Yesterday we focused on the citation advantage for open access articles, particularly for repository-based articles. Today’s post is a guest post by Fiona Glasgow of our Research Support Unit.

There are many ways to make your work openly available. One option is to deposit your work in an institutional repository; at Otago we have OUR Archive. An institutional repository aims to collect, preserve, and make available digital copies of the intellectual output of an institution.

Around 80% of journals will allow you to deposit your research in an institutional repository after a certain time has elapsed from the date of publication – for free! This time period is often 6-12 months, though some but you will need to double check the contract you signed with the publisher or the policies on their websites. Alternatively, you can check this information on SHERPA/RoMEO. This site is a great way to find publisher copyright and self-archiving policies. As mentioned in Richard’s posts earlier this week, 84% of Otago-authored articles from 2017 that are currently behind a paywall could now be legally deposited in a repository.

Some benefits of using OUR Archive include:

• Making your research visible and accessible. Publications are indexed by search engines (Google, Google Scholar, DigitalNZ, etc); this can increase the ranking of your publications in Google searches and help them reach a broader audience.
• Providing persistent access. Each item is assigned a unique handle (persistent URL).
• Gathering statistics on views and downloads. Usage statistics are available for all items and department collections in OUR Archive, and include statistics based on city and country.

Associate Professor Janet Stephenson, Director of the Centre for Sustainability, makes a succinct and compelling case for the benefits of using OUR Archive in this short video interview. She talks about how using OUR Archive has been a critical part in getting the right kind of profile and impact for the Centre’s research outputs, and how increasing access to their work is important for PBRF.

Currently, the majority of research that is deposited into OUR Archive are theses, but it’s possible to deposit a wide range of research outputs, and file types. Over the coming months, the Library is going to focus on increasing the number of non-thesis deposits in OUR Archive. If you have questions or need assistance with the depositing process, please contact your subject librarian.

In early 2020, the Research Support Unit is planning upload-a-thons where librarians will help you deposit your research outputs in OUR Archive. These upload-a-thons also aim to demystify copyright and open access. By going into departments, we hope to tailor the events to your own domain-specific research needs – so bring along any questions you have and works you want to deposit. We hope to see you there!

So far this Open Access Week, there’s a chance I’ve depressed readers of this blog. “2 out of 5 Otago articles are free-to-read?” I hear you moan. “Access to research costs us how much?!?” you wail. But I did suggest that there are reasons to be positive.

Open access is often framed as being the right thing to do: the public paid for it so they should have access. That’s not my focus here and the research I’ve been discussing this week sought to look for measurable ways in which we can assess the effect that making your work open has on its impact.

Increasingly the agencies that fund our work are looking at impact, in particular with a new focus on impact outside academia (see MBIE’s recent position paper on research impact). Research that is referenced by policymakers and the media is more likely to have real-world outcomes than research that is cited only by the academic community. In our sample we found that open articles were cited in the media 3.5 times more than closed ones and mentioned in policy documents twice as often.¹

Another, more traditional, way to asses this is our old favourite academic citation rates.

Let’s examine the graph in some detail.

• Closed access articles (n=1480) – that is, those available only via subscription – fare worse than all the types of open access apart from the bottom one (Diamond).
• Hybrid articles (n=89) – those in subscription journals where you have the option of paying for your single article to be open – achieve the highest average. This result is not surprising as they are likely to be high-profile publications (a topic for further investigation for us). It comes at a cost, with the average Article Processing Charge (APC) being $4260 and totaling $93,000 in 2017.
• Green OA (n=237) – self-archiving or repository-deposited work – is almost on a par with Hybrid. Self-archiving incurs no APC, of course, and its average citation rate in our Otago sample is 93% higher than closed research (compared to a 66% advantage for all New Zealand universities).
• Bronze articles (n=162) – those articles whose open status is uncertain but are currently free-to-read – and Gold have similar average citation rates. Gold cost us an estimated $643,000 dollars at an average APC of $2873.
• Diamond journals are those in which is free to publish and free to read. These represent a small subset of our data (n=41) and are mostly small, independent journals.

The Green result is most interesting in the context of yesterday’s discussion, where we saw that we could be depositing the majority of our closed research in repositories, avoiding APCs. But we are not and thus we’re missing out on the citation advantage we see for Green OA here. To compound this, we’re missing out compared to other countries we would normally like to compare ourselves to, which have much higher rates of openness.²

The result for Green OA is also interesting in the context of the common attitude that the final, published version is the only one with value. Our findings suggest that that doesn’t matter to people who don’t have access to that version. Here we’re seeing that Green OA achieves a higher citation rate than the Gold/Bronze/Diamond forms of OA and that almost-double average compared to closed articles. Remember that what we’re counting as Green has been published in a closed journal, it’s just that a free version has been made available. You can still cite the published version even if all you’ve had access to is a free version. And that’s the heart of it here: I’m not going to cite something at all if I couldn’t read it.

