UI Current LIS Clips: Electronic Resources

April 2003 - Compiled and annotated by Lynn Hanson

Commercial Electronic Resources - For this issue of UI Current LIS Clips, we focus only on commercial electronic resources, primarily databases and electronic journals. Libraries also deal with two other types of electronic resources: digital materials created from library holdings; and web-based gateways that link to selected internet resources in the public domain.

In This Issue:

1. Highlighting UIUC Research - Paula D. Watson
2. Electronic Resources, What are the Issues?
3. Introduction to Electronic Journals
4. Electronic Books
5. Selection
6. Consortial Purchasing: Pros and Cons
6a. Watch Out for the Big Deal
6b. Big Deals are Beneficial
7. Licensing Issues
7a. A Review of License Terms
7b. Informing Staff and Users of License Terms
8. Providing Access
8a. Access Via the OPAC
8b. Access Via Web Pages
8c. Linking to Full Text Resources
9. Staffing
10. Usage Statistics
10a. The ARL E-Metrics Project
10b. Issues: Privacy and Value
10c. Do-It-Yourself Statistics
11. Cancelling Print Subscriptions
11a. Creating a Policy for Cancellations
11b. Quality of Electronic Journals
12. Archiving
13. Roles in the Electronic Information Environment

Commercial organizations or products mentioned in this issue reflect the choices of the authors being summarized and do not constitute endorsements by the University of Illinois.


1. Highlighting UIUC Research - Paula D. Watson

Meeting the Challenges of Electronic Resources

By Paula D. Watson, Director of Electronic Information Services, University of Illinois at Urbana-Champaign Library.

Building and managing e-journal collections is a complicated and labor-intensive activity that isn't getting any easier despite dramatic progress on several important fronts.

Who's Minding the Store?
Library classifieds frequently carry several ads for "innovative, energetic" individuals to coordinate and support the acquisition, processing and maintenance of electronic resources. Typical requirements include a knowledge of the e-marketplace, emerging technologies, licensing, negotiation, and a willingness to work nights and weekends on the reference desk. Large institutions usually have several people formally assigned to e-resources acquisitions and management. Trouble shooting access problems continues to be a major headache.

Do We Still Need to Pay for Print?
Pricing models continue to evolve and, while annual increases have modified somewhat, maintaining e-journals continues to strain library budgets - especially now that both public and private institutions are feeling the pinch of an uncertain economy.

While many libraries now feel far more comfortable replacing print journals with e-versions, moving titles online still does not result in the savings many libraries need to keep pace with increasing demands in an environment of diminishing resources. And there is no definitive solution yet to the e-archiving problem.

Use-Based Pricing and Project COUNTER
Recognizing that the "Big Deal" may be going the way of the dinosaur, publishers (and many librarians) are increasingly drawn to use-based pricing. Tying costs to usage requires accurate measures. Project COUNTER, http://www.projectcounter.org/, a bold new initiative, has been launched to address the need for reliable and truly comparable vendor-produced statistics. Issuing its first-phase Code of Practice in January, COUNTER has kept to its aggressive development timetable, but much remains to be done before apples-to-apples comparisons of trusted data are routine.

More and More Access
OpenURL, Herbert Von de Sompel's brainchild, has dramatically expanded access capabilities. Link resolvers based on OpenURL technology allow citations to be tied to the appropriate full text and can initiate interlibrary loan or document delivery transactions or launch searches for related material. See: http://www.niso.org/committees/committee_ax.html

As libraries add more different types of resources to their online collections, helping users find what they need is posing new challenges. E-content trackers that can generate web-based journal lists, table-of-contents-based search utilities, and portals have all been developed with remarkable rapidity over the last couple of years to help libraries cope with an increasingly complex array of service issues.

While e-journal finding tools, link resolvers, and meta-search engines offer wonderful advances in access, they also add significantly to the cost of maintaining e-collections.

Read More
Read more by Paula Watson in her recently published "E-Journal Management: Acquisition and Control." Library Technology Reports, March/April, 2003. 80p.

Copies of the report can be ordered from: https://www.techsource.ala.org

From the Introduction
"The report provides librarians with an introduction to the knowledge needed to build, maintain, and manage e-journal collections. It is divided into three chapters:


2. Electronic Resources, What are the Issues?

Jewell, Timothy D. Selection and Presentation of Commercially Available Electronic Resources: Issues and Practices. Digital Library Federation and Council on Library and Information Resources. July 2001. 55p. http://www.clir.org/pubs/reports/pub99/contents.html

The author reports on a survey of 30 libraries which are members of the Digital Library Federation and the Association of Research Libraries.

