make good decisions. Faculty usage and multidisciplinary aspects of databases can be mapped to disciplines and can be a factor in cancellation and retention decisions.
APPlatform for Discovery: Building a Unique Experience Through Discovery Apps
Platforms are environments for running other software. The network used to be the platform, and one could run an Internet connection on it. With today’s modern phones, the platform has shifted to the device. Apps on a platform add significant value, and it is easy to develop new ones, which is why we see millions of apps available today. An open platform lets anyone distribute apps and leads to increased functionality, which is where we are now with discovery platforms.
Lynne Grigsby, Manager, Library Applications and Publishing Group, University of California-Berkeley, said that undergraduates are their biggest users online but they check out the fewest items. They wanted the OPAC to look more like Amazon, so widgets were added to make it easier for users to find items on other sites. They also cleaned up the advanced search functionality because users were intimidated by it.
Sarah Stang, Web Services Librarian, Okanagan College, Kelowna, BC, described some of the unique circumstances her library faces because it serves a large geographical area and most people they serve will not go to the library. So with the help of EBSCO, changes in the search platform made the links more intuitive:
- It was not easy for students to progress from results to items, so icons were added to help them navigate.
- Evaluating print books was made easier using a Google Books preview, which allows a student to decide whether it is worth a trip to the library to obtain it.
- In-context access to related content was provided using Google Analytics. Links to style guides were added.
- Widgets down the side of the screen made it easy to access research help without needing to ask a librarian in person.
Eric Frierson, Director of Field Engineering, North America for EBSCO said that he has lots of games and apps on his phone, and discovery should have the same flexibility. Here are some of the apps that libraries have requested:
- Custom limiters so researchers do not need to know Boolean logic. One of the most frequently requested features is the ability to limit a search to printed books.
- Book series information. For libraries with an emphasis on leisure reading, this app provides a good way to scroll through the collection.
- Course reserves highlighting allows students to enter a professor’s name and get the reserves for a course.
- Digital archives. Libraries want to search them in different ways. (This is a standard feature of EBSCO’s EDS system.)
- Make EDS fun. An upcoming app will show movie trailers within EDS. The power of discovery will be expanded to allow results to be obtained even if a search term is misspelled.
Apps must be evaluated to ensure that they are meeting a need (there is an app for that!); one of EBSCO’s apps uses Google Analytics to show how often people click on results shown to them. Apps serve a unique role allowing EBSCO to respond to users’ needs very quickly; if many people use them, they are integrated into the EDS platform.
Securing Your Library’s License Legacy: Best Practices for Record Retention
Licenses are a necessary part of libraries’ access to electronic content, but in many libraries, retention of licenses is by benign neglect, and there are no established policies. In this session, Betsy Appleton from St. Edward’s University and Susan Davis from the University of Buffalo reviewed the legacy of licenses (how the agreements are stored and shared), who has access to them, and what are institutions’ retention policies. Some of the steps that must be followed are:
- Verify perpetual access,
- Verify titles covered by the license, and
- Maintain a history of the negotiations.
A survey of 250 academic libraries in 13 countries, mostly in the US and Canada, revealed that in about half of the cases, the e-resources librarian reviewed the license and half were reviewed by the library director. In about 2/3 of the libraries, the librarian signed the license. Most of the surveyed libraries did not have a specific license retention policy, but in practice most keep them “forever”. About ¼ of the libraries destroy old licenses as soon as they are superseded.
At the University of Buffalo, the licenses are managed by the E-resources Librarian, and the terms are integrated into their A-Z list of e-journals and e-books. Searchable scanned copies of the agreements are placed in the institutional repository. Superseded and cancelled licenses are retained if needed for title lists. St. Edward’s University follows similar practices.
Here are some best practices that were developed from the survey results:
- Whenever possible, the library should take the lead in license review and negotiation.
- Licenses should be accessible to more than one staff member and on more than one computer.
- Licenses should be retained indefinitely to verify title lists and perpetual access rights.
- A periodic review of all licenses should be planned, with attention given to changes over the past 5 to 10 years.
15 Student Data Secrets That Could Change Your Library
Nevada State College, a small undergraduate institution, has 3,700 students, many from low-income families. The college has the first digital library in Nevada; there are no print books in its library. One of the goals of the college is to become a national model for closing equity gaps in education, which led to its heavy reliance on data from EZProxy logs, ILL requests, library computer use, study room reservations, and research consultations. (These data capture most of the activity in the digital collection.) Tiffy LeMaistre, Electronic Resources and Data Services Librarian, described her data journey:
- 2014 was the “why year” of planning, research, and testing to see if there was a measurable connection between library use and students’ educational success, as measured by GPAs, graduation rates, etc.
- 2015 was the “whoa year” and the start of data collection. Positive feedback from initial analyses and the overwhelming amount of data in EZProxy logs led to mission creep.