None of this is rocket science but it’s the first time we’ve had evidence that is specific to our university and the New Zealand university sector as a whole.

No doubt, after reading the above, Otago researchers will be clamouring to deposit their closed access research in OUR Archive. We’ll look at the practicalities of that in tomorrow’s post, a guest post from the wonderful folks in our Research Support Unit.

Richard White is the Manager, Copyright and Open Access at the University of Otago.

This is one of a series of posts for Open Access week 2019. The comments can be used below for discussion or debate. Otago staff can refer to our Open Access Policy and associated Guidelines.

Notes

For the national results of the research referred to here, including an infographic and full report, refer to the Universities New Zealand website for:

• Infographic: CONZUL OA Project Group (June 2019). NZ open access: what do we know?
• Full report: CONZUL OA Project Group (August 2019). Open Access in New Zealand universities: an environmental scan.

¹ See a fuller discussion of this in our full report, cited immediately above, pp. 9-10.

² The Leiden Ranking tool uses a different method to that employed by our group, including using data from 2014-17, but is a useful tool to evaluate global trends and compare its results to our own. Leiden’s figure for the proportion of NZ research that is openly available in some form is 38.4%, close to what our research found at 41% nationally and 39% for Otago. Leiden’s NZ figure of 38.4% compares to Canada 42%, Australia 42%, Germany 48%, Ireland 49%, Norway 54%, United States 54%, and United Kingdom 71%. 34 of the top 50 universities for proportion of OA research are from the UK; New Zealand’s top-ranked university is the University of Canterbury at number 416 in the list.

When I talk to researchers about open access, cost is often the first thing that comes up. We know that researchers are in principle overwhelmingly in favour of their work being free to read (with 87% in a large survey backing open as default)¹ but, as we saw yesterday our practice in making our work open is hugely at odds with this since 39% of Otago-authored articles are openly accessible. Cost is definitely one barrier but lack of understanding of the scholarly publishing ecosystem is just as much a factor. Today’s post looks at these issues – it’s going to get detailed further down, so buckle up.

The short version is: we pay a lot for subscription access still and a not-insignificant amount on top of that for open access publications but we could be doing a lot more to make our work open in other, cheaper ways that are just as good (if not better).

The longer version? Here we go.

Most readers will know that publishers charge us subscription fees for access to research. This is still how we get access to most electronic material we use in teaching and research. In 2017 New Zealand universities combined spent $68.5 million on access to electronic resources and this goes up each year.² Open access came along but in some of its models this actually introduced a new cost, where publishers charge the authors/researchers to make it open as opposed to libraries (with fees known as Article Processing Charges or APCs). As I said yesterday that’s where our work estimated $735,000 spent by Otago researchers in 2017.³ The figure for the eight universities combined was $2.1 million so Otago’s share was about one-third.

This estimated $735k was spent for two types of open access, as indicated by the black line with the curved line at the end of it. The vast majority of this money was for what are termed Gold OA journals, where there is no subscription fee and all articles are open access with an APC charged to the authors (about $642k or 87% of the total), like Public Library of Science or Biomed Central. The remainder was in Hybrid journals, which charge libraries subscriptions but allow researchers the option of paying an APC to open up that particular article to all readers (e.g. the Lancet, Nature). This is an area of interest for further investigation: why did our researchers choose to publish in these venues and pay this fee, especially where it was optional in the Hybrid cases?

But there are other ways to make your work open. There are several different ‘shades’, as indicated in the smaller arch in the graphic,⁴ but the Green OA proportion is of particular interest because there are no APCs. My title for this post was deliberately provocative and not strictly true: it’s doesn’t cost ‘nothing’ to self-archive in that it requires time and effort to do it and repositories must be developed and maintained. But it’s much cheaper at scale than paying publishers $3000 to $4000 on average per article, with one piece of work identifying the cost per-article of depositing in a repository to be NZ$62.⁵

Most publisher policies now allow you to deposit an accepted manuscript in a non-commercial repository, sometimes with the proviso that you have to wait for a period after publication, most commonly 12 months. (Note: you can actually check the policy of any journal here). The figure on the right here shows what we could have deposited perfectly legally in a repository but haven’t:

People don’t realise they can do this or they don’t feel it’s worth the time and effort to do so. But they likely also don’t understand that on average there appears to be a big advantage in self-archiving your work in terms of impact, which is the subject of tomorrow’s post for OA Week 2019.