Economic Pressures Lead to Consortial Purchasing
Library spending for electronic resources is increasing each year, and it is likely to accelerate as the volume of resources available increases. The main method libraries have used to deal with this economic pressure is consortial-based purchasing, which allows costs to individual libraries to decrease as the size of the group increases. Prices based on total student FTE for the consortia are common, with the higher the total FTE, the lower the cost to each library. Consortial purchasing also makes more extensive resources available to the smaller member libraries. However, consortial purchasing reduces local control and creates added complexity in managing the resources, especially as libraries can belong to several buying groups. These 'big deals' also take larger shares of the acquisitions budgets, leaving less money for purchasing individual titles and smaller packages from scholarly associations or university presses.

Operational Costs are Substantial
In addition to any subscription costs, operational costs for electronic resources can be substantial. Providing title-level access to electronic resources by cataloging the resources so they can be accessed via the library OPAC and/or maintaining lists of links to the resources via the library web site, requires substantial staff time. Libraries at Drexel found that the cost to obtain and organize electronic journals was higher than the costs for other formats. In addition, there are costs for interface design, application development, server capacity, workstations, connectivity, and staff time to provide user support.

Managing the Resources Takes Time
The libraries surveyed have invested heavily in creating workflow procedures to gather and manage the information and documentation involved in acquiring electronic resources. Especially important is establishing an efficient way for library staff to track the status of a resource while its acquisition is in process. Some libraries have created their own computerized systems for acquiring, managing and supporting electronic resources, and 13 systems are reviewed by the author.

Licensing Issues are Challenging
Some vendor licenses may restrict interlibrary loan, prevent archiving, provide no ongoing access rights to the resource, place severe restrictions on photocopying, some of which cannot be enforced, and fail to indemnify the library against 3rd party damage claims. Fortunately, vendors have become more responsive to libraries' concerns, perhaps due to pressure from the International Coalition of Library Consortia and its important "Statement of Current Perspective and Preferred Practices for the Selection and Purchase of Electronic Information" at http://www.library.yale.edu/consortia/2001currentpractices.htm. In addition to negotiating the licenses themselves, it is also a challenge to inform users of license restrictions, as they vary extensively, and some even require a user click-through page. In consortial purchases, license terms may even vary among the member libraries.

Web Presentation Strategies - OPAC or Web Site?
To alert users to the availability of electronic resources, libraries either catalog the items to present them in the OPAC and/or present them on library web pages. The relative merit of OPAC versus web page access is debated. Some libraries use metadata in the OPAC to create a database of electronic resource descriptions, which is then used to generate web pages. Some libraries create customized access methods based on the needs of user groups, and offer methods for users to set their own resource defaults on the web access pages.

Linking to Fulltext is Desired, but Takes Effort
Another challenge is the difficulty of providing links between records in abstracting and indexing databases and full text articles in electronic journals. Although several vendors now provide products and services to accomplish this task, it requires substantial staff time to set up and maintain, effort is duplicated when the library subscribes to multiple vendors, and links to local content cannot be presented.

Users Need Support
Patrons require extensive support to use electronic resources. Information must be kept up-to-date about which resources are available to which user groups and how to connect to the resources. Assistance with access problems must be provided, and there are many opportunities for technical failure as more reliance is placed on multiple remote vendors and linking.

Usage Statistics are Difficult to Obtain
Obtaining useful statistics on the usage of electronic resources is difficult. Vendors may provide no data, and for those that do, the data provided is rarely comparable among vendors. Some vendors are now trying to meet the International Coalition of Library Consortia (ICOLC) Guidelines for Statistical Measures of Usage of Web-Based Information Resources http://www.library.yale.edu/consortia/2001webstats.htm. Libraries or consortia with locally-loaded resources can develop their own data-gathering methods, although resources are not often locally-loaded due to the cost.


3. Introduction to Electronic Journals

Curtis, Donnelyn, Virginia M. Scheschy and Adolfo R. Tarango. "Introducing Electronic Journals." Developing and Managing Electronic Journal Collections. Neal-Schuman, 2000. p1-18.

