Document Delivery Task Force

 

Findings and Recommendations

 

Members: Sherri Jones (chair), Carol Brach, Thurston Miller, Carole Pilkinton, and Kathy Ryan-Zeugner

 

LEC Liaison: Nigel Butterwick

 

 

Massive journal cancellations, as well as the desire to enhance service to users, are driving many universities to implement document delivery projects. These projects vary from large scale services where the library fully subsidizes unlimited, unmediated article purchasing for all affiliated users to smaller scale pilot projects involving selected departments and/or selected user groups. Some libraries choose to offer a mediated service, where users’ requests are filtered through the ILL office or branch library before being filled via a commercial document supplier. Both the sending of the requests and the receiving of the articles are done through the ILL/branch staff.  Some libraries impose a fee while other libraries fully fund the service. With unmediated subsidized document delivery, the users’ requests are sent directly to the commercial provider and the receiving of them is done at their desktop or personal fax machine, thus bypassing the ILL/branch office totally. No fee is imposed for articles purchases. However, some libraries set limits to the service, such as the number of articles that can be purchased per user per semester or the maximum cost of the article purchased, or they will limit participation to selected groups. A variety of commercial vendors are being used, including CISTI, Ingenta, Infotrieve, and the British Library Inside service, to mention a few.

 

No One Size Fits All Method

 

In June 2004, a Document Delivery Task Force was created and charged with identifying the critical features required for a successful document delivery service and then selecting one (or more) commercial document delivery system(s) to use during a trial period to test the service. The task force is responsible for planning and implementing the trial (pilot) service including determining the length and time for the pilot project, the restrictions, if any, placed on the service, and patron eligibility for the service. The task force is also responsible for evaluating the service upon completion of the pilot.

 

The members of the Task Force have spent a great deal of time reading the journal literature on document delivery, talking to commercial vendor representatives, and examining practices at other institutions to determine the type of document delivery service that would best suit our needs. As a result of our investigation and after careful deliberations, The Task Force has concluded that there is no one perfect model for meeting document delivery needs across all disciplines nor is there one perfect vendor. Therefore, we are recommending the implementation of three different pilot projects, staggered at least one month apart, in order to reduce the amount of stress on staff members implementing the services and to allow for a smooth transition from one service to another. Also, offering a variety of services will allow us to better evaluate the services by making comparisons between them

 

 

Phase I

 

 Self Service (unmediated) library-funded commercial document delivery using CISTI Source

 

Purpose: To test the feasibility and user acceptance of library funded, self-service ordering of articles to enhance local ownership by providing users with speedy access to journal articles at the time of need.  Specifically, we hope to

 

  1. Improve service by extending timely journal access outside the present collection.
  2. Determine what a satisfactory delivery time is.
  3. Use the project as a way to measure user satisfaction with document delivery as an alternate to journal ownership.
  4. Study the feasibility and costs of a permanent subsidized, unmediated document delivery service

§         How much staff intervention is needed?

§         Are users willing to take the time to request articles?

  1. Evaluate previous cancellation decisions based on data collected from the pilot project.
  2. Determine if library-subsidized unmediated document delivery benefits some disciplines more than others.
  3. Deliver all documents to the desktop in electronic format.
  4. Compare mediated with unmediated document delivery by doing a cost/benefit analysis.

 

 

Duration: six months, to begin in February 2005 (or whenever Ariel software can be installed successfully at the Engineering and Life Sciences Library).

 

Limits:  No limits on the number of articles ordered; no maximum cost per article.  However, the library will not subsidize the purchase of articles from journals to which the library subscribes.

 

Participants: Faculty, post-docs, graduate students, and staff in the Department of Biological Sciences and all departments in the College of Engineering.

 

Participation is voluntary. However, since effective user training will be critical to the success of the pilot, class attendance or individual training will be required before a user can obtain an account password for ordering articles.

 

Participants will be asked to complete a survey at the end of the pilot. Survey questions might include the following:

 

  1. Would you anticipate using this service in the future if it is continued?
  2. Would you be willing to pay a small flat fee for this service in the future?
  3. How would you rate the document delivery service overall?
  4. How would you rate the delivery time from order to receipt of your article? Did the delivery time meet your expectations?
  5. How useful did you find the article(s) you ordered?
  6. Is this type of document delivery an adequate way of supplying articles from occasionally used or previously cancelled journals?
  7. Would you support cutting subscriptions to high cost/low use journals and substituting this service for those subscriptions as a library strategy to reduce ongoing serials and inflation costs?

 

 

Why limit participation?

 

In order to control expenditures and to discover if use patterns emerge across disciplines, it was decided to test the service on a select group of users, with the hopes of expanding to additional groups in the future. By limiting participation to a select number of groups, the potential for abuse can be gauged and the final decision about totally or partially subsidizing document delivery can be made a later date. Also, by limiting participation to selected groups, we can assure that each group has a library liaison to serve as a contact person to assist them with questions or problems.

