I read the article as well and was hoping we could kick this around on the list.  The article has some great points. I do detect a subtle cautionary tale about canceling journals or Big Deals based on naïve interpretations of cost-per-use calculations. Cui bono?

 

Here’s what I pulled out of the article:

 

·        Rigid standards of cost-per-use: applied across the board without nuance, are not useful. Agreed.

·        HTML/PDF Shenanigans: it’s interesting to learn that some publisher platforms will automatically land patrons on an HTML article, which will count as a “hit” in the JR1 download statistics, and then register another hit if the patron chooses to download the PDF. (there’s no way to get to the PDF without routing through the HTML first).

·        Readers in different subject areas exhibit different behavior: STM readers tend to download more articles,  and read more quickly (some articles may be skimmed over); humanities/social sciences readers tend to download fewer articles and read more slowly. Is your library prepared to analyze journal usage by subject area?

 

·        Hybrid journals: if a subscription journal has some open access articles, COUNTER reports won’t take the open access part into account. Also your patrons may not route through the link resolver or proxy server to hit the open access and it may not register as a JR1 article download at all.

 

·        Group titles: yes, it’s important to consider whether cutting a small package to get rid of underused titles will also eliminate the one popular title.

·        Some other phenomena that complicate usage:

o   New journals have less content to draw from, so usage statistics should be handled carefully there

o   Statistical fluctuations – the impact of individual behavior on small samples. Duly noted.

o   Title changes. ‘nuff said.

o   Journal transfers interfering with access

o   Journals like Nature have a lot of short articles and reviews, and also invite reading from cover-to-cover. Are the short articles and reviews as valuable as longer, more substantial articles?

o   Backfile vs frontfile content

o   Currency fluctuation and ongoing price adjustments (not a big deal for my library, your mileage may vary).

·        Aggregator platforms: do you try to combine statistics from a publisher and from an aggregator for the same journal? I wouldn’t combine them but it would be useful to compare them. As the author points out, coverage often varies between publisher and aggregator platforms for the same titles.

·        “Knowing” journals: librarians may have to become acquainted with the more popular and important journals to their faculty in order to know what is valued.

·        Nothing’s perfect: matter how hard you try to gather and weigh usage data correctly, there is always going to be some angst and mistakes made when cancelling journals.

Where I disagree with the author is the notion of correcting for usage spikes (legitimate usage). It doesn’t make a difference to me whether a journal is used throughout the year or heavily in a few weeks (perhaps used by a class for an assignment), the total usage is what makes it valuable. In either case it stands a chance of being used at similar rates again the next year. If anyone has an opposite view on this, please jump in and discuss.

 

I also think the author relies too heavily on ISSN as the magic matching device for journals. There are too many journals without ISSNs out there, and too many erroneous ISSNs floating around in various systems.

 

I don’t think the author takes document delivery that much into consideration. We have the Rapid ILL system at my institution and many articles are available within 24-48 hours (on business days) after request.  Supply and demand in our ILL department is somewhat elastic but balanced by the overall system.

 

I’d love to hear what others think about this article.

 

Diane Westerfield, Electronic Resources & Serials Librarian

Tutt Library, Colorado College

diane.westerfield@coloradocollege.edu

(719) 389-6661

(719) 389-6082 (fax)

 

 

 

 

***********************************************
* You are subscribed to the SERIALST listserv (Serials in Libraries discussion forum)
* To unsubscribe, send an email to the server address: LISTSERV@LIST.UVM.EDU . Do NOT include a subject line. Type as an email message these two words: SIGNOFF SERIALST
* For additional information, see SERIALST Scope, Purpose and Usage Guidelines.
***********************************************