What is an Electronic Journal?
An electronic journal should perhaps meet the AACR2 definition of a serial [a publication in any medium issued in successive parts bearing numerical or chronological designations and intended to be continued indefinitely], but in practice, it is not easy to determine if a publication is an electronic journal and libraries may define the term differently. Some of the varieties of publications to be considered include:

Faculty may be reluctant to accept electronic journals due to the perception that web content is of poor quality, although some electronic journals are as rigorously peer-reviewed as prestigious print journals.

Package Deals are Common

In most cases, electronic journals are offered by vendors as packages of multiple titles. A package subscription has the advantages of a lower price per title, access to more titles than a library is likely to subscribe to in print, one search interface for the lot, and one order process and license to negotiate. However, a package subscription has several disadvantages:

Can Aggregator Databases Function as Electronic Journals?
Abstracting and indexing (often called "aggregator") databases may include full text articles, so libraries want to use them as electronic journals are used. However, these databases usually do not include all the content from each publication that is indexed, and the content and publications indexed change frequently as contracts with publishers change. Vendors are now responding to libraries' interest in using these databases as electronic journals and are providing more complete content for each title.

No Single Source for All Electronic Journals
Every library would like access to a complete database of all the electronic journals available, but unfortunately, there is no such single source as yet. Instead, electronic journals are made available through a variety of channels. Electronic journal publishers which provide access to their own journals include:

Re-publishers which provide access to the electronic journals of other publishers include: Third parties providing access to journals include: Gateway services from subscription agents which provide access to journals include: Electronic journals may also be accessed via journal databases from database vendors, including: Finally, free electronic journals may be identified via the following web sites:

4. Electronic Books

In order to limit the length of the issue, we have not included summaries of articles on electronic books. However, the following two articles are both recent and useful:

  • Hyatt, Shirley and Lynn Silipigni Connaway. "Utilizing E-books to Enhance Digital Library Offerings." Ariadne Issue 33 2002. - The authors "describe the rationale and background of OCLC's acquisition of netLibrary, the information environment that is being pursued, and the resulting benefits that libraries may experience." http://www.ariadne.ac.uk/issue33/
  • Sawyer, Susan K. "Electronic Books: Their Definition, Usage and Role in Libraries." LIBRES Library and Information Science Research Electronic Journal v12 n2 2002. - An extensive literature review. http://libres.curtin.edu.au/libres12n2/Index.htm

5. Selection

Holleman, Curt. "Electronic Resources: Are Basic Criteria for the Selection of Materials Changing?" Library Trends v48 n4 2000 p694-710.

Jewell, Timothy D. "Selection Issues and Practices." Selection and Presentation of Commercially Available Electronic Resources: Issues and Practices. Digital Library Federation and Council on Library and Information Resources. July 2001. p4-13.

It is commonly said that due to the increase of electronic resources in libraries, collection development is now becoming merely collection management. Librarians are becoming interpreters not selectors, and selections are now made only on macro scale rather than in terms of individual titles.

Traditional Selection Criteria are Still Valid
Libraries are now pressured to abandon traditional selection principles. There are also pressures to abandon print, to stop collecting and respond to user demand, and to purchase vendor collections without regard for library needs. However, Richard Gardner's four criteria for selection of materials as originally published in 1981 (quality, relevancy, aesthetic/technical aspects, cost) are still valid today regarding electronic resources, although their meanings and context have changed.

Other criteria to consider in selecting electronic resources over print versions include:

Purchase By the Drink or With the Kitchen Sink?
Purchases may now follow the 'by the drink' model or the 'kitchen sink' model. The 'kitchen sink' model refers to the large packages of electronic journals or other resources that are sold as a 'big deal,' including both titles the library has selected as well as those that are not wanted. 'Kitchen sink' implies a worthless, unnecessary addition to a package; and this is true in that package deals commit libraries to purchasing whatever a vendor publishes regardless of price or quality. On the other hand, package deals make a large number of resources available to more users, who are not offended by having access to too many low quality journals.

The 'by the drink' model of purchasing is newer, in which libraries can purchase pay-per-view access to individual articles. A charge is incurred when the article is viewed by the user. It is often the case that articles can be purchased this way for the same cost as subscriptions, although this may not hold for future cost models, since vendors can be expected to change pricing to retain profits. Unlike the 'kitchen sink' model, with 'by the drink' arrangements, libraries do not purchase unneeded materials. However, because users make their own selections from the package of titles offered by the vendor, librarians are removed from the position of determining the information needs of the community. Equal access to information may also be compromised; since users can increase costs by unrestricted requesting, the status (for example: faculty or graduate student versus undergraduate) of the requestor is often used to determine if he/she can request articles automatically.

