Unsub and COUNTER5 and trying to figure out why the data is

143 views
Skip to first unread message

Alice Fairfield

unread,
Sep 16, 2021, 3:49:43 PM9/16/21
to Unsub discuss

Hi Folks,

We're starting to use Unsub to analyze our Springer collection for the second time, now with COUNTER5, and the results are baffling compared to last year with COUNTER4.

Since UNSUB with COUNTER5 uses turnaway stats, would I be correct in assuming that those are counted as "downloads" in some way?

Because we are seeing large amounts of "downloads" for journals that we don't have access to.

The second part is that many of these odd journals are ones that are very, very, unlikely to have any turnaways from UConn folks (tiny German-language only niche journals).

Our best guess of the source of the data anomalies are the periodic SciHub attacks using purloined UConn credentials, and the bots trying to sweep up everything, including journals we have no access to.

Is there anyway in Unsub to sanitize (and sanity-ize) the data to account for this?

Thanks!

Alice

***************************************************************

Alice Fairfield
Serials Librarian
Librarian for Gender and Sexual Minorities Studies
Acquisitions and Discovery Unit
University of Connecticut Library
http://s.uconn.edu/Fairfield
She/Her/Hers

Lisa Hinchliffe

unread,
Sep 16, 2021, 5:34:56 PM9/16/21
to Alice Fairfield, Unsub discuss
I think it may be that those are journals that Springer Nature opened up during the pandemic pivot and so though you don't subscribe your users had access?
 
I did a webinar with Bob Boissy (Springer Nature) and Lettie Conrad (Consultant) on this general topic of user journeys and impact on data reporting earlier this year on April 22 where we predicted that libraries were going to see these sorts of anomolies:
 
User Journeys with Online Learning and Research During the Pandemic: A Scholarly Communications Symposium
The scholarly content discovery landscape turned almost completely digital during the pandemic period we have all experienced.  But have user journeys and user experiences to find and use digital content changed?  Some libraries report unexpected downturns in digital resource use during this period and look for explanations.  Were downturns experienced in specific sectors of academic and corporate entities?  Were users disconnected from content they needed during this time?  Were they disconnected from their libraries?  What measures were taken to mitigate disconnection?  What effect did the massive surge of freely available published Covid-19 research have on the overall picture of use, and how much of what happened has been captured by library and publisher COUNTER data?  Is there an effect relating to use of the relatively new release 5 of the COUNTER Code of Practice?  Are there differences between the academic and corporate sectors when it comes to our understanding of the user experience?   Some facts are known and some theories are being tested about digital resource use during the pandemic, and the time for discussion and learning is now.
 
A recording is available: https://register.gotowebinar.com/recording/5010125257568129800
___

Lisa Janicke Hinchliffe
lisali...@gmail.com
___

Lisa Janicke Hinchliffe
lisali...@gmail.com





--
You received this message because you are subscribed to the Google Groups "Unsub discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to unsub-discus...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/unsub-discuss/e0bdaee4-9ab8-4a8a-a0de-b38035f5b6b7n%40googlegroups.com.

Alice Fairfield

unread,
Sep 17, 2021, 9:45:55 AM9/17/21
to Unsub discuss
There's probably some usage from the open COVID materials, but that doesn't explain the inexplicably high numbers for these German language, non-medical titles. And probably others, but these all have negative CPU so they stand out.

Alice

Alice Fairfield

unread,
Sep 20, 2021, 3:27:50 PM9/20/21
to Unsub discuss
Looking at our data, I'm further mystified.

One title is "Zorg + welzijn", which is not present in our COUNTER5 JR2, JR3, or JR4 reports, and yet Unsub claims we have 363 downloads!

Clearly Unsub is pulling these numbers from somewhere, but why, and how do we make it stop!

Ian Robson

unread,
Sep 20, 2021, 3:52:48 PM9/20/21
to Alice Fairfield, Unsub discuss

Alice,

 

Could the downloads be a result of Unsub’s use of weighted usage data as a representation of citation and authorship data? While you may not have any actual downloads, perhaps your faculty are citing / publishing in the journal?

 

Best,

 

Ian

_______________________________________________________

Ian Robson | Head, Information Resources | Queen's University Library

Working Remotely | ian.r...@queensu.ca | library.queensu.ca

Pronouns: he / him / his

Queen's University is situated on traditional Anishinaabe and Haudenosaunee Territory.

Alice Fairfield

unread,
Sep 20, 2021, 4:03:08 PM9/20/21
to Unsub discuss
Hi Ian,

For Zorg + Welzjin, Unsub lists no citations or authorships by UConn folks. I assume it's only showing up in our Unsub via the public price list data, but it should have 0  usage based on the reports we've uploaded.

