Usage report issues

Issue Report: APS is now using a service called Inqwell to generate their COUNTER reports. There are some issues with the reports I just retrieved for my institution:
1. Incorrect column headers in JR1
• H8 is called “Retrievals” instead of “Reporting Period Total”
• I8 is called “HTML” instead of “Reporting Period HTML”
• J8 is called “PDF” instead of “Reporting Period PDF”
2. The xls version of JR1 has the same header naming error, and it has footer information (explanation of the report type).

Usus Response:
These are issues of noncompliance with COUNTER. Inkwell and APS were contacted.

Issue Report: How does COUNTER address reporting of prior titles for publishers adhering to PIE-J — separately, in both the prior and current title, or one or the other?

Usus Response: If the title or ISSN changed then usage should be reported separately for each instance.

Issue Report: It appears that Oxford Journals may be leaving titles with 0 use off of the JR1. I have two titles (Proceedings of the Aristotelian Society for the Systematic Study of Philosophy ISSN 0066-7374 and British Journal for the Philosophy of Science ISSN 0007-0882) which I know are on the Oxford Journals platform, and which are not on the JR1, which prompted me to look at the Reporting Period Total column — and no titles report 0 use. My notes from last year indicate that Oxford had previously been leaving blank cells instead of entering 0s, so possibly they’re trying something new this year?

Usus Response: Usus contacted ScholarlyIQ, which processes OUP’s statistics. They replied that OUP does not provide them with subscription data that they could use to create the zero entries. According to Note 11 in the COUNTER Code of Practice, dated September 2016, “[I]f a platform is unable to provide zero usage for entitlements only, they should not include the zero usage.” In this case, therefore, ScholarlyIQ/OUP is not required to include the zero usage in their JR1 (and CR1) reports.

Issue Report: In past years, IngentaConnect had not reported titles with 0 usage at all; now they do — and their COUNTER JR1 has more than 20,000 titles (we subscribe to two). Many of these titles are not real; eleven are duplicated titles but with typos in the ISSNs.

Usus Response:
Usus contacted Ingenta, which recently brought their usage statistics program in-house, to let them know that publishers should limit zero usage to subscribed titles only and that there were title issues in the JR1 including missing titles, duplicate titles, and invalid ISSNs. Ingenta responded that they have development work already scheduled to restrict the titles listed to just those where subscriptions are held for an institutional registration (including zero usage titles) and that they would address the title issues.

Issue Report: Manually downloading reports from EDP Sciences (Vision4Press) shows empty columns “Publisher”, “Print ISSN” and “Online ISSN”.

Usus Response: Usus contacted ScholarlyIQ, which processes stats for Vision4Press, and they saw that these values are not currently being provided to them by the publisher. They have notified the publisher and will work with them to provide these values.

Issue Report: Downloading an Excel JR5 from the new Cambridge Core Admin Site, I found that the columns for YOP are in the wrong order – starting with “Articles in Press” from the left, but then YOP 1999, YOP 2000 … finishing with YOP 2016 and YOP unkown as the last two columns on the right. YOP pre-2000 is missing.

Usus Response: With regard to the ordering of YOP columns in JR5, while the Code of Practice does not explicitly state the order in which these columns appear, the common expectation is that they be in descending order when viewed left-to-right. Usus will contact Cambridge University Press to request that they update the order of the columns.

Issue Report: Manually fetching reports from Silverchair publishers (e.g., AMA, Annals of Internal Medicine, many other small societies), instead of storing the statistics in the file as integers, they are in formulas (e.g. “=400” instead of just 400).

Usus Response: COUNTER will be contacting Silverchair and the relevant auditor about the formulas in the cells.

Issue Report: My library doesn’t have a Project Euclid package; instead we subscribe to more than a dozen journals available on the Project Euclid Platform. Project Euclid started to offer COUNTER 4 reports for 2015, but their interface requires librarians to download a separate JR1 file for each subscribed journal. This set up of course requires extra time since librarians need to download several reports from the same platform and then they have to compile those files into one file before loading them into an ERM. The vendor should provide one JR1 file for each customer.  Could you please check with Project Euclid?  Thanks!
Usus Response:  Project Euclid has been contacted and is now in the process of bundling all of the individual journal reports into a single download.  They hope to have this deployed in the next few weeks.

Issue Report: I’ve noticed that several publishers whose content is hosted by the same usage processing vendor do not include zero use titles on the JR1 and JR1a reports if these data are harvested via SUSHI protocol. Whereas, if I request the same reports manually, these do include zero usage titles.
Usus Response: The COUNTER Code of Practice does require publishers to include zero use titles in the Journal Reports regardless of delivery mechanism.  (Only providers of aggregated full text databases, such as EBSCO and ProQuest, are exempt from providing zero use titles.)  In this case, the usage processing vendor has been notified of the problem and is in the process of correcting it.  They will be working closely with their publisher clients and notifying each one once the problem has been resolved.  It is important to periodically compare SUSHI reports to manually harvested COUNTER reports.  The community is encouraged to alert Usus to issues such as these.

Issue Report: Errors have been found in JR5 reports harvested using SUSHI from IEEE, IOPscience, and RSC. These errors include empty identifier elements, publication year value errors, and issues with usage dates.

Usus Response:  All of the publishers identified with these problems have their statistics handled by the same usage processing vendor.  The vendor has been contacted, and they have agreed to make the changes.  They will need around 2 weeks to complete this work, but the updated reports should be available to the libraries in the next month’s reporting cycle.