 


6. Consortial Purchasing: Pros and Cons

6a. Watch Out for the Big Deal

Frazier, Kenneth. "The Librarians' Dilemma: Contemplating the Costs of the 'Big Deal'" D-Lib Magazine v7 n3 2001 http://www.dlib.org/dlib/march01/frazier/03frazier.html

The author urges libraries to avoid making 'big deals,' large scale purchases of journal packages, with commercial publishers. These deals weaken the library collection with titles that would not normally be selected. In order to obtain those titles they truly want, libraries must accept the full package, including irrelevant or poor quality titles. Instead, it is possible to continue print subscriptions, select only electronic subscriptions for titles that are most needed, or perhaps provide document delivery for articles from other titles.

A big deal does create the short term benefit of providing access to a large number of resources. However, a big deal is dangerous in that it will "weaken the power of librarians and consumers to influence scholarly communication systems in the future. Librarians will lose the opportunity to shape the content or quality of journal literature through the selection process. Those who follow us will face the all-or-nothing choice of paying whatever publishers want or giving up an indispensable resource." In a big deal, libraries have few options to cancel individual subscriptions; their loyalty is thus ensured. With this loyalty in place, publishers will gain great power to control the market.


6b. Big Deals Are Beneficial

Scigliano, Marisa. "Consortium Purchases: Case Study for a Cost-Benefit Analysis." The Journal of Academic Librarianship v28 n6 2002 p393-399.

The author conducted a cost benefit analysis for the purchase of electronic subscriptions to the entire package of 29 Annual Reviews Online titles for the consortium, Ontario Council of University Libraries (19 members). Prior to the purchase, the total subscription cost of the consortium's 167 print and electronic subscriptions to various Annual Reviews titles was $70,063 Canadian. Only four libraries owned the complete series. The consortial purchase allowed all member libraries to subscribe to the entire series in electronic format for $72,962 Canadian and increased the number of subscriptions to 406.

An estimate of the value of time saved by patrons (faculty) due to the availability of the electronic format was calculated based on 15 minutes saved per article accessed electronically as compared to accessed in print format. Based on average faculty salaries, the savings in time was valued at $525,505 Canadian. The value of time saved by patrons is important because the purpose of the library is to serve teaching and research.

Additional savings in avoided interlibrary loan costs (since all libraries can now access all titles) as well as the costs of handling print versions could be expected, but were not calculated. In this cost benefit analysis, the cost savings did not affect the total outlay, but brought increased benefit to the consortium in terms of enhanced access to resources and savings in patron time.


7. Licensing Issues

7a. A Review of License Terms

Curtis, Donnelyn, Virginia M. Scheschy and Adolfo R. Tarango. "Licensing and Legal Issues." Developing and Managing Electronic Journal Collections. Neal-Schuman, 2000. p95-116.

Metz, Paul. "Principles of Selection for Electronic Resources." Library Trends v48 n4 2000 p711-728.

Most electronic resources are not purchased outright, but accessed for a period of time under terms governed by a license agreement with the publisher or vendor. It is essential that libraries review the license carefully during the acquisition process. The topics below are generally addressed by licenses and should be evaluated for clarity and to determine that they meet the needs of the library and its users.

If a license is unclear or does not meet the library's needs as presented, do not hesitate to suggest changes. Select one representative from your institution to negotiate with the vendor. Several organizations have prepared sample licenses and other guidance for libraries negotiating licenses for electronic resources. Especially useful are:

7b. Informing Staff and Users of License Terms

Cochenour, Donnice. "How Will They Know? Libraries' Responsibility to Inform Users of License Restrictions for Electronic Resources." Colorado Libraries v26 n4 2000 p45-6.

Emery, Jill and Renulfo Ramirez. "Tackling the Monolith: Licensing Management at the Consortial and Local Levels." The Serials Librarian v42 n3/4 2002 p275-280.