 

While unmediated, self-service ordering of articles will be limited to selected participants, the Table of Contents/subject alerting service offered by CISTI and the ability to search the CISTI Source database will be freely available to the entire ND community.

 

Why limit participation to the sciences?

 

Based on our experience with journal usage and previous cancellation projects, we have speculated that faculty and students in the scientific disciplines, and to a slightly lesser extent, the social sciences, would benefit most from an unmediated commercial document delivery service.

 

Literature on the subject indicates that the sciences, including engineering and medicine, and the statistical-related social sciences depend heavily on journals as the primary mode of scholarly communication. Also, at Notre Dame, like at most other institutions, science, engineering, and social science journals cost more compared to humanities journals and have experienced the greatest inflation in periodical prices. As a result, it is in these areas that the library has been forced to cancel the most journal subscriptions.

 

Cited studies support the view that science users would benefit the most from an unmediated commercial document delivery service. A study involving a library-subsidized self service document delivery service at the University of Tennessee Libraries using CARL Uncover showed that “Uncover benefited users in the sciences and social sciences most as a collection development resource and had little impact on the humanities.”  Specifically, during the third year of their pilot, January through August 2000, a total of 858 articles were requested by faculty in the sciences, as compared to 300 in the social sciences and only 5 in the humanities. [1]

 

Why CISTI Source?

 

CISTI, the Canada Institute for Scientific and Technical Information, is one of the institutes of the National Research Council of Canada. It was mandated in 1929 to be the national science library for Canada, and as a result, has the strongest collection in the field of science, technology, and medicine in Canada and one of the strongest in the world. The collection is particularly strong in journal and conference proceedings.

 

Document Delivery is a major service for CISTI. The primary service is delivery of copies of articles from the CISTI collection and from the Canadian Agricultural Library (CAL). Items not held at CISTI or CAL may be obtained from CISTI’s partners, the British Library Document Supply Centre (BLDSC using their LINK service, or from any appropriate supplier in the world using their Global service. [2]

 

CISTI Source is an integrated current awareness Table of Contents (TOC) database and document ordering service, consisting of over 17,000 journal titles from 1993+. The coverage is primarily science, medicine, and technology literature as the database is produced by CISTI. (52 % Science, Technology, and Medicine; 21% Social Sciences; 9% economics and business; 5% humanities).  

 

Why unmediated?

 

Unmediated document delivery provides for speedier access to information because it eliminates the middle man and thus decreases processing and delivery time.  Ordering articles for themselves offers a convenience to those users who prefer to take charge of their own document delivery needs or are under severe time constraints. Furthermore, unmediated document delivery reduces ILL expenses and staff workload by transferring the ordering and receiving processes to the requestor. It can also soften the blow of journal cancellations by streamlining the process of obtaining articles. [3]

 

A benchmark study by New Mexico State University indicated that a majority of users expect delivery of articles within 48 hours. [4] Allowing users to order directly from vendors will guarantee a faster turnaround time than with interlibrary loan or a mediated document delivery service and should allow us to better meet users expectations for quick delivery.

 

What are the associated costs?

 

$2750 for a six month subscription ($5500 per year for annual subscription) to CISTI Source

§         Includes unlimited use of Table of Contents and subject alerts.

 

$2000 for integration of custom holdings (one time fee)-“Holdings matching” is critical to prevent unnecessary ordering of materials owned by the library.

 

 

Phase II

 

Self Service (unmediated) library-funded commercial document delivery using Ingenta Connect and Ingenta “Pay per View” service

 

 

Purpose: Same as above.

 

Duration: six months (to begin approximately one month after first pilot begins)

 

Limits:  No limits on the number of articles ordered; no maximum cost per article. However, the library will not subsidize the purchase of articles from journals that the library subscribes to.

 

Participation: Faculty, graduate students, and staff in the departments of Psychology and Life Sciences and all departments in the College of Business.

 

Participation is voluntary. However, since effective user training will be critical to the success of either pilot, class attendance or individual training will be required before a user can obtain an account password for ordering articles. Participants will be asked to complete a survey upon completion of the project.

 

Why are no humanities represented in the trial?

 

While Ingenta is a multidisciplinary database, coverage is primarily in the sciences and social sciences. Libraries offering document delivery services have found that users in the sciences and social sciences benefit from the service more than those in the humanities.

 

Why Ingenta?

 

Ingenta (formerly Carl Uncover) is an automated alerting service that makes available current tables of contents from a database of over 28,000 periodicals covering a wide variety of disciplines, delivered via fax or Ariel, and another 6000+ titles available electronically on a “Pay per view” basis.  Ingenta offers the most comprehensive collection of academic and professional publications, with over 16 million articles.