Richard White is the Manager, Copyright and Open Access at the University of Otago.

This is one of a series of posts for Open Access week 2019. The comments can be used below for discussion or debate. Otago staff can refer to our Open Access Policy and associated Guidelines.

Notes

For the national results of the research referred to here, including an infographic and full report, refer to the Universities New Zealand website for:

• Infographic: CONZUL OA Project Group (June 2019). NZ open access: what do we know?
• Full report: CONZUL OA Project Group (August 2019). Open Access in New Zealand universities: an environmental scan.

¹ Blankstein, M., & Wolff-Eisenberg, C. (2019, April 12). Ithaka S+R US Faculty Survey 2018. doi: 10.18665/sr.311199

² The Consortium of New Zealand University Libraries reported that subscription to electronic content in 2017 cost universities NZ$68.5 million. See the Universities NZ submission to the Copyright Act review p. 28. For context the Marsden Fund gave out  $84.6 million (Source: Royal Society).

³ This estimate is an estimate because, while we know how many 2017 articles were published where the corresponding author was an Otago researcher in a journal that would require an APC to be paid, we don’t know for sure if that fee was waived or paid by someone else.

⁴ Getting into the minute detail, we can break down the white 39% open section in the graph above into sub-groups to show how the articles were made open. The largest proportion was for Gold OA (16.91% of all articles); next comes Green OA at just under 10% (sometimes called self-archiving, where articles are published in a closed journal but an accepted manuscript version is deposited in a repository like OUR Archive); Bronze OA (6.7%) is where the article is currently freely available but it’s status is uncertain and could change; Hybrid (optional APC) and Diamond (free to publish, free to read) made up the remainder.

⁵ Johnson, R., Pinfield, S., & Fosci, M. (2016). Business process costs of implementing “gold” and “green” open access in institutional and national contexts. Journal of the Association for Information Science and Technology, 67(9), 2283-2295. doi.org/10.1002/asi.23545

How many of Otago’s research articles are free-to-read on the web? Not as many as you would hope.

‘Access provided by the University of Otago’: when you’re reading an online article how often do you notice that little piece of text at the top of the screen? It’s ubiquitous (and tiny) so we hardly notice it but, of course, we can read most of the research we’re interested in because we’re paying for access. The University of Otago has a very high level of access compared to many other teaching or research organisations – not to mention all the decision-makers in government or local bodies, practitioners, business/innovators, media, iwi groups and other stakeholders and the general public who have little or no access to a lot of research publications.

So how much of Otago’s own research is free-to-read online for those who are interested in it? This is something we haven’t had a good idea about – until now. You can see from the above that 3 out of every 5 (61%) are only available to those who can afford to pay for access. This finding comes from a national project looking at the current state of open access in New Zealand, the results¹ of which I’ll be blogging about over the course of Open Access week (21 – 25 October 2019). Out of the 2418 journal articles in our sample published in 2017 by Otago researchers, 938 were online for anyone to read for free (39% or, roughly, 2 out of 5).² The other 1480 papers were only available via a subscription, meaning all those groups I listed above generally won’t have access.

We’re interested in knowing how much of our work can be accessed without barriers because we know that this benefits us not only in scholarly terms but also because it can benefit the wider impact of our work outside of academia. But we’re also interested in how we made our work open and whether it cost extra to do so: that’s the estimated $735,000 in the callout box on the right spent by Otago researchers in 2017 on what are termed Hybrid and Gold open access journals. More on these interesting questions in tomorrow’s post!

Richard White is the Manager, Copyright and Open Access at the University of Otago.

This is one of a series of posts for Open Access week 2019. The comments can be used below for discussion or debate. Otago staff can refer to our Open Access Policy and associated Guidelines.

Notes

¹ For the national results of this work, including an infographic and full report, refer to the Universities New Zealand website for:

• Infographic: CONZUL OA Project Group (June 2019). NZ open access: what do we know?
• Full report: CONZUL OA Project Group (August 2019). Open Access in New Zealand universities: an environmental scan.

² This study used a dataset comprising all the journal articles published by the eight New Zealand universities in 2017 with a Digital Object Identifier (DOI). The graphic and data discussed in this post represent a subset that had at least one Otago author on the paper.

Nature has published a feature article that provides good overview of the current state of data publication and sharing in science. Despite the title, which suggests evangelism in favour of open access, it’s a generally well-balanced view of the challenges facing us right now:

“…the current state of science: partly open, partly closed, and with unclear and inconsistent policies and expectations on data sharing that are still in flux.”

Popkin, Gabriel. Nature 569, 445-447 (2019) doi: 10.1038/d41586-019-01506-x