Managing Information for Library Staff
As a library subscribes to greater numbers of electronic resources, managing information about their license terms becomes more and more difficult. All staff must have ready access to the licenses. At the University of Texas, an LDAP (Lightweight Directory Access Protocol, an open source software) database was created containing license content, which was made available to staff on the library web site
via iPlanet. Copies of the licenses are now posted as HTML files, which are more readily updated than PDF files. Information on the acquisition status of the products is included and is the most heavily used component of the database. The records can be queried, and it is possible to copy and paste from the agreements so extracts can be attached to interlibrary loan materials. Creating the database was labor-intensive, but it has been worth the effort.

Informing Users
Users must also be informed about the terms of use for each resource. Vendors monitor usage, and in the event of violation, a library may be required to document how users are informed. Yale University, for example, provides a checklist of allowed uses, with links to some license agreements (in some cases, vendors do not allow license terms to be made public) http://www.library.yale.edu/journals/licensing.html. Although this checklist is useful, patrons may not take the time to link to this separate page to read terms.
Ideally, every user access point should pass through a method of informing users of the terms that cover that particular resource.


8. Providing Access

8a. Access Via the OPAC

Curtis, Donnelyn, Virginia M. Scheschy and Adolfo R. Tarango. "Cataloging and Access." Developing and Managing Electronic Journal Collections. Neal-Schuman, 2000. p145-192.

Martin, Charity K. and Paul S. Hoffman. "Do We Catalog or Not? How Research Libraries Provide Bibliographic Access to Electronic Journals in Aggregated Databases." The Serials Librarian v43 n1 2002 p61-77.

Riemer, John and Jina Choi Wakimoto. "Taming the Aggregators: Providing Access to Journals in Aggregator Databases." The Serials Librarian v42 n3/4 2002 p157-163.

Wakimoto, Jina Choi. "Utilization of a Set of Vendor-Supplied MARC Records to Provide Access to Journals in an Aggregator Database." The Serials Librarian v43 n1 2002 p79-95.

Title Access is Important
Libraries often seek to provide access by title to journals included in electronic journal collections. Similarly, they may also provide access by title aggregator databases, thus allowing the aggregator databases to function as electronic journal subscriptions. It has been debated whether libraries should provide title access via the OPAC or via a web list or both. A 1998 survey of 62 libraries found that 71% preferred to provide access via the OPAC.

Single Versus Separate Records for Electronic Formats
For libraries that catalog their electronic journal titles in the OPAC, a decision must be made to use either the single record approach or the separate record approach.

Using Vendor-Supplied Records for Titles
Some vendors now provide sets of electronic records representing the journal titles in their databases, which can be downloaded into a library OPAC. If the single record approach is used by the library receiving the records, the records must be altered and updated manually, which is time-consuming.
The Cal State Northridge libraries were the first to load vendor-produced records for electronic journals into a library OPAC. EBSCO provided the record set for its product, Academic Search Elite. The libraries used the single record approach, which is preferred by patrons because it returns only one record when a title is searched in the OPAC. However, the manual updating process proved to be too time-consuming, so the libraries returned to a separate record approach.


8b. Access Via Web Pages

Cohen, Laura B. and Matthew M. Calsada. "Web Accessible Databases for Electronic Resource Collections: A Case Study and its Implications." The Electronic Library v21 n1 2003 p31-38.

Jewell, Timothy D. "Web Presentation Strategies." Selection and Presentation of Commercially Available Electronic Resources: Issues and Practices. Digital Library Federation and Council on Library and Information Resources. July 2001. p18-21.

Metcalf, Cameron. "An Open Source Solution to Managing Electronic Journal Links with Database-Generated Web Pages." The Serials Librarian v43 n2 2002 p21-28.

Withers, Rob and Rob Casson. "Providing Web-Based Listings of Electronic Journal Titles the Low-Maintenance Way: Or, Automating Ourselves Out of a Job." The Serials Librarian v42 n3/4 2002 p249-254.

The relative merits of providing access to electronic resources via library OPACs versus web page access is debated. Web pages may be more familiar to users and they can be organized in a variety of ways to meet user needs; however, creating and maintaining the web pages requires substantial staff time because the titles available and their URLs change so frequently. Using a database to store the information about electronic resources and a web-based interface program to allow users to query the database is a solution many libraries are adopting. Static web pages of links to electronic resources require extensive staff time to maintain. An active web page solution reduces the staff maintenance burden and greatly increases the access points for users.

