3 things we can do to make open access better

Open access is being pulled and pushed in different directions by groups who each have their own intentions and motivations:

  • Research funders want to maximize their investment and – by holding the purse-strings – are the change instigators accelerating the pace of adoption of open access. Some are more proactive than others pushing scholarly communication towards Gold OA in certain subject disciplines, whilst other funders are less active preferring change to be more organic.
  • Publishers, as gate-keepers of the scholarly written record, influence how open access happens through innovation (developing new business models and products), control of intellectual property (open licensing or imposing journal embargoes) and controlling the spiraling costs. Some publishers are profit-driven and seek the highest returns that the market can burden. Others are more motivated by the academic community
  • Libraries are change agents who can help to enable open access in institutions, for example through implementing repository platforms and offering support services and expertise. Their motivations to be involved are many; Library core values are well-aligned with open scholarship, they have a strong interest in and are well-placed to ensure institutional funds are efficiently allocated, and there is a drive to enhance their relevance through redefining roles within research institutions.
  • Academics. It is easily to fall in to the trap that academics are passive actors in all of this. It feels like the silent majority go along with the status quo as research is their prime concern, and scholarly communication is a side-show with which they have little interest in how it works. Because publishing is increasingly being outsourced they lack a sense of agency or ownership. However, some researchers are driven to innovate and change their scholarly communication practices.

The interaction of each of these players in the scholarly communication game has led to the development of a system driven by interlocking policies, platforms and processes, which we have shown over the course of the last few blog posts, is unnecessarily complex, expensive, inefficient and increasingly at risk of being not fit for purpose.

What steps should libraries be doing to improve scholarly communication?

1. Remove complexity

The problem with Green OA is – it’s not immediate (journals embargoes are far too long), it’s not compliant with all funders policies and it’s unnecessary complexity (checking and matching funders policies and journal embargoes) is inefficient and has many hidden costs.

Help your institution to adopt the UK-Scholarly Communications Licence and most of these problems are diminished. [Read more here]

2. Reduce OA publishing costs

Hybrid OA Gold is the most popular and expensive route for paid open access. A side effect of lowering embargoes is that authors can comply with their research funders open access policies via Green OA.

Where possible, stop paying Hybrid OA costs, and use the open access block grants for pure Gold OA only. [Read more here]

3. Innovate and nurture academic-led publishing

Academic and National Libraries should support academic-led publishing and open access initiatives that are inclusive and open to scholars who do not have budgets for publishing.

Help your staff and students set up their own open access journals using software like Open Journal Systems. Support initiatives like the Open Library of Humanities by becoming a supporter member, and if you are from a larger institution then you should offer to support at a higher rate. [Read more here]

University of Edinburgh 2016/17 Gold open access spend analysis

The University of Edinburgh receives two main block grants from Research Councils UK (RCUK) and Charity Open Access Fund (COAF) to support researchers funded by them to comply with their open access policies. The Library’s Scholarly Communications Team manages these block grants on behalf of the institution. In 2016/17 we spent £1,167,966 to make 697 articles open access via the Gold open access route. In addition, £28,951 was spent on other publication costs, for example page and colour charges.

A summary of the open access charges made to the top 15 publishers is shown in the table and figures below:

Row Labels Count of APC paid Sum of APC paid (£) Ave APC (£)
Elsevier 108 £283,413 £2,624
Wiley 71 £122,685 £1,728
Nature 47 £97,296 £2,070
Springer 107 £70,240 £656
Oxford Journals 28 £56,742 £2,026
Royal Society of Chemistry 49 £51,840 £1,058
BioMed Central 33 £46,295 £1,403
BMJ Publishing Group Ltd 24 £43,650 £1,819
Public Library of Science 29 £43,210 £1,490
Frontiers 16 £24,508 £1,532
American Heart Association 6 £20,693 £3,449
Taylor & Francis 28 £20,605 £736
American Chemical Society 8 £18,278 £2,285
BioScientifica 8 £16,215 £2,027
Royal Society Publishing 9 £15,984 £1,776

