OA APC variation: english language editing

MDPI provides a good example and explanation of a variation on article processing charges that reflect the work involved, that is, english language editing services. From the MDPI APC website (August 5, 2014):

For journals with an APC of 500 CHF or lower, a charge of 250 CHF will be applied to articles requiring extensive English language editing or formatting. To avoid surcharges, authors are recommended to carefully follow the instructions for authors and use the MS Word or LaTeX template files available on the instructions for authors page of the journal website. We encourage non-native English speaking authors to send their manuscripts to a professional English editing service prior to submission. If you use a service that provides a confirmation certificate, please send a copy to the Editorial Office. Authors from developing countries should consider registration with AuthorAid, a global research community that provides networking, mentoring, resources and training for researchers in developing countries.

This illustrates an important point about scholarly publishing when viewed as a service rather than as a good for sale: there is an inverse relationship between quality and the amount of work involved, i.e. the higher the quality, the less the work that is needed. This is because publishers do not pay for the largest portion of the work, conducting the research and writing the article. A well-researched and well-written article is less work for a journal at every step, as high quality articles make for easier editorial and peer review decisions as well as less work at the copyediting stage.

This approach provides an incentive for authors to submit articles in much better shape along with clear instructions about what that means, and also points to assistance for authors from developing countries.

One suggestion that fits with this approach is that it may be more effective to shift much of the support work of formatting and copyediting from publishers to the author’s institution. This way, institutions could hire local assistants and pay at rates appropriate for their own country and in their own currency, as well as creating local jobs, perhaps jobs for their own graduates in the case of universities.

For authors and copyeditors, there are advantages to working together over multiple projects. The copyeditor then has an opportunity to learn the terminology and approach preferred by the author, lessening the workload for both parties, as well as an opportunity to observe the growth of the author and research project over time. Perhaps a staff person in this position can help researchers with similar administrative tasks such as filing paperwork for grant proposals. This would free up the time of researchers to focus more on research. Where would the money come from? My suggestion is that in the process of transition to OA, we should not be looking to or funding publishers for services such as copyediting and formatting.

Of the 124 journals listed on the MDPI website, 124 or 85% offer english language editing services, generally at 250 CHF.

As a methodological note: while MDPI listed 124 journals, as of May 15, 2014 DOAJ listed 48 titles for MDPI. Correction August 5, 2014: note that DOAJ lists 104 journals for MDPI; the 48 titles are ones for which there are APCs. Many MDPI journals do not charge APCs, so the discrepancy is much less than I had thought.

I am finding that these large variations in title lists between websites of publishers relying on OA APCs and DOAJ are quite common. Simply collating these lists is proving to be a fair bit more challenging than anticipated.


Open access article processing fees: variation by type of article

One of the results emerging from the OA APC study is that behind the “APC” there is actually a variety of sub-models. An element of APC charges that makes sense when you think about is that sometimes the fee varies depending on the work involved.

One model that likely reflects the volume of work: some open access publishers charging varying fees depending on the type of article. Following are the fees charged by Bentham Open for their Group One journals (from the publisher’s website as of May 2014):

  • Letters: The publication fee for each published Letter article submitted is US $600.
  • Research Articles: The publication fee for each published Research article is US $800.
  • Mini-Review Articles: The publication fee for each published Mini Review article is US $600.
  • Review Articles: The publication fee for each published Review article is US $900.

For Group Two journals, there is a flat fee of $250 per article regardless of type.

BioMedCentral has a similar model, except that instead of differential fees, there is a standard 20% discount for certain types of articles (published protocols and case studies).

I argue that although these variations add complexity to calculating costs, these experiments are healthy and a better reflection of publication charges based on the service of publication than would be the case with a flat fee. Encouraging this approach may be wise to facilitate transparency and to work towards affordable pricing for full open access. Consider, for example, if we push publishers like BMC and Bentham Open to charge a flat fee per article, regardless of type, to simplify accounting. In this scenario, I would predict an APC at or near the highest current per-article price. What do you think?

Other examples of variations in pricing models that reflect the amount of work involved (blogposts to come):

  • page charges
  • flat fee up to a certain number of pages, page charges for overage
  • language editing services

Genome Biology: not listed in DOAJ

Genome Biology, a well-established open access journal published by BioMedCentral with an impressive impact factor, is not listed in the Directory of Open Access Journals (DOAJ).  Presumably this is because one of DOAJ’s criteria for inclusion is “All content freely available”. In Genome Biology, the research articles are open access, but subscriptions are required for other content. Genome Medicine uses the same approach and is similarly not listed in DOAJ.

This is just one illustration of a methodological conundrum for the open access article processing fee research project. We are using DOAJ as the main source list for open access journals, however the DOAJ title lists for open access publishers using OA APCs do not quite match the publishers’ own lists, at least not the Hindawi and BioMedCentral title lists. In some cases this is likely due to recent changes at the publisher (new journals, older journals that have ceased to exist, changed titles or merged). However, the omission of a journal like Genome Biology is significant for this type of research because it is well-established, with an impact factor and a relatively expensive APC.

Traditional journals that use a hybrid approach (some articles open access while the journal as a whole continues as a subscription journal) are appropriately described as “double-dipping” by the open access movement. Are we giving publishers like Springer (the owner of BioMedCentral) an unwarranted free ride for doing exactly the same thing? Considering the high cost of publishing in Genome Biology or Genome Medicine ($2,835 US), this may be a question worth asking.