Final report

Improving Access to Usage Statistics for Electronic Journals

Kathleen Bauer, Karen Reardon and Jennifer Weintraub

Goals of the Project

  1. The Yale Library e-journals database (YELMO) will be expanded to include fields for usage statistics from both SFX and vendors.  (accomplished)
  2. Journal usage data for the last two years (at minimum) from the selected publishers will be entered in YELMO.  (accomplished)
  3. A report generator will be written 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.  (partially accomplished, subject areas and keyword searching is still being worked on)

Problem Addressed

Historically, there have been barriers for selectors who wanted to use electronic journal usage data.  These data come in a wide variety of not always compatible forms.  This wide variety of data makes it difficult for librarians to know which specific pieces of data to use and how to compare data from different vendors.  Also, given a journal title, librarians must locate the vendor supplied usage data, download it, and then input the data into an Excel spreadsheets.  Selectors who would go to all this trouble still could not generate reports for journals in a particular subject area.

Methodology:

Katie and Jen reviewed all the data available from a set of our major vendor publishers.  These publishers provided many different types of data, including articles downloaded, searches, turnaways, PDF versus HTML, and web hits.  The one piece of data in common to all the vendors was articles downloaded, and this was the data point selected to be downloaded to the new table in YELMO. 

Katie and Jen met with Karen Reardon and described what we wanted and gave her an example of some raw journal statistics.  Karen developed a new table for YELMO database and found a way to provide cross-tab display of the statistics as well as adding the statistics together.  To accomplish this, usage statistics had to be formatted in a particular way in an Excel spreadsheet. 

Jen and Katie then created procedures for each vendor and tested them.  Finally, we hired a student who formatted all the statistics.  After the spreadsheets were created the student uploaded them into DBOW using an ODBC connection. 

Budget

$1000 was used to hire a student for eight hours per week to download data, input data in Excel, and then export to the database. 


Results

Over the last year, we entered usage data from eight journal vendors (Blackwell, Highwire, Kluwer, Muse, Nature, Oxford, Science Direct and Wiley) and SFX into a new table in the DBOW database.  The database report generator written for this project allows any selector to look at the journals by publisher and fiscal year.  In the next phase of the project we will implement a keyword search.  The keyword search will allow a selector to look, for example, at all the journals with the word “history” in the title or subject heading, across all vendors. 

We will continue to add new usage statistics to this database. As part of the training of the student who downloaded usage data into DBOW, we now have written instructions in place and procedures for inputting the monthly usage data into the DBOW database.   In addition, if Yale does purchase an Electronic Resource Management software package, we will know from this experience what features we are looking for in choosing how to collect, manage, and present statistics about journals. 

The report generator can be accessed http://resources.library.yale.edu/online/statsdb.asp

Issues Remaining

  • We were not able to include two important vendors in this project.  The first vendor, Ovid, does not provide ISSNs with their statistics, just journal abbreviations that are not as reliable.  The second, Ebsco, provides some good and some poor ISSNs.  Further work may make Ebsco statistics more useful.  Without good ISSNs it is not possible to compare across vendors.
  • Vendors are increasingly providing some statistics in COUNTER format, which will be helpful in the future.  The ability to dump COUNTER statistics into this database with just a few changes would speed up the maintenance and management of statistics.
  • We would like to expand this database and report generator to include database statistics in the future.
  • The use of this report generator and the usage statistics will have implications for any implementation of an Electronic Resource Management software package.  Selector reaction to this report generator will help guide us in our requests to the ERM vendor for report capabilities. 
  • It would be most useful if the usage data could be combined with cost information so that librarians and administrators could easily check cost per use for individual titles, subject areas, and vendor packages.  This is not yet available.