Alice

Ian Robson

unread,
Sep 20, 2021, 4:09:54 PM9/20/21
to Alice Fairfield, Unsub discuss, Heather Piwowar, Jason Priem

Alice,

 

Thanks for the clarification. I am using Unsub to analyse a couple of big deals now and would like to understand this issue, too. @Heather Piwowar and/or @Jason Priem – are you able to shed any light on this?

Scott Chamberlain

unread,
Sep 21, 2021, 5:33:28 PM9/21/21
to Unsub discuss
Hi all, 

First, I'd like to introduce myself. I'm Scott, the new Unsub product owner. I'm still newish (I've been here for about 6 weeks), but I'm getting up to speed quickly :) 

Second, I'll dig into this and reply here as soon as possible!

Best, 
Scott Chamberlain

Ian Robson

unread,
Sep 23, 2021, 8:28:39 AM9/23/21
to Scott Chamberlain, Unsub discuss
Scott,

Thanks very much for your reply. Looking forward to working with you.

Best,

Ian

Ian Robson | Head, Information Resources | Queen’s University Library
Pronouns: he / him / his
Queens’ University is situated on traditional Anishinaabe and Haudenosaunee Territory
From: unsub-...@googlegroups.com <unsub-...@googlegroups.com> on behalf of Scott Chamberlain <sc...@ourresearch.org>
Sent: Tuesday, September 21, 2021 5:33:28 PM

Alice Fairfield

unread,
Oct 8, 2021, 10:20:41 AM10/8/21
to Unsub discuss
Hi Scott,

I just wanted to check in and see if you had discovered anything yet.

Thanks!

Alice

Scott Chamberlain

unread,
Oct 11, 2021, 5:26:44 PM10/11/21
to Unsub discuss
Hi Alice, 

I replied off thread to you personally to discuss details. Did you get my email?

Best, Scott

Ian Robson

unread,
Oct 12, 2021, 8:54:15 AM10/12/21
to Scott Chamberlain, Unsub discuss

Scott,

 

If possible, it would be desirable to respond to the list about questions we have posed. That way, everyone benefits from the knowledge and can review it later on.

 

Best,

 

Ian

_______________________________________________________

Ian Robson | Head, Information Resources | Queen's University Library

Working Remotely | ian.r...@queensu.ca | library.queensu.ca

Pronouns: he / him / his

Queen's University is situated on traditional Anishinaabe and Haudenosaunee Territory.

 

Alice Fairfield

unread,
Oct 13, 2021, 12:17:16 PM10/13/21
to Unsub discuss
The mysterious stats for Zorg was explained by me not looking hard enough at the COUNTER reports (dang case-sensitive searches).

Scott is still looking into how Unsub treats turnaways compared to actual downloads.

Alice

Strudwick, Jane

unread,
Oct 13, 2021, 12:21:01 PM10/13/21
to Fairfield, Alice, Unsub discuss

Freakin’ Zorg.

 

From: unsub-...@googlegroups.com <unsub-...@googlegroups.com> On Behalf Of Alice Fairfield
Sent: Wednesday, October 13, 2021 12:17 PM
To: Unsub discuss <unsub-...@googlegroups.com>
Subject: Re: Unsub and COUNTER5 and trying to figure out why the data is

 

*Message sent from a system outside of UConn.*

 

Strudwick, Jane

unread,
Oct 13, 2021, 12:23:28 PM10/13/21
to Strudwick, Jane, Fairfield, Alice, Unsub discuss

Sorry, list. The joke was not for the list ☹- a Reply All snafu. For the record, I have nothing against Zorg.

 

Best,

Jane

 

 

Jane Strudwick
Head of Acquisitions & Discovery

UConn Library
369 Fairfield Way, Unit 1005B | Storrs, CT 06269-1005
lib.uconn.edu

Scott Chamberlain

unread,
Oct 14, 2021, 7:15:22 PM10/14/21
to Unsub discuss
Ian, Good reminder. I wanted to talk with Alice off line to keep private any sensitive discussions - and then come back here to recap what we found. 

I'm going to write up an article about what data is included in Unsub scenarios, including what Alice refers to above as "turnaways vs. downloads" for our knowledge base here http://help.unsub.org/en/ and will ping back here when that is up!

Best, Scott

Scott Chamberlain

unread,
Dec 6, 2021, 5:50:23 PM12/6/21
to Unsub discuss
Hi all, 

Sorry for the delay in following through on what I said above. 

We've written up two Unsub help articles that explains what data is included in Unsub scenarios:
We'd love any feedback you have on the above two help pages!

Best, Scott

Reply all
Reply to author
Forward
0 new messages