Points and trends to note:

  • Our overall spend on open access in 2016/17 was £1,167,966 which bought 697 papers, with an average Article Processing Charge (APC) of £1,676.
  • The open access expenditure is dominated by three large publishers: Elsevier, SpringerNature and Wiley. These 3 publishers account for 53% of our total spend. Note that we have shown SpringerNature as three individual publishers in the table and figure as they have distinct publishing identities – Springer, Nature and BioMedCentral.
  • The top individual publisher (Elsevier) alone accounted for 24% of the total spend, with their APCs 56% more expensive than the average.
  • Generally speaking, we found the highest individual APCs were from US society publishers, like the American Heart Association or American Chemical Society.
  • Best value APCs were from publishers where we have offsetting or national agreements with – for example Taylor & Francis, Springer and Royal Society of Chemistry.
  • 70% of our expenditure is on Hybrid Gold OA  – where journals charge a subscription and APCs – rather than pure Gold OA (30%) where APCs are the only charge.
  • Looking at the whole dataset Hybrid Gold OA is more expensive (£2,075) than pure Gold OA APCs. (£1,581).

Discussion

The University of Edinburgh publishes 6 -7,000 journal articles and conference proceedings per year. In the current market conditions if we were to fully transition to Gold OA then we estimate it would cost in the region of £10M per year. To put this in context Edinburgh University Library spends approximately £4M per year on e-journal subscriptions so this represents a significant uplift on the total cost of publication. I suspect this figure is similar for other research-intensive universities.

At its current scale and intensity, open access as purely delivered by Hybrid Gold OA and Gold OA alone is not likely to happen. Any increase to the total cost of publication is not acceptable.

If we wish to transition to full open access then significant cost mitigation for paid open access needs to occur, and alternative low cost or no cost options need to be investigated and adopted. Some options are described below.

Reducing cost recommendations

  1. Push for offsetting deals : Libraries should be a) requesting offsetting deals, b) pushing for significant discounts, and c) helping publishers adopt transformative business models.

Examples of publishers offering good offsetting schemes are IOP – where an actual rebate is given depending on Gold OA spend – and SAGE Publishing who globally discount the subscription rate of journals where more than 5% of articles are published as Gold OA. SAGE also offer steep open access discounts of 80% for some members of consortia subscriptions. Taylor & Francis are another publisher which significantly reduce APC cost (by 75%) to consortia members. These steep discounts are welcome, but have a fundamental problem in that they rely on privilege. Discounts are not available to all authors as they rely on institutions having a subscription.

Transformative business models such as the UK Springer Compact national agreement, or the Royal Society of Chemistry Read and Publish agreement, are also welcome as they lower transactional barriers for authors to participate in Gold OA. This leads to a greater coverage of open access articles in journal titles, rather than the patchy open access that Hybrid Gold OA delivers. The overall costs are also appreciably lower than the cumulative costs seen with Hybrid Gold OA – look at the difference in our expenditure between Elsevier and Springer for a similar amount of Gold OA articles published.

  1. Support ‘low cost and no cost’ Gold OA : Academic and National Libraries should support open access initiatives that are inclusive and open to scholars who do not have budgets for publishing.

Gold OA does not necessarily equate to paid open access. Many academic or library-led publishing initiatives do not charge Article Processing Charges as the costs are covered through alternative mechanisms. Some examples of innovative academic or library-led publishing activities are described below.

Edinburgh University Library provides an Open Journal Service which is available free of charge to University of Edinburgh students and academics. It provides a hosting platform for academic and student-led groups to create and publish their own Open Access journal. Currently the service has a portfolio of 16 journals and is looking to grow and develop in the coming year.