University of Albany Provides Active Web Pages
The University of Albany libraries found that use of electronic journals was very low, perhaps because these resources were presented on web pages separate from other electronic resources. An integrated search interface for all types of electronic resources was wanted. The libraries used Active Server Pages and Visual Basic Script to create a database and search interface for all electronic resources; these programs were chosen because they are reliable, scalable and have good response time. Three search types were offered: alpha browse by title, keyword and exact title (string) search, search by subject and/or resource type. A relational database was used so that multiple subject headings and document types could be assigned to individual titles. A web site was created for data maintenance so staff could edit the database contents from any location. Links are checked by writing the URLs from web pages and using the link-checking module in Web Trends Analysis Suite. The improved integrated search interface now receives high use. Future goals are to accommodate common misspellings or alternative names for titles, and to create a unified list of all print and electronic journals by providing links from the web list to catalog records, or by generating catalog records to the web list.

Miami University Provides Both OPAC and Web Access

At the Miami University libraries, electronic journal usage statistics were also very low. Access was provided via the Innopac OPAC via title, keyword and LCSH. To enhance access to the electronic journals, a web page was created by exporting data from the catalog using PHP (Hypertext Preprocessor) to create the interface and using Postgres to create the supporting database. Both are open source software programs. Data is extracted from the OPAC monthly to update the web pages. Because the libraries use the single record method of cataloging electronic journals, URL changes must be made manually. The use of electronic journals has now increased and subject specialists can select the titles for display in real time. In the future, links may be provided from the web pages to the OPAC so that holdings information can be readily accessed.


8c. Linking to Full Text Resources

Aaron, Amira, Jonathan Helmke and Eve Davis. "Articles, Articles Everywhere..But Where? And Does it Matter?" The Serials Librarian v42 n3/4 2002 p171-176.

Jewell, Timothy D. "Web Presentation Strategies." Selection and Presentation of Commercially Available Electronic Resources: Issues and Practices. Digital Library Federation and Council on Library and Information Resources. July 2001. p18-21.

Curtis, Donnelyn, Virginia M. Scheschy and Adolfo R. Tarango. "Introducing Electronic Journals." Developing and Managing Electronic Journal Collections. Neal-Schuman, 2000. p1-18.

Libraries want to provide users with easy access to full text articles by placing links in the records of A&I databases directly to the full text articles in electronic journals. Links are also desirable from the list of references at the end of articles to their full text sources. Several types of initiatives are underway to help meet this demand for linking directly to full text sources.


9. Staffing

Duranceau, Ellen Finnie and Cindy Hepfer. "Staffing for Electronic Resources Management: The Results of a Survey." Serials Review v28 n4 Winter 2002 p316-320.

Libraries Agree that Staffing is Inadequate
The authors surveyed 15 academic libraries to collect information about their staffing for management of electronic resources and found that staffing for electronic resources is at a crisis level. In the years 1997-2002, the number of electronic resources acquired on average by the libraries increased tenfold; however, the number of staff members added to manage the resources only doubled during that time period.

The libraries reported that they had adequate staffing for invoicing functions and systems support. However, the following task areas were understaffed in all the libraries:

Selecting and marketing electronic resources, collecting and analyzing usage statistics, and user instruction are additional task areas that were not included in this survey.

Staff Take on Extra Duties
Responsibility for these task areas had been assigned to staff members in addition to their existing duties. They are complex tasks and require additional expertise such as knowledge of the products, of library systems, of the campus network, and of the proxy server. Thus, savings in support staff time that might accrue by cancelling print subscriptions and avoiding the processing tasks associated with print, are simply transferred to the professional staff as more complex tasks. The libraries often distributed task areas among several staff members, creating a team approach.


10. Usage Statistics

10a. The ARL E-Metrics Project

Miller, Rush and Sherrie Schmidt. "E-Metrics: Measure for Electronic resources." http://www.arl.org/stats/newmeas/emetrics/miller-schmidt.pdf

ARL E-Metrics Project. http://www.arl.org/stats/newmeas/emetrics/contract00-01.html

Because electronic resources are expensive, it is essential that libraries be able to answer the question, "What difference does this tremendous outlay of resources make to the users of libraries?" Most electronic resources are maintained by vendors rather than loaded locally, where libraries could develop methods of collecting user statistics to help answer this question. This means that libraries are dependent on statistics provided by vendors.