- 2016 saw the first official research assessment and refinement of the data collection process.
- In 2017, the way forward is being charted as reflection and sharing results with the library community takes place.
From the initial years of library data collection, LeMaistre derived the following 15 principles governing the use of student-level data.
- There is no handbook for this kind of work, and there is not much data about how other people conducted a project of this type.
- Look beyond the library for help in managing the data, anonymizing it, etc.
- Start by reproducing other research, and don’t be hesitant to reinvent the wheel. Determine what questions people have asked in analyzing this type of data and think about whether they would be useful in your environment.
- Be willing to take a detour. Understand what resources students use and the uniqueness of your collection.
- Brace yourself—privacy is coming! The library has become an important voice in data collection efforts and can contribute significantly to what we know about privacy.
- Be transparent with your users. If you will be collecting student data, let them know and give them the option to opt out of the study. Use your marketing skills and apply them to your data collection efforts.
- Encryption is a good place to start, but it is not the end in protecting the data. There are many open source encryption tools available, and your IT department should be able to help you.
- There is no such thing as too much data.
- Do not create backups. There might be a temptation to save the raw data, but you must protect the privacy of your users.
- Anonymize IP addresses and student IDs as early as possible in the data collection process.
- Aggregate the data into groups of at least 10 students before you share it so that a student cannot be identified from the data set.
- Continue to improve the process. Keep thinking about what you are doing with the data, how it is being protected, and what new technology has become available.
- Tell everyone about your results once you are confident about your processes.
- Correlation still has value and is persuasive to campus administrators.
- Visualization matters; people remember visual presentations.
The results of the data survey were that library users are significantly more likely to earn a higher GPA, maintain a good academic standing, and be retained in their jobs.
Evidently Rising: New Providers, Models, and Lessons in Evidence-Based Acquisitions of E-Books
A panel of 5 speakers organized by Harold Colson, International Relations Librarian and E-Books Coordinator, University of California-San Diego, discussed evidence-based models (EBMs) for acquiring e-books and informing administrators. Josh Petrusa from Butler University said that a small institution like Butler must cover many subjects with a small budget. They started a patron-driven acquisition (PDA) program in 2016 and studied usage data to determine what users wanted. About a quarter of the monograph budget was quickly spent on acquiring titles in an evidence-based selection project (1,700 e-books were purchased from a single publisher—Taylor & Francis). Initial usage was disappointingly low because of only a relatively few titles in the collection, but recent purchases are finding higher usage as time progresses.
Arielle Lomness, Collections Librarian, University of British Columbia (UBC)-Okanagan, noted that 80% of their budget goes to US vendors or publishers. The budget was flat for many years, but recent exchange rate changes have had a major negative impact. So the library had to consider how it was buying e-books, and an EBM was launched in 2015 for purchase of social sciences and humanities content from Taylor & Francis and extended for 20 months. As with Butler’s experience, early low usage caused anxiety, and many imprints for the same title caused confusion. A second program with Cambridge Books was started and extended for 12-monts. One lesson learned was to ask for usage data from publishers early in the project because the longer duration projects are better for seeing accurate usage. For the future, EBMs will be continued.
Lynn Wiley, Head of Acquisitions at University of Illinois, Urbana-Champaign (UIUC), described how her large research library tried to get DRM-free e-books. They have about a million e-books and a long list of items they want. They buy using several models: direct from the publisher with no user limits and perpetual ownership, by subject collection and e-book series subscriptions, and DDA programs on DRM platforms with user and print limits and minimal downloading. Evidence-based acquisition has the advantages of being strategic, data driven, and user focused. UIUC signed up in 2016 to participate in two pilot programs: one with Project MUSE to obtain access to all of its university press e-books and another with Oxford University Press, both of which are still ongoing. UIUC’s experiences are similar to those of Butler and UBC: low initial use that grows slowly as users discover the e-books. Wiley’s opinion is that evidence-based acquisitions work and provide an opportunity to experiment with various prices and subjects to build a model that works for her library.
Melanie Schaffner, Director of Sales and Marketing, Project MUSE, noted that for some small publishers, they are the only distributor. Project MUSE has content from 240 non-profit scholarly publishers who provide content from more than 600 journals. Over 100 scholarly presses have added over 49,000 books to the platform, and nearly 3,000 customers deliver MUSE content to their users. Books on MUSE are DRM-free and provide unlimited simultaneous usage, downloading, and printing. They are searchable at the chapter level, and no special reader is needed. ILL is allowed for books that are purchased. MUSE has launched an evidence-based acquisition pilot program in response to customer requests and to demonstrate revenue potential to publishers. So far, 40 institutions are participating, and publishers’ fears of lower revenues appear to have been unfounded; 90% of them have benefitted. The content is being discovered and bought; the most expensive book purchased cost $2,400, and the oldest book used was published in 1920. Lessons learned:
- EBA is labor-intensive and uses a lot of resources.