The Open Library of Humanities (OLH) is a charitable organisation dedicated to publishing open access scholarship with no author-facing article processing charges. The OLH publishing platform supports academic journals from across the humanities disciplines, as well as hosting its own multidisciplinary journal. The University of Edinburgh has opted to support the Open Library of Humanities at a higher rate than required. This additional support will enable the OLH to continue its growth mission to convert subscription journals to a solid, ongoing, open-access model, with no author-facing charges.

  1. Reduce reliance on Hybrid Gold OA : Hybrid Gold OA – where journals charge both subscriptions and Article Processing Charges – is too expensive.

Unfortunately, Hybrid Gold OA which accounts for 70% of our expenditure are significantly more expensive than pure Gold OA costs. It is not uncommon to receive invoices of $5,000 for individual APCs. Our open access block grants are supported either directly by taxpayers money (RCUK), or from charitable organisations (COAF), so it is imperative that anyone managing these funds should be seeking best value for money. I would find it very difficult to explain to someone who has raised money for Cancer Research UK the hard way by running the London Marathon that their sponsorship money has only paid for half an APC. Where research funders policies allow we should be seeking to reduce our reliance on Hybrid Gold OA by utilising alternative mechanisms like Green OA, which we will visit in more detail in tomorrow’s blog post.

Managing publisher’s open access memberships with Trello

Screenshot - 14_06_2016 , 16_18_06

As more and more publishers have started to offer open access institutional memberships we have started to struggle with the effective management of the various schemes. We are currently signed up to 12 active deals and are scoping half a dozen more. All of the memberships are at different stages of maturity leading to a fairly complex situation if you are trying to stay on top of things.

Currently this work is all carried out by one member of the Scholarly Communications Team; however, we are trying to move towards a more transparent decision making and management process which is available to the whole team and other associated staff members.

After reading an interesting post about the use of Trello within Libraries we decided to try out this project management tool for managing our open access membership schemes. I won’t go into detail about about Trello, as this is covered by the blog post linked above, but I will talk about how we are using the tool for this specific purpose.

Screenshot - 14_06_2016 , 16_35_12

Each project you wish to manage with Trello gets assigned it’s own board which you can keep private to an individual or share with a team. In the screenshot above you can see that I have 3 private boards and am sharing 2 boards with the Scholarly Communications Team. Looking in more detail at the OA Prepay membership board (screenshot below) you can see that we have created 4 vertical columns, known as lists, each populated with a number of cards. You can assign tasks to cards, but in this case we have chosen to allocate them to publishers membership deals. If you want to you can even assign cards to be monitored by specific people within the team, but we have not yet opted to do this.

Screenshot - 15_06_2016 , 10_50_29

The lists are a way to keep the cards organised – for example each list could represent a workflow step. If you like the ‘Getting Things Done‘ methodology you could have To Do, Doing and Done lists.

When looking at publishers’ open access membership deals there are a number of tasks that need to be carried out which can be arranged into 4 general groups:

  1. Evaluating the initial offering
  2. Sign-up procedures
  3. Day-to-day administration and usage monitoring
  4.  Renewal/ deactivation decision making

We decided to use 4 lists to represent the workflow of moving from evaluating new publisher deals, through completing the sign up procedures and monitoring usage, finishing with renewal/deactivation processes:

Scoping -> Current Memberships <-> Action required ->Dropped memberships

Each new potential open access membership deal would be given a new card (see below for an example). The card is firstly populated with a brief description of the membership. We then add a Label (to enable filtering by publisher), a Checklist of to-do items, and a Due Date when a decision is needed. It is possible to add attachments to the card, for example the offer documents or contracts, but we have decided not to do this. Instead we show the file path where the documents are saved on shared network drives accessible only to the team. We don’t trust giving a third party our business documents in case of a data breach.

Screenshot - 15_06_2016 , 10_59_19

Once the checklist items have been worked through, if successful and the membership was activated, the card would move over right to the Current Membership list. If unsuccessful we would place at the bottom of the Scoping list with a sticker indicating this (a big red thumbs down).