 

Like CISTI, Ingenta has the ability to set limits on cost per article or per user per semester to control expenditures. Also Ingenta provides gateway clients with monthly user statistic reports to track expenditures. Other features that we found attractive are:

 

§         Electronic subscription matching allows patrons to seamlessly access the electronic full text of the journals we subscribe to from a single location.

§         Institutional Holdings Matching prevents unnecessary ordering of articles by blocking orders on titles already owned.

§         User Profiles Management to match groups with various user rights and purchasing power.

 

Branch Library staff have been using Ingenta for several years to order articles for users and are familiar with the search interface.[5] Also, many users are already familiar with the Ingenta gateway from using it as an article database and for accessing ND’s online subscriptions.

 

What are the associated costs?

 

Annual gateway fee with holdings matching- $3570 for six months or $7140 per year.

 

§         Includes a 15% discount on all document delivery fees.

 

§         50 user license for Table of Contents/Alerting Service Allows for table of Contents and keyword search alerts for up to 50 patrons (additional users can be purchased).

 

§         “Holdings matching” is critical to prevent unnecessary ordering of materials owned by the library.

 

$1000 initial set up fee.

 

Data Collection and analysis:

 

Data from management reports received monthly from CISTI and Ingenta and from user surveys will be collected and analyzed. Each vendor will be evaluated on the following:

 

  • Usage patterns across disciplines and user status
  • Users’ perception of the value of the service across different disciplines
  • Comparison of the cost of document delivery versus the cost of journal subscriptions for same time period for each title ordered.
  • Document quality
  • Amount of staff intervention needed
  • Analysis of orders placed according to several measures:

 

§         Journals having five or more articles requested

§         Percentage of orders for articles from cancelled journals

§         Average cost per article

§         Average turnaround time for receipt of article

§         Degree to which turnaround time met users’ needs and expectations

§         Monthly expenditures

§         Number of orders per month

 

 

Phase III. Somewhere down the road….

 

Direct ordering using Open URL LINKING

 

As the final project, we hope to work with a commercial document supplier such as CISTI, Ingenta, or Infotrieve, to integrate direct article ordering into our online databases, using outbound and Open URL linking. By using Open URL linking, we could create direct order links to the commercial vendor from our Find Text databases, such as Sci-Finder Scholar, Biological Abstracts, Compendex, etc, or from our e-journal locator for selected titles.

 

Other Task Force Recommendations

  1. The science branches should continue the practice of decentralized, mediated, library-funded document delivery, using a variety of commercial vendors and publishers. Branch libraries will be encouraged to use CISTI Source and Ingenta to fill requests during the pilot period.

 

  1. Interlibrary Loan staff will begin using commercial document delivery to fill article requests for faculty and graduate students in Psychology, Anthropology, and all departments in the College of Business.

 

  1. A mechanism allowing subject bibliographers to use their university or endowment funds to purchase articles should be established.

 

  1. Install Ariel software (Receive only) at the branch libraries so that articles can be received electronically from commercial vendors.[6] Ariel delivery provides better quality copies than fax, especially for graphics and charts.

 

Ariel is a software program developed by the Research Libraries Group (RLG). Ariel sends scanned documents via FTP in TIFF (Tagged Image File Format), from one Ariel station to another. Recent versions can also send via e-mail from Ariel-to-Ariel, or from Ariel-to-patron.

 

  1. The library must continue to promote a variety of document delivery services in order to raise awareness at the point of need with faculty and students.

 

  1. The information gathered during the trial period should be used as another factor in deciding whether to cancel journal subscriptions and electronic journal packages.

 

 

 

Prepared by S. Jones, 1/26/05

 

 

 

 

 

 

 

 

 

 



[1] Roe, Jane, Shrode, Flora, and Rita Smith. “Across the Disciplines: Does Subsidized Document Delivery Meet the Challenges?” Collection Management 26(2):2001: 13-29.

[2] VanBuskirk, Mary and Diane-Helene Caouette. “Ariel in a High-Volume Environment: How CISTI has integrated Ariel into its Document Delivery Business.” Journal of Interlibrary Loan, Document Delivery, and Information Supply. 10, #4 (2000): 113-119.

[3] Haslam, Michaelyn and Eva Stowers. “Library-subsidized unmediated document delivery.” LRTS 45, #2 (2000): 80-89.

[4] Stabler, K. “Benchmarking interlibrary loan and document delivery services: lessons learned at New Mexico State University.” Journal of Interlibrary Loan, Document Delivery & Information Supply 12 #3 (2002): 57-73.

[5] From January 03 to September 04, library staff ordered a total of 619 articles from Ingenta for a total cost of $18470.26 or $29.83 per article.

 

[6] Four copies of Ariel 4.0 have been ordered. Priority should be given to installing the software at branch libraries participating in the pilots (Engineering, Life Sciences, and BIC).