- Publishers can change prices at any time and can add or delete single titles, which is confusing.
- It is important to ensure that benefits accrue to every publisher participating in an EBA program.
Freely Available Articles from Gold, Green, Rogue, and Pirated Sources: How do Library Knowledge Bases Stack Up?
Michael Levine-Clark, Dean and Director, University of Denver Libraries, reported on a recent study of free access to journal articles. He began with the well-known fact that many information users do not start their searches with the library but use Google and Google Scholar. Levine-Clark’s study looked at 300 articles indexed in Scopus (100 each from 3 broad disciplines), that were randomly selected and checked to see where they were accessible (accessibility was defined as being available within 3 clicks). Access definitions and results were:
- “Gold” OA: Open access on the publisher’s website (26%).
- “Green” OA: Open access in a repository or on an author’s website (20%).
- “Rogue” OA: Freely available via an academic social network such as ResearchGate (37%).
- Pirated: Freely available on Sci-Hub (87%).
Only 39% of the content originated in a library. Many people are downloading articles from Sci-Hub, even though they are at large universities where they should have access to the content in more legitimate ways. Most Sci-Hub users do not think it is wrong to download pirated papers because of convenience or objection to publishers’ profits. Levine-Clark’s conclusions are:
- It is hard to follow the rules!
- Starting a search with Google Scholar and supplementing it with Google gives slightly better results than starting with ResearchGate.
- Starting a search with Sci-Hub gives the best results; therefore publishers should be very concerned.
Levine-Clark concluded his presentation with a proposition: All content, both OA and licensed, should be discoverable through library systems. That combination should rival what is available through Sci-Hub.
John McDonald, Associate Dean for Collections, University of Southern California, described an extension to Levine-Clark’s study which examined how well libraries and their vendors provide access to Gold OA articles through their discovery systems. 144 libraries and 8 publishers were invited to participate, and 320 articles (no more than one per journal) were examined, of which 270 were Gold OA. They were tested for indexing in the discovery services and access to the full text at 40 North American libraries to discover whether library discovery systems index and provide access to the full text of as many Gold OA articles as Google Scholar does. Articles not found were then checked in aggregators’ systems. Here are the general conclusions and test results:
- Indexing of the articles in discovery systems is reasonably good; 85-100% of the articles were indexed.
- 50-90% of the articles were accessible from library discovery systems.
- Smaller libraries do not have as many articles available on their discovery services as large libraries do, which could be due to a number of factors.
- For major publishers, indexing is fairly consistent across libraries.
- We should try to leverage DOI links more.
- Some small and medium sized schools have done a very good job of improving low rates of access.
- We can support the OA movement by encouraging usage.
Regaining Control During Vendor Platform Changes
Kim Maxwell, Electronic Resources Management Librarian, MIT Libraries, and Angela Sidman, Electronic Resources Librarian, Yale University, have both been through vendor platform changes and discussed some of the problems they encountered. Migration on content and publisher platforms is a unique challenge for an e-resources librarian because although they are responsible for the systems installed at their institution, they do not really control them. When a migration goes badly, poor PR results for both the library and the publisher. Migrations are difficult and are no small job. They entail a tremendous amount of work by the publishers because large volumes of content are moving from one place to another. It is very hard to do a migration without something breaking.
Good information from vendors ahead of the migration is appreciated by librarians. A good migration guide includes a timeline, changes in access and content, potential changes in administrative and user accounts, and marketing and training tools that can be shared with public services staff. The guide should be publicized on all relevant listservs. Libraries should respond by tracking changes and responsibilities on a migration checklist, keep communicating, and support their staff members by connecting them to training materials.
All parties in a migration program can improve by:
- Communicating well and working together to ensure a smooth user experience. Publishers should post known problems on their websites. Advisory Boards can be used to advantage.
- Making migration guides living documents and publishing updates as they happen.
- Providing a preview environment where librarians can test changes and be partners in the migration.
- Acknowledging that migrations will continue to occur, so amending the licenses as appropriate.
- Making the migration widely known to subscription agents so they can inform their customers about the changes.
Impact Analytics: Measuring and Driving Meaningful Use of Electronic Resources
Do your collection development policies specifically address the growing area of streaming video? Jesse Koennecke, Director, Acquisitions and E-Resource Licensing Services at Cornell University, noted that more video content is becoming available, there are more vendors, and we are spending a lot of money on it. Streaming video is being purchased by academic libraries for entertainment, collection building, and course use. Many entertainment videos are also used for courses; they are either purchased directly or ripped from DVDs and hosted on a streaming server. Areas to be considered are the potential user base for the video, areas where deeper collections are needed, and costs. Cornell started a PDA program for videos in 2015, and it generated a lot of use: 10,793 uses (278,000 minutes) over 2,870 titles, of which 7,471 uses triggered licenses for 529 titles.