Screenshot - 15_06_2016 , 11_43_21

Once on the Current memberships list (2) the cards are ordered by dates, with the membership deals that are ending first placed at the top. Any member of the Scholarly Communications Team can then see which deals are active, when the membership runs out, find out details of the deal including costs, sales rep contact details and links to other documentation. This is preferable to having all the membership knowledge retained by one member of the team.

Screenshot - 15_06_2016 , 13_05_36

Towards the end of the membership period the cards are moved over to the Action Required list (3) to indicate that a renewal or deactivation decision is needed to be made. If renewed the card is placed back into the Current Memberships list with details updated and the Due Date reset (4). If the renewal is rejected then the card is passed right to the Dropped Membership list (5), with the decision details recorded so they can be revisited at a later date if the renewal is questioned.

In summary, the workflow we present here might not work for every institution however we have found Trello to be a very useful tool to manage the process of managing open access membership schemes within a team setting. It’s main benefits are that it is highly visual and thus easy to use with minimal training, very adaptable to fit different circumstances and the basic version is free.

 

APCs paid in the wild

As part of the Lessons in Open Access Compliance for Higher Education (LOCH) project we have been looking at the problems of identifying Article Processing Charges (APCs) that the Library is not currently aware of – which we colloquially call ‘APCs paid in the wild’. We have written a short case study that is currently under open review at The Winnower:

Improving estimates of the total cost of publication by recognising ‘APCs paid in the wild’

The take home message from our ‘APCs paid in the wild’ case study can be summarised as:

We estimate that these costs could account for up to 20% extra in the total cost of publication that is not currently being accounted for. This additional cost is important to take into account when institutions are negotiating fair offsetting agreements for open access publishing.

We would welcome any constructive criticism on the work so please have a read and leave an open review so that we can improve the article.

Open Access in the College of Medicine & Veterinary Medicine

My name is Anna Krzak, and I am an Open Access Research Publications Administrator for the College of Medicine and Veterinary Medicine at the University of Edinburgh. I have been in this role since March this year. Previously, I worked as an Open Access Publications Assistant (also for MVM) so I am not entirely new to the University and its Open Access (OA) project. I have been assisting academics within the College of Medicine and Veterinary Medicine with the RCUK OA policy implementation since April last year but, since I used to work mostly from home, I should introduce myself properly now.

The main purpose of my new role is to gather Open Access full text versions of research papers and the accepted peer-reviewed manuscripts and to upload the files to the Institutional Repository. As part of that, I ensure that the licensing terms and conditions are adhered to, including any embargo periods, and that any licenses or set phrases are acknowledged in PURE. In addition, I often advise academic staff on research funders’ Open Access policies and relevant Open Access options. If necessary, I consult the publishers in regard to their often unclear self-archiving policies (this is probably my least favourite bit..). As such, my role combines both theoretical and practical aspects of the OA implementation project that’s currently being undertaken throughout the University.

As the RCUK OA policy has been in force since April 2013, I thought it would be a good idea to evaluate the progress of its implementation in my College:

Please note that the collected evidence refers only to peer-reviewed research articles (including review articles) and conference proceedings that were submitted for publication after 1 April 2013 and that acknowledge the RCUK funding (as per the RCUK OA policy).

RCUK Compliance for the reporting period 1/04/13 – 4/06/14:

Approximately 224 research outputs have been identified, of which 192 have open access documents available to the general public. This means an 85% open access compliance rate (as of 4th June 2014).

All outputs All Open Access Gold/Gratis OA Green OA
         223      191      164            27

However, a more detailed analysis of the RCUK requirements for OA has revealed few secondary problems:

  • Licensing: Although the majority of all OA articles have been published under the CC-BY and CC-BY-NC licence (as required by the RCUK), in approximately 18 cases the articles were published under the CC-BY-NC-SA or CC-BY-NC-ND licences
  • Length of embargo periods: In 18 out of 27 cases the embargo periods were 12 months and longer
  • Self-archiving issues: In several cases journals didn’t offer any green options

If we take these points into consideration, the compliance rate for the specified period stands at approximately 67%, as compared to the required 45%. Overall, it’s quite a good result for MVM.