Vendor-Supplied Statistics Are Problematic
Unfortunately, vendor statistics are problematic for libraries. Some vendors provide no statistics. Among those that do, the data provided is not comparable; for example, one vendor may count a search in five databases as one search while another counts the same query as five searches. The data elements provided may not be clearly defined and the method of collecting the data may not be explained. Data may be provided in various formats (email, web, printed) which requires staff time to collate. Vendors may also make errors and supply data from another institution, or technical problems may make data unavailable for certain periods. Extensive staff time is needed to organize vendor data to make it useful. And finally, it might be asked if vendor data is trustworthy rather than inflated.

ARL Study Creates Standard Data Elements
To address this situation, the Association of Research Libraries (ARL) initiated a study in which 24 ARL libraries as well as 12 vendors participated. The study addressed four areas:

To help vendors and libraries standardize their definitions of data elements, a field test was conducted using 18 data elements covering number and type of electronic resources, usage and activity, costs of providing the resources, and support costs. A procedures manual with data definitions, techniques for collecting standardized data, and an instruction module were developed as a result of the field test and are available at the URL listed above.

The next stage of the project is planned to recommend strategies for assessing the impact and value of electronic resources on user behavior and effectiveness in terms of outputs or outcomes.


10b. Issues: Privacy and Value

Okerson, Ann. "Are We There Yet? Online E-Resources Ten Years After." Library Trends v48 n4 2000 p671-693.

Tracking how patrons use electronic resources raises concerns about user privacy, which must be protected. Equally important is the issue of how to preserve resources that are little used and thus have the least economic value in the current market. These same resources may be of essential value to future readers. The value can be astronomical, for example, in the case of preliminary discovery of a key scientific principle. Using usage data to value information puts emphasis on its immediate use rather than future value.

Usage data also supports a change in emphasis, making electronic resources user-centered rather than collection-centered. Publishers monitor usage closely on their free sites to learn what users want. Libraries should also focus on their user needs and provide options for users to customize their own access to electronic resources.


10c. Do-It-Yourself Statistics

Cox, Fannie M. and Weiling Liu. "What's One to Do when Vendors, Publishes and Aggregators Do Not Meet Your Usage Reporting Needs? Do It Yourself!" The Serials Librarian v42 n3/4 2002 p223-228.

Duy, Joanna and Liwen Vaughan. "Usage Data for Electronic Resources: A Comparison Between Locally Collected and Vendor-Provided Statistics." Journal of Academic Librarianship v29 n1 2003 p16-22.

Are Vendor Statistics Inflated?
Individual libraries have created their own methods of obtaining usage statistics to supplement those supplied by vendors. The North Carolina State University libraries were interested in verifying if vendor statistics were realistic or inflated. They recorded data for a 12 month period from a representative sampling of electronic resources, based on user attempts to access those resources from the library web site. This method, of course, missed counting users who connected directly to the vendor sites via bookmarks.

The data collected showed a pattern of increases/decreases of usage similar to the pattern displayed by the vendor data. However, for some of the vendors the library data and the vendor data did not match in terms of values. Some vendor statistics (Elsevier, Emerald, Project Muse) were much higher than the library statistics, due to users accessing the resources at the title level rather than through the main vendor page. On the other hand, use of ProQuest was overcounted by the library due to differences in the way user time outs were treated. The study thus demonstrated that the vendor and library data showed the same pattern of use but not the same statistical values.

Gathering Additional Data
The University of Louisville libraries developed a program which gathered usage data for electronic journals by title, subject and vendor. This is significantly more information than can be obtained by counting user connections to the vendor home pages. The authors collected data using a CGI program adopted from AXS free software into a MicroSoft Access database. Their method provides reports organized in the following ways:


11. Cancelling Print Subscriptions

11a. Creating a Policy for Cancellations

Rupp-Serrano, Karen, Sarah Robbins and Danielle Cain. "Cancelling Print Serials in Favor of Electronic." Library Collections, Acquisitions and Technical Services v26 n4 2002 p369-378.

Publishers originally offered electronic journals at no or minimal cost if libraries also maintained a print subscription to the same title. Libraries happily accepted the electronic versions without thinking about their equivalence to print. Now that pricing options have become less advantageous, libraries must carefully evaluate the costs and benefits of maintaining subscriptions to one or both formats. The questions that were not asked in adding electronic journals now have to be asked in reverse as subscriptions are subtracted. Many libraries do not yet have written collection development policies to guide the cancellation of print subscriptions.

Based on a survey of 47 ARL libraries, the authors suggest the follow points to include in a policy regarding print subscription cancellations.

