E-Journals Publishers Roundtable Ann Jensen Morna Conway (Moderator) In an attempt to determine what publishers provide now, what their future plans are, and what librarians want and need in the way of usage statistics, each speaker was asked to address the same broad topic of usage statistics for electronic products. Representatives from IOPP, AIP, SIAM, IEE, Turpion, ScienceDirect, and ISI spoke and interacted with the audience on issues related to electronic usage statistics. Tony O'Rourke (IOPP), Douglas LaFrenier (AIP) and Mary Rose Muccie (SIAM) all reported that their organizations gather usage data for two major reasons; to meet subscriber demands and to verify the strength of their journals to their edtiors, authors and subscribers. Most count full-text downloads only and provide statistics either via password, or e-mail to registered users. They each recognize that publishers statistics are incomplete since they do not capture usage from third-party providers. Statistics can also be skewed by the actions of robotic harvesters. Jim Ashling reported that IEE is in the planning stage only for usage statistics. He expressed concerned about how widely such statistics should be shared, since there are different privacy statutes in the UK and USA. He is also concerned about the incompleteness of such statistics, since they do not reflect use by third party services (INSEPC, Crossref, etc). Lev Malov of Turpion, a publisher of translations from major Russian journals in mathematics and physics, demonstrated Turpion's web-based usage statistics. He too, mentioned concern about their use, since they do not reflect all usage from aggregators and other higher volume users. True statistics would have to include downloads from all providers. Turpion's statistics currently show date ranges, numbers of fulltext downloads, abstracts, table of contents, and searches, in a user created profile on line. Amanda Spiteri, from Elsevier ScienceDirect, also voiced the need for the creation of industry standards in what she terms e-metrics - the counting and reporting of statistics related to electronic usage. ScienceDirect currently has a team of 5 creating usage reports from what is an enormous database of statistics. They offer 12 standard reports, with password access, over a 13-month floating window. Keith MacGregor reported that ISI started compiling usage statistics in 1997. They provide site-wide summaries, include number and duration of queries, maximum users, turn-aways, at the IP address level. Statistical reports are delivered via e-mail, not web access. These are flat files which users can manipulate. Librarians said it would be helpful to know time of day when turn-aways occur. Publishers need to know how librarians would like this data reported. Many different filters can be applied. There is currently no comparability among publishers, so all expressed concern about how usage statistics can be used erroneously. Useful statistical enhancements requested by librarians included: All agreed that standardization in statistics gathering and reporting is essential, otherwise the numbers cannot be used meaningfully. Usage statistics should not be used in isolation, as they are only one measure of relative value of titles. Several of those present are actively involved in standards organizations to try to create meaningful cross national standards for reporting statistics. Requiring all vendors to speak to the same topic made for a highly constructive conversation about needs, wishes and publisher capabilities, where issues common to all in the room were raised and discussed. This was a very worthwhile session!
ajensen@library.berkeley.edu