I’m afraid that my introduction has come across all too serious. However, in a face-to-face conversation you may find out that I am not really that bad 😉

-Anna Krzak, Open Access Research Publications Administrator, MVM

Three different traits of open access publishers

This week I’ve been compiling some data for the next meeting of the RLUK Ethical and Effective Publishing Working Group. Some of the data itself is pretty interesting so I thought I would write a quick blog post and share some preliminary thoughts on what it means. The table below shows the top 5 publishers in terms of money spent on article processing charges (APCs) from the RCUK open access block grant in 2013-14.

Publisher Total spend No. of APCs Average APC Discount on list price
Elsevier £52,596 36 £1,461.00 25%
Wiley £51,781 35 £1,479.46 25%
Public Library of Science (PLOS) £23,737 24 £989.04 0%
Nature Pub Group (NPG) £21,226 8 £2,653.25 0%
BioMed Central (BMC) £20,746 16 £1,296.63 15%

Article processing charges (APC) for the most popular journals for Edinburgh authors.

We found that 2 publishers stood head and shoulders clear from the rest of the field. In terms of gross spend and number of articles published the top publisher was Elsevier, with £52.6k and 36 articles. In second place, with a similar publisher profile was Wiley with £51.8k and 35 articles. Both of these publishers were followed by PLOS, NPG and BMC who all had broadly similar spends of around £20k. Whilst the total cost per publisher is interesting, what is really noteworthy is the number of articles that money pays for, revealing something of the publisher’s strategy in the open access market place.

The lowest APCs are incurred from the open access journals – PLOS and BMC – who have fees roughly a third less than the other publishers. The highest APCs are incurred by hybrid journals, who also make money from subscriptions, and article reprints. NPG stand out from the crowd as they charge nearly double compared to their competitors.

In summary, what we see here are broadly 3 groups of publishers with different traits:

Money Makers – traditional publishers with the biggest market share, the highest number of articles published, APC set to the highest they think market can bear without losing submissions, initially offering biggest discounts for institutional deals to get sign ups (and easier access to authors).

Prestige reputation – traditional publishers trading on their reputational status. Significantly less articles published but with larger APCs levied to publish in the journals with the highest impact factors. Strategy of selling high end products and services to those that can afford them.

Emerging challengers – new business model and products, more reasonable APCs to attract a market share. However, it is worth noting that since being bought out by Springer, BMC have attracted criticism for raising APCs much quicker than the rate of inflation.

When we get round to submitting the final RCUK report we’ll release our full dataset of article processing charges.

[Minor edits made to original to correct grammar, headings and stylesheet]

University of Edinburgh Open Access update: March 2014

As of 31st March there are approximately 76,800 records in our Current Research Information System (PURE), of which 16,795 have open access documents available to the general public (22% open access). In addition there are 170 records with documents waiting for validation.

Looking specifically at just journal articles and conference proceedings:

All time OA docs Open access % 2008 onwards OA docs Open access %
Medicine & Veterinary Medicine 6513 33 4476 41
Humanities & Social Science 3002 22 2509 36
Science & Engineering 5687 22 3826 30

Monthly application figures to the Gold Open Access funds:

Month Applications to RCUK Applications to Wellcome
January 2014 32 13
February 2014 24 13
March 2014 23 14

Status of the RCUK fund – currently there is £367,400 left in the fund*, with an additional £74,400 committed on articles submitted for publication. Altogether the fund has 35% left in the account.

(*this figure was slightly wrong last month – apologies!)

Status of the Wellcome fund – since the start of the new reporting period (November 2013) the cumulative open access spend has been £137,078

University of Edinburgh Open Access update: Feb 2014

As of 27th Jan there are approximately 75,700 records in our Current Research Information System (PURE), of which 15,486 have open access documents available to the general public (20% open access). In addition there are 71 records with documents waiting for validation.