Kanopy and Alexander Street, have set up new analytic models to allow their users to see what their usage is, where it comes from, referral URLs, etc., which leads to different ways of thinking about usage, especially counting embeds when a user embeds a video on their site. Alexander Street can show “curated views” beyond simple counts, such as the average percent of a video that was played, which helps to determine how important it was to the viewers and whether a subscription to it should be purchased or continued. A collections strategy is being developed for long-term video usage at Cornell; questions to be answered include how video should be incorporated into the collection development policy, how it should be funded, and different needs for collection building and course use.
Andrea Eastman-Mullins, COO of Alexander Street, said that we do not have any standards yet for determining how important a subject is; COUNTER remains the only way to compare usage, but it does not give title information. So they created an “impact statistics portal” to show the subject area of the video, paging reports, engagement (which goes beyond watching the video), clips created, play lists, etc. They also ask end users to provide feedback, rate videos, and indicate how they were used. Usage will determine what the standards will be in the future; opportunities include expanding beyond video, partnering with other libraries, and creating heat maps to show the most relevant videos.
Boaz Nadav-Manes, Director of Technical Services at Brown University, said that the library is like an organism, and we must standardize assessment beyond statistics. What do users actually get out of an engagement? We must be satisfied that we have a better understanding of the research process, and we need more statistics about groups of researchers working together because they have a different impact on usage than one person watching something. Information from different platforms should be put into a tool to measure engagement.
Explore the Hidden Cache of Statistics at Your Library: Data Mining and Visualization Techniques for Collection Development and Assessment
Librarians are spending increasing amounts of time working with large sets of data, but according to Stephanie Hess, E-Resources Librarian at Binghamton University (SUNY), we have not moved along to the analysis part. Collecting is only part of the battle in winning financial support; we need to provide evidence and make it shine. Hess suggested that data should be presented in exciting formats, and visualization can help convey complex data. She quoted Information Dashboard Design (Analytics Press, 2013) by Stephen Few, a data visualization expert who said,
“A dashboard is a visual display of the most important information needed to achieve one or more objectives; consolidated and arranged on a single screen so the information can be monitored at a glance.”
To create an appealing dashboard, frame your questions, parse the data, and select appropriate tools. Hess’s library created an Access database to analyze electronic resource data, but this just created another silo, and many reports were not being used. But when Tableau was used to analyze the data visually, the outliers became readily apparent. Forecasts of potential usage can be made and vendor reports can be incorporated into the analysis process. OverviewDocs is a useful tool for visualizing documents.
Monica Bulger, who leads the Enabling Connected Learning initiative at the Data & Society Research Institute, presented the closing keynote on “Fake News, Reliability and Questioning: A Researchers Struggle to Navigate the New Information Landscape”. She began with the observation that information is social, dynamic, and depends on us to give it life. Our minds are working against us in this new information environment because when they get overloaded, they start to function automatically. It is important to realize that our minds are not objective recorders of information; we are using our prior experiences and what we already know to make sense of events. We also do information status slicing; getting fast results is better than perfect ones because most of the time we have competing demands, which we do not like.
Not only are our minds working against us, but so are advertisers. They understand the psychology of engagement, and their programs are designed to keep us engaged. How much do we record when information gets overwhelming? We need to figure out our sources and whom we can trust. Check confirmation biases in real time, and step back to see both sides of a story. Once we are trying for more clicks or more links, how does that impact someone’s view of the information? We need to empower people to be critical consumers, sort through information, and find the truth.
What are the criteria of reliability? Are we teaching people to question everything? Sometimes we must turn off our feeds of news, etc. because there is too much to be engaged with. We cannot make sense of everything; choose your issue and focus on what you will worry about. Crowds can have a silencing effect, which is the dark side of the internet. Our minds like things that are easy, and our brains like the increasing personalization of content, so when fake news came, we were conditioned to receive it.
Recordings of many of the presentations are available on the ER&L website. The 2018 ER&L Conference will return to the AT&T Conference Center in Austin on March 4-7.
Donald T. Hawkins is an information industry freelance writer based in Pennsylvania. In addition to blogging and writing about conferences for Against the Grain, he blogs the Computers in Libraries and Internet Librarian conferences for Information Today, Inc. (ITI) and maintains the Conference Calendar on the ITI website (http://www.infotoday.com/calendar.asp). He is the Editor of Personal Archiving: Preserving Our Digital Heritage, (Information Today, 2013) and Co-Editor of Public Knowledge: Access and Benefits (Information Today, 2016). He holds a Ph.D. degree from the University of California, Berkeley and has worked in the online information industry for over 45 years.