Show me the data!
Header

Who actually has access to paywalled research?

September 16th, 2015 | Posted by rmounce in Open Access

Yesterday, I tried to read a piece of research, relevant to my interests that was published in 1949. Sadly as is usual, I hit a paywall asking me for £30 + tax to read it (I didn’t pay).

Needless to say, the author is almost certainly deceased so I can’t simply email him for a copy.

The paper copy is useless to me, even though my institution probably has one somewhere. I need electronic access. It would probably take me an hour to walk to the library, do the required catalogue searches, find the shelf, find the issue, find the page, re-type the paragraphs I need back into a computer, walk back to my desk etc… That whole paper-based workflow is a non-starter.

I noted the article is available electronically online to some lucky, privileged subscribers – but who? Why is the list of institutions that are privileged enough to have access to paywalled articles not public information? It would be extremely helpful to know what institutions have access to which journals & which journal year ranges.

So I thought I’d do an informal twitter poll of people on twitter about this issue:

I received an overwhelming number of responses. Probably over a hundred in total. Huge thanks to all those who took part.

Given such a brilliant community response it would be remiss of me not to share what I’ve learnt with everyone, not just those who helped contribute each little piece of information. So 24 hours later, here’s what I now know about who can access this 1949 paper (data supporting these statements is permanently archived at Zenodo):

Mounce, Ross. (2015). Data on which institutions have access to a 1949 paper, paywalled at Taylor & Francis. Zenodo.

I’m not pretending the following analysis of the data is rigorous science. It’s not. It’s anecdata about access to a single paper at a single journal (a classic n=1 experiment). Of course it also relies on each contributor correctly reporting the truth, and that some potential responses may have self-censored. The sampling is highly non-random and reflects my social sphere of influence on Twitter; predominately US and UK-centric, although I do have single data points from Brazil & Australia (thanks Gabi & Cameron!). Nevertheless, despite all these provisos it’s highly interesting anecdata:

The United Kingdom of Great Britain and Northern Ireland:
Of responses representing 41 different UK institutions including my own, only 3 have access to this paper, namely: University of Cambridge, University of Oxford, and University of Glasgow.
Had I got more responses from a wider variety of UK HEIs like the University of Lincoln and University Of Worcester where I also have friends, I suspect the overall percentage of UK institutions that have access would be even smaller! I’m particularly amused that it appears that no London-based institution has electronic access to this paper!

North America:
Of responses representing 29 different institutions in Canada and the United States, only 7 have access to the paper, namely: Virginia Tech, University of Illinois, University of Florida, North Carolina State University, Case Western Reserve University, Arizona State University, and McGill University. It’s intriguing that North American institutions appear to have slightly better access to this journal as originally the journal was published in London, England!

The ‘rest of the world’ (not meant in a patronising way):
Of responses representing 23 different institutions not based in the UK, Canada, or the United States, only 2 definitely have access to this paper: Wageningen University and Stockholm University. I note that the person who contributed data on Stockholm University access does not have an official recognised affiliation with Stockholm university and that they used alternative methods *cough* to discover this (just for clarity and to further demonstrate the sampling issues at play here!).

Despite asking far and wide. I only found 11 different institutions that actually have electronic access to this paper, and none from London where the paper was actually published.

I’m fascinated by this data, despite its limitations. I’d like to collect more and collect it more efficiently. Perhaps the librarian community could help by publishing exactly what each institution has access to? Although one conversation thread seemed to indicate that libraries may not even know exactly what they have subscribed to at any one point in time (Seriously? WTF!).

Why is this stuff important to know?

I often hear an old canard from certain people that we don’t need open access because “most researchers have access to all the journals and articles they need”. Sometimes some crap, misleading survey data is trotted-out to support this opinion. Actual data on which actual institutions have actual access to subscription-only research is pivotal to countering this canard. For example, it is extremely useful to point out that institutions like Brock University and University of Montreal do NOT have access to the bundle of Wiley journals.  Particularly at a time when maddeningly many societies have decided to start publishing …with Wiley e.g. the Ecological Society of America! It’s not very joined-up thinking and it’s going to create a lot of pain for a lot of people. Both Montreal & Brock & many other institutions with ecologists do not have access to the big Wiley bundle of journals. I’m sure there are useful examples in other subject areas too of mismatch between subscriptions held & needed access. The solution to this of course is NOT to re-subscribe, but to fix the problem at its source; to fully-recognise that access is a global issue and many people need access to a very wide variety of different journals, that a proper transition to an open access availability model is needed.

If I wait 26 years, it will be available for free in the Biodiversity Heritage Library. I hope I live that long!

What to do next?

If your institution isn’t listed in my dataset so far, please do still try and access this article and let me know if you can or cannot instantly access it via your institutional affiliations from Taylor & Francis.

Given we have researchers coming from all corners of the globe for OpenCon later this year, I will soon explore whether together, as the OpenCon community, we can do something like this on a grander scale to more rigorously document the patchy nature of subscription-provided access.

The final word

I’ll leave the final word, to the obvious ‘elephant-in-the-room’ that I haven’t discussed much so far, they are the 99.99% relative to us privileged institutionally-affiliated lucky-ones. I am very obviously aware of and do care about, independent researchers & readers of the ‘general public’; neither of which can afford subscription-access to most paywalled journals:

  • HI Ross
    here is a little experiment to access taxonomic information of few groups of scientific names, dedicated to rock stars, the species in my sons fish tank, http://plazi.org/wiki/Taxon_Search_Portal

    The goal is to figure out how many of the scientific descriptions /(treatments) are accessible. First, whether I can actually find the respective reference, then whether there is and I can obtain a digital copy of it, and finally, whether it is open access, the name in Wikipedia and Plazi (and with this in Wikidata).

    It is work in progress, but it shows not only how difficult it is to get access to such baseline information, but more in general, a big break in the information flow from the species names the public is aware of to the scientific evidence. Open Access is just one barrier.

  • Hi Ross,

    “Perhaps the librarian community could help by publishing exactly what each institution has access to?”

    Unfortunately, many uni libraries *have* to sign agreements with the bundle publishers that prevents them disclosing this useful information.

    Some researchers have put in a lot of effort trying to get this info, e.g.,
    https://gowers.wordpress.com/2014/04/24/elsevier-journals-some-facts/

    Interesting reading!