This article has been reproduced in a new format and may be missing content or contain faulty links. Contact zsrweb@wfu.edu to report an issue.

More live blogging…

  • MPS Technologies provides electronic resources in the publishing market.
  • Librarians said they wanted a portal with all stats in one place, that’s self-generating.
  • This is Scholarly Stats.
  • First thing they did in development was to send out an email to a list in the US and UK asking folks to help develop project.
  • Questions included: “How many of you are collecting usage data as frequently as you’d like?” “How easy is it comparing usage across multiple platforms?” “How often would you like to collect usage data?”
  • Several said “every month” in academia. Want to follow peaks and valleys in usage to understand marketing, find out the bang for buck for each database, etc.
  • Explained basic development process…reported some findings:
  • Stats are playing a big roll in subscription decisions, next most in justifying expenses. Much lower was “reporting” and “other” (which includes promoting, marketing, training, administrative, strategic planning, curiosity, etc).
  • Wanted to combine vendor stats, other data gathered from web logs, consortia reports, etc, fun code/subject, cost, etc.
  • Found biggest challenge from lack of standards, takes too much time, COUNTER standards help but there are some drawbacks.
  • Most useful statistics are: number of full-text downloads (not necessarily views), number of searches, number of sessions, COUNTER statistics, number of turnaways, others.
  • Tools that are out there now: ERMS (Innovative, ExLibris, Serials Solution), Thomson Scientific, ScholarlyStats, Project COUNTER, SUSHI, UKSG Usage Factor
  • On the ScholarlyStats site, they generate monthly reports available on your portal: Full text use on month-by-month and on a calendar year. Number of searches and sessions from all databases. Number of turnaways by database (for concurrent usage databases, not because of a technology issue on their end). Searches and Sessions for entire system.
  • Reports come in excel file and csv format. In the future, maybe xml, too.
  • On the ScholarlyStats site, they generate monthly dashboards reports available on your portal: most used titles by platform, least used titles by platform, etc.
  • Going forward, will allow you to save up to five years of data. (Not retrospective.)
  • There is no standard for consortia reporting by institution. EBSCO, ProQuest do a good job of it, but a lot of databases done. Since not all do, ScholarlyStats can’t.
  • How does this differ from Serials Solution’s option: SS gathers from federated search engine, but not from independent search interfaces.
  • Breaks out the same journal from various resources so you can see which is getting most usage.
  • Can generate a cost per use or search with this data.
  • Federated search searches each database, when you pick an item and go to it, that’s when it gets counted in the statistics.
  • Rank and list from most to least use, percentage of that use compared to the rest of what you’ve asked to collect, etc (long tail OR 80/20 rule thinking)
  • Can get a discount through SOLINET.