11b. Quality of Electronic Journals

Henebry, Carolyn and Ellen Safley. "Before You Cancel the Paper, Beware: All Electronic Journals in 2001 are NOT Created Equal." The Serials Librarian v42 n3/4 2002 p267-273.

At the University of Texas, Dallas, librarians were given a mandate to remove duplications in the journal collection. They reviewed sample pages from print and electronic versions of approximately 350 journal titles to determine if the quality and timeliness in the electronic versions were acceptable. When the electronic quality was acceptable, the print subscription was cancelled. If the publisher did not allow cancellation of the print version, the librarians determined if the print version should continue to be bound, or if it should be discarded. The print subscriptions were kept for all journals with inferior electronic versions.

Problems with Graphics and Content are Common
The librarians evaluated the electronic versions based on legibility of the text, timeliness, quality of graphics and color, and completeness of content. No problems were found with legibility of the text, but significant problems were found with the quality of graphics and the accuracy of the content. Some electronic versions included no graphics or only black and white graphics instead of color. Many problems were found with incomplete content; articles were missing, some issues were listed in incorrect chronological order, one publisher misrepresented the date that full text coverage was available, and some titles had no journal home page. Problems such as these create significant impact on reference and public service staff who must help users, and in increased requests for color photocopies or interlibrary loan to obtain color versions. The librarians notified the publishers of errors, but found that errors had not been corrected five months later. In evaluating electronic versions of journal titles, first determine which features are important to the library, report all errors to the publishers, and demand higher quality.


12. Archiving

Curtis, Donnelyn, Virginia M. Scheschy and Adolfo R. Tarango. "Archiving Electronic Journals." Developing and Managing Electronic Journal Collections. Neal-Schuman, 2000. p19-25.

Access to Back Files is Not Guaranteed
Unlike print journal subscriptions which a library owns and can access in perpetuity as physical volumes, there is no guarantee that the back files of electronic journal issues will be accessible in the future. When a library buys access to an electronic journal with a subscription over a time period, access to the journal on the publisher's server is often terminated when the subscription is cancelled. The library can no longer access the journal issues covered by the subscription period even though fees were paid for access to those issues. A publisher may offer access to back issues, but then go out of business or be purchased by another company which does not maintain the back files, and access is lost. Or perhaps the back files are maintained, but the service providing access to the files is no longer offered by the publisher; a library which has purchased copies of the back files must then provide its own access method.

Archiving Files is a Complex Task
To be maintained as a relatively secure archive, electronic journal files must be held in duplicate locations, be periodically refreshed, and, more importantly, be migrated to new formats as technology changes so they can be accessed by current software and hardware. These processes can be complex when there is interlinking between the files or to external web sites, file updates, multimedia components of the files, and data files requiring specific software.

Who Should Maintain the Archives?
Among libraries, library consortia, national libraries, utilities, publishers, and vendors, who should take responsibility for preserving archives? Libraries tend to believe that publishers should take the responsibility because they already produce and house the files. Publishers tend to believe that libraries should take the responsibility because libraries have always preserved the written record. The lack of a workable, cooperative plan for archiving electronic journals is the largest obstacle to libraries fully accepting electronic journals in place of print journals.

Some current experiments in electronic journal archiving projects include those listed below:


13. Roles in the Electronic Information Environment

Neie, Philipp and Heather Steele. "Infomediaries in the Internet Era: Subscription Agents as Intermediaries and Aggregators in the Electronic Publishing World - Agents of Change and Tradition." The Serials Librarian v42 n1/2 2002 p59-77. The authors are co-CEOs of Swets Blackwell, Inc.

New Features of the Electronic Information Landscape
Electronic resources exist in a constantly changing, complex environment involving new suppliers, new formats, new types of relationships, new platforms, new pricing and licensing models, new ways of buying and selling, consortia, increasing demand for individual articles and linking between document types, the end of the traditional journal and subscription model, and new processes for operations and systems. Every participant in this environment must try to identify the key trends and prepare for what will become important in the future.

Future Trends
Some of the future trends that might be expected include:

The Role of the Subscription Agent
The subscription agent was traditionally used by libraries to handle print journal subscriptions because they offered convenience and reduced costs. Agents now provide services for electronic journals such as:
New Services Could be Developed
Given the complex and changing nature of the electronic resources environment, there is a need for the following services, which might be developed and offered by subscription agents:

April 22, 2003