Looking specifically at just journal articles and conference proceedings:

  All time OA docs Open access % 2008 onwards OA docs Open access %
Medicine & Veterinary Medicine 6450 26 4537 32
Humanities & Social Science 2667 19 2239 32
Science & Engineering 5514 22 3688 30

Monthly application figures to the Gold Open Access funds:

Month Applications to RCUK Applications to Wellcome
October 2013 23
November 2013 27 20
December 2013 19 9
January 2014 32 13
February 2014 24 13

Status of the RCUK fund – currently there is £327,222 left in the fund, with an additional £65,011 committed on articles submitted for publication. Altogether the fund has 39% left in the account.

Status of the Wellcome fundsince the start of the new reporting period (November 2013) the cumulative open access spend has been £108,372

Gold Open Access: Counting the Costs

The Digital Library team at the University of Edinburgh have written a new paper analysing the cost of Gold OA over the past few years. The data arises from managing the Wellcome Trust’s Open Access grant awarded to the university.

Some of the article’s main take home messages are:

  1. Hybrid journals seem to be more popular venues for Open Access  publication, and
  2. Hybrid journals generally charge more than full OA journals independent of journal impact factor, and
  3. There is a positive correlation between APC cost and impact factor for both hybrid and full OA journals.

Some more reflective points arising from the work:

  1. It appears that Open Access policies require rigorous compliance monitoring to be successful, and seem to be more effective when punitive sanctions are imposed.
  2. Research-intensive institutions are likely to be hit by a cost ‘double whammy’; they not only publish more articles, but they also publish them more frequently in high-impact-factor journals.
  3. Institutions need to be more open about costs, and publish the data in a format that allows reuse.

The full article is available here:

http://www.ariadne.ac.uk/issue70/andrew

The data set upon which the article is based is available here:

http://datashare.is.ed.ac.uk/handle/10283/250

 

Gold OA: how much will it cost us?

As the administrator of our institutions Wellcome Trust open access grant it’s that time of year when we have to prepare a report on last years spend on open access publication charges.

Given that the recent RCUK OA policy has a strong steer towards Gold OA it gives us a chance to look at some real facts and figures at what this policy change could cost for a research intensive institution.

Looking at the basic figures for last year we spent approximately £170,000 on 100 articles published in 75 different journals, working out at an average cost of £1700 per paper.

Extrapolating upwards the University of Edinburgh publishes in the region of 4000 peer reviewed journal articles per year – this figure does not include other publication types like conference papers not affected by the RCUK policy. Assuming an average Article Processing Charge (APC) of £1500 (more in line with other institutions) the total publication costs will be in the region of £6M. It is clear that even with guaranteed funding from HEFCE large research intensive universities will not be able to pay for all of their research to be published under Gold OA. How to allocate funding to researchers will be a difficult choice that many institutions are currently asking themselves – will it be on a first come first served basis, funder specific, or will REF submitted material take priority?

A more detailed analysis of Gold OA costs from the past 5 years is being written up for publication in a forthcoming article of Ariadne. Here is a taste of some of the data we will be looking at:

Price range JIF Range JIF Mean JIF Median Number of articles* Type
£0-£999 2.263 – 10.472 4.849 4.537 41 44% hybrid 56% full OA
£1000-£1999 0.856 -15.389 6.328 5.117 108 63% hybrid 37% full OA
£2000-£2999 0.986 – 12.594 4.411 3.441 61 100% hybrid
< £3000 5.971 – 15.710 12.363 10.881 17 100% hybrid

Table.1: The range of journal impact factors and mean/median values for the 4 main prices ranges (*discounting 3 journals which were not allocated a JIF in the Journal Citation Report). Journals with high impact factors seem to charge significantly more than other types of journal for gold open access