Improving Access to Usage Statistics for Electronic Journals

 

The Yale University Library has moved toward its ultimate goal of integrating print and electronic resources in its collections.  This change in the Library’s collections has challenged traditional methods of gathering usage data.  The Library previously gathered all usage data internally, but now must depend on external vendors to supply data on usage of electronic collections.  This has sometimes been a difficult process, particularly in the lack of standardization among vendors. However, collecting the usage statistics allows the library to learn more about how electronic resources are used and how we can exploit their features to better serve our readers. This proposal seeks to enhance our ability to use and analyze electronic journal usage statistics by incorporating existing Library resources (e.g., the Ejournals database) and by providing a friendly Web front end to produce usage reports for individual selectors. 

 

Information from vendors is currently gathered and processed by Electronic Collections staff.  The data are transferred to Excel spreadsheets and are then mounted on the Web, often with charts showing the change in use over time.  (http://www.library.yale.edu/ecollections/access/UsageStats.html).  Many Yale librarians use this page and the data it contains to find usage data for a particular journal or vendor-supplied group of journals.  Usage data are currently not easily viewable in a way that would compare use across resources.  

 

This is particularly true for electronic journals.  Most electronic journals arrive as part of a package subscription from vendors and are not naturally grouped by subject. Our current presentation is not helpful in spotting usage trends relative to other resources or publishers.  A helpful tool would generate, given a specified subject area, a report detailing electronic journal usage in a certain subject at Yale.   This would be particularly useful when negotiating with vendors for electronic journal packages, which has become increasingly important as journal prices rise.   

 

We propose to take vendor and SFX usage data available for Yale subscribed journals and combine it with other data contained in the Ejournals database; i.e., subject classification, selector, ISSN and title.  By combining this information in a database we will create the ability to run subject-wide reports on electronic journals usage. Along with the data repository, a web front end will be designed that will allow the user to specify the desired data, time period, and output format (Web page or delimited text file).  An example of this type of front end for one set of usage statistics can be seen on the Cushing/Whitney Medical Library’s intranet, http://info.med.yale.edu/library/intranet/statistics/ovid/

 

Methodology

Journals will be selected for inclusion in this project based on the publisher’s adherence to established guidelines (such as the ICOLC guidelines and COUNTER).  Journal usage statistics must be provided on a monthly basis in an electronic format.  Data will be included based on their ability to be compared across vendors; that is data must be provided with clear definitions of how they are gathered, and those definitions must match across vendors.  Journal packages identified as candidates for this project include Elsevier, Highwire, Project Muse, Ebsco journals from Academic Search, JSTOR and journals we receive from Ovid.  This selection of several thousand journals will cover many disciplines and fields in humanities, social sciences, and sciences.  Data from the SFX server will be included based on requests to the SFX server for a particular journal. 

                       

Web and Workstation Support will create new tables for our project in the Ejournals database.  The Ejournals database includes the information that runs the electronic journals list and the databases lists on the Library Front Door.  We will then load the usage statistics for 2002 and 2003 directly into the database ourselves without bothering the systems office further. 

 

A significant portion of the time will be spent developing processes to load the data into the database in a consistent and efficient manner.  We will also train a student in the use of these processes and create new ones for resources we’d like to add to the database.   The student will also be responsible for documenting the process for staff that will take over this responsibility in the future.

 

The student will also help us create the Web front-end.  This Web form will allow the selector to specify journals in a particular subject area, a particular month (or months) and out format (html or a comma delimited text file). 

 

At the end of this project:

  1. DBOW will have been expanded to include fields for usage statistics from both SFX and vendors. 
  2. Journal usage data for the last two years (at minimum) from the selected publishers will be entered in the Ejournals database. 
  3. A report generator will exist that will allow a user to create a report by month for all journals in one or more subject areas.  The report will generate a Web report or a text file suitable for importing into Microsoft Excel or Access.  Data generated from this Web form will be compared to the original data for accuracy. 

 

Timeline

 

January-April:  Analyze electronic journal usage statistics, select statistics for inclusion in the project.  Set up new tables in Ejournals database

 

May-September:  With student help, establish procedures, import usage statistics, develop schedules for retrieving information, create web front-end.

 

Autumn:  Announce project, assess its success (see Project Assessment below). 

 

 

 

Budget

Student Labor (learning importing processes, creating documentation, helping design front end)

$11.00/ hr x 100 (10 hours a week for 10 weeks over the summer)

$1100

 

 

Benefit and Impact

As the Library moves into a time of large percentage increases in pricing for electronic journals, selectors may be faced with more difficult decisions as to where to spend Library collections funds. This prototype system will provide better access to journal statistics.  This may help us to negotiate more effectively with vendors over pricing and content of packages.

 

Even where cost is not an issue, it is increasingly important for us to assess the use of electronic resources and to find ways to measure use and the impact electronic resources have on scholarship and research.  This project will help Yale better determine if the data we receive from vendors is useful, or if we need to ask for or develop different data about electronic resources.