June 30, 2015
In the humanities there are usually no author-side fees for publication in open access, but when they do occur they are more frequently paid by researchers themselves. Fees in the humanities also seem to be relatively low. In biomedical sciences the situation is the opposite. Fees are common, and relatively high, but they are most often covered by grant money. What are reasons for this phenomenon? Is it going to change?
One of the biggest problems of open access publishing is funding. Open access advocates, publishers and governmental bodies are looking for a sustainable economic model for academic publishing that will make scientific communication effective. The easiest open access funding model for publishers to introduce is the so-called author pays model, which simply reverts the standard economy of academic publishing. However, this model has a major disadvantage – it engages the author in the transaction and requires him or her to have some knowledge about the funding available. In an ideal situation, the money paid by the author as an Article Processing Charge would be the very same money that is saved by a library thanks to a decrease in subscription based content. However, it is not always so simple, and as research has shown, funding APC is often a problem for authors. Today, I will present some more important data from previous research on this phenomenon.
According to the huge on-line survey called SOAP,, conducted in 2010, 52% of authors have published at least one open access article in previous 5 years and for 50% of them paid author side fees.
Different fields, different situations
With regards to disciplinary differences, it emerges that the last open access article was free for only 32% of authors working primarily in Biological and Medical Sciences*, and for 58,7% of authors working mainly in other Natural Sciences and Engineering and as much as 78,9% of authors for whose main field is Arts, Humanities or Social Sciences. The “author pays” model is therefore clearly dominant in the field of biomedical sciences only. And arts and humanities especially tend to use different solutions.
The author’s of the SOAP survey also asked how the fee was covered in the case of the most recent open access article, if it was non-free for the author. Here the disciplines have reversed positions. Only 13,3% of authors working primarily in Biological and Medical Sciences have the fee covered at least partially from their own pocket, which occurred in the case of 14,5% of researchers from other Natural Sciences and Engineering and in the case of 36,5% of authors in Arts, Humanities and Social Sciences. This means that the APC based model is not really “author pays”.
The second important conclusion is that in humanities there is usually no author-side fee for open access articles, but when they occur they are more frequently paid by researchers themselves. Fees in humanities also seem to be relatively low. In biomedical sciences the situation is the opposite. Fees are common, and relatively high, but usually they are covered by grant money.
SOAP was conducted on a non-randomized sample, which means that these statistics might not be true for the whole population of authors of open access articles. Although, the results from other research are similar, which means that they are probably correct.
Similar results was obtained by Solomon and Björk in their study conducted apparently in more or less the same time as SOAP survey.They contacted corresponding authors of articles published in selected APC charging journals indexed by DOAJ. They found that only 8% of authors working in Health Sciences, Biology and Life Sciences covered publication costs from their own pocket. It was 19% in the case of Education, Social Sciences and Political Sciences and as high as 46% in the case of authors working in Business and Economics. In this study also authors working in Agriculture and Forestry frequently claimed that they cover fees from their own pocket (35%).
In his later study, Solomon sent the survey to authors who published in four megajournals (BMJ Open, Plos One, Sage Open and PeerJ), all charging APCs. He found out that in the case of Sage Open, which operates in the social sciences, 62,6% of authors paid articles with personal funds against 10,8% for BMJ Open, 7,8% for PLOS ONE, and 32,3% for PeerJ (all three journals operate in biomedical sciences).
In fact, Sage knew about this fact earlier and it drastically dropped its fees in 2013, justifying it with the fact that the majority of their authors paid them the fees themselves. Why?
Why in humanities researchers pay from own pocket?
Why do researchers in the humanities pay fees from their own pocket more frequently? SOAP data suggests that it is because they have funding less frequently for unspecified goals that they can spend on APCs. Humanities researchers answered that their institutions have covered part or whole of their fee almost as frequently as biomedical scientists. The difference in the answer “My research funding includes money for paying such fees” is bigger (33% to 25%), but the popularity of the answer “I used part of my research funding not specifically intended for paying such fee” really makes the difference between these two fields (40,1% to 25%). Also, data collected by Solomon and Björk shows that in the disciplines where fees are covered by personal funds more frequently, they are significantly less frequently covered with “grant/contract” money, while using institutional and national funds was equally popular.
The most important question that might be raised here, and which needs further examination is how many APCs are covered with money which is intended for this purpose. This will show what is the real influence of policies supporting gold open access on the APC market. At this moment it seems that the real reason for the success of the APC based model in biomedical sciences is not the institutional support for this solution, but rather the high level of general funding in this field. This is also an obvious reason for the low popularity of this model in less wealthy fields. Will increasing the number of funding opportunities designed strictly to support gold open access alter this disproportion?
Is searching for funds a problem for authors?
SOAP also provides data about researchers who do not know what the fee was for their recent open access article. It is 17,0% in Biological and Medical Sciences, 11,8% in other Natural Sciences and Engineering and 9,7% for Arts, Humanities and Social Sciences. This means that some researchers have nothing to do with paying fees, which is quite positive since their job is to do research.
What is also worth remembering is that the authors who paid a fee for their last article (by any means) usually (54,3% according to SOAP) thought it was “difficult” to find money for this purpose. The differences between the disciplines are not big.
The researchers who declared in the SOAP survey that they published no open access articles within the previous 5 years were asked if there was a specific reason for not publishing such articles. Those who answered “yes”, where asked to specify the reason. Among them the biggest group pointed to funding as one of, or the only obstacle for publishing in open access. This was said by 62,1% researchers from Biological and Medical Sciences, 44,8% from other Natural Sciences and Engineering and 23,4% of those working primarily in humanities. This means that the popularity of the author pays model in biomedical sciences limits the publishing possibilities for some researchers who have difficulties in gaining funds. This happens less frequently in fields where APCs are smaller and less popular.
All the studies I have mentioned in this post were based on samples that were not representative of the entire research community, but they obtained quite a similar result, which makes them quite credible. However, there is still a need to conduct a large, representative survey about researchers’ experiences with APCs.
Dallmeier-Tiessen, Suenje, et al. “Highlights from the SOAP project survey. What scientists think about open access publishing.” arXiv preprint arXiv:1101.5260 (2011).
Solomon, David J. “A survey of authors publishing in four megajournals.” PeerJ 2 (2014): e365.
Solomon, David J., and Bo‐Christer Björk. “A study of open access journals using article processing charges.” Journal of the American Society for Information Science and Technology 63.8 (2012): 1485-1495.
* – I have conducted some analysis of data archived by SOAP project on my own. I downloaded CSV file containing raw data and import it to IBM SPSS 21. Then I aggregate data for variable “Main research field” into 3 groups of disciplines, to make fractions bigger and to allow more detailed comparison between them.
1) Biological and Medical Sciences consists of following original SOAP categories: Biological Sciences, Agriculture and Related Sciences, Medicine, Dentistry and Related Subjects.
2) Other Natural Sciences and Engineering consists of following original SOAP categories: Astronomy and Space Science, Chemistry, Earth Sciences, Mathematical and Computer Sciences, Physics and Relative Sciences, Architecture, Building and Planning, Engineering and Technology.
3) Arts, Humanities and Social Sciences consists of following original SOAP categories: Historical and Philosophical Studies, Business and Administrative Studies, Creative Arts and Design, Education, Language and Literature Studies, Law, Social Sciences, Mass Communications and Documentation.
First image: A meeting of doctors at the university of Paris. From the “Chants royaux” manuscript, Bibliothèque Nationale, Paris. This image is in Public Domain.