Will your paper be more cited if published in Open Access?

By Ernesto Spinak

Photo: SciELO.

Academia.edu is a well-known social network for scholars, established in 2008, which currently informs over 30 million registered users. The platform is used to share research papers, monitor their impact and follow up on any research in a particular area of ​​expertise. Its repository contains more than 8 million full-text articles published in open access (OA) and receives 36 million visitors per month.

In April 2015, a research conducted by six Academia.edu employees and the consulting company Polynumeral¹ on the growth of received citations to research publications that were deposited in its open access repository was distributed to 20 million users registered on its website, stating that the articles there deposited increased citations received by 83% within five years.

This work raises at least two questions:

• Is there really a positive relationship between OA and the increase in citations over time?
• Why, with so many similar websites functioning as repositories, is the increase reported by Academia.edu so impressive?

The reason why it is expected that OA has, as a consequence, more citations, is founded on three postulates:

• OA articles are easy to obtain, and are easier to read and to cite.
• In many cases they are published online first of formal publication, therefore, they have more time in favor to be cited that their commercial version.
• If prominent authors publish their articles in OA, then these articles would have a favorable bias in relation to subscription articles.

To prove the truth of these postulates, in the past decade in which the OA movement took off, very many papers that analyzed the correlation between OA and citations were published. A very interesting repertoire was recently compiled by the Scholarly Publishing and Academic Resources (SPARC Europe)² on this topic, and features 70 studies, of which 46 show evidence of advantage on citation increase, and the other 24 show negative or not significant results.

Therefore, the work published by Academia.edu should not be surprising, however, its results are rather surprising. It is interesting that this article, which has been published for about a year, has received almost no mention from other papers. A search on Google and Google Scholar only reveal documents that mention it, except for a work published by Phil Davis on the blog The Scholarly Kitchen, which makes a critical analysis of the results, “Citation Boost or Bad Data? Academia.edu Research Under Scrutiny”³.

The main criticism, also supported by the 35 following comments, has to do with the experimental design, i.e. a randomized control group, that is similar to the sample in every aspect, was not used. For example, in the control group there were editorials, corrections, retraction notes, letters to the editor, book reviews, obituaries, etc., that were not present in the sample. The conclusion of the first version of Phil Davis’ article is that “the big citation boost in Academia.edu could have nothing to do with open access, but rather fully explained by [use of] inadequate data”.

Therefore, the authors of the Academia.edu paper published on July, 7^th, a note seeking to clarify the differences with Phil Davis, “Academia.edu, Citations, and Open Science in Action”⁴, which motivated Phil Davis to add an addendum to his article dated August, 18^th, which states that the original results published by Academia.edu, which reported an increase of 83% had been reduced to 73%, which is currently informed by the Academia.edu website.

However, little above or below the increased number (73%, 83%, or whatever), the question is how to explain that Academia.edu get these results which are far greater than its larger competitors, with more years of operation, such as PubMed Central, arXiv, ResearchGate, Mendeley, which report values ​​around 3% growth. The authors of the Academia.edu paper indicate that actually it is not OA that is pushing the citation results, but the associated services they provide and the ease of articles recovery (discoverability), notifying users about all the news that are published on its repository. But… all similar services do the same… including journals individually to their subscribers.

So would Academia.edu be as great with a proportionate market share as important in relation to its competitors? If we stick to a research published in Nature in 2014⁵, only 29% of scientists and engineers consulted knew Academia.edu and only 5% visited it regularly, compared to 88% and 29% for ResearchGate, respectively, the social sciences having more participation in Academia.edu than other disciplines, so-called “hard sciences.” Most of the answers in the field of social sciences indicated that the reason to register for these services was to maintain a presence with a professional profile.

My reflection

Most published studies show unequivocally that OA increases the use of papers (downloads, tweets, likes). SciELO is a good example – the Brazil collection receives over 700,000 downloads per day. But that does not mean the increased use translates directly into increased citations.

According to David Crotty, one of Phil Davis’ article commentators,

OA does not greatly increases access to active researchers in a particular field of science, but it expands access to a wide spectrum of readers outside this formal community of selected research. This is a major OA benefit, putting publications in the hands of people like clinicians, decision-makers, politicians, ordinary citizens. This use is not reflected in citations because all these people do not publish research. However, within a community of researchers in a certain area, which already has good access to most journals on their specialty, until the present date, there are not enough measurable differences.

What then of the figure of 73%? I don’t know, let it be.

Notes

1. NIYAZOV, Y., et al. Open Access Meets Discoverability: Citations to Articles Posted to Academia.edu. Academia.edu. 1995, 39p. Available from: http://www.academia.edu/12297791/Open_Access_Meets_Discoverability_Citations_to_Articles_ Posted_to_Academia.edu

2. The open access citation advantage: list of studies and results to date. SPARC Europe. 2015. Available from: http://sparceurope.org/wp-content/uploads/2015/01/open-access-citation-advantage.xlsx

3. DAVIS, P. Citation Boost or Bad Data? Academia.edu Research Under Scrutiny. The scholarly kitchen. [viewed 08 January 2016]. Available from: http://scholarlykitchen.sspnet.org/2015/05/18/citation-boost-or-bad-data-academia-edu-research-under-scrutiny/

4. NIYAZOV, Y., et al. Academia.edu, Citations, and Open Science in Action. Academia.edu. 2015. Available from: http://medium.com/@academia/academia-edu-citations-and-open-science-in-action-4a24a6376573#.q15h1bu7g

5. VAN NOORDEN, R. Online collaboration: Scientists and the social network. Nature. 2014, vol. 512, nº 7513. pp. 126-129. DOI: 10.1038/512126a

References

DAVIS, P. Citation Boost or Bad Data? Academia.edu Research Under Scrutiny. The scholarly kitchen. [viewed 08 January 2016]. Available from: http://scholarlykitchen.sspnet.org/2015/05/18/citation-boost-or-bad-data-academia-edu-research-under-scrutiny/

NIYAZOV, Y., et al. Academia.edu, Citations, and Open Science in Action. Academia.edu. 2015. Available from: http://medium.com/@academia/academia-edu-citations-and-open-science-in-action-4a24a6376573#.q15h1bu7g

NIYAZOV, Y., et al. Open Access Meets Discoverability: Citations to Articles Posted to Academia.edu. Academia.edu. 1995, 39p. Available from: http://www.academia.edu/12297791/Open_Access_Meets_Discoverability_Citations_to_Articles_ Posted_to_Academia.edu

The open access citation advantage: list of studies and results to date. SPARC Europe. 2015. Available from: http://sparceurope.org/wp-content/uploads/2015/01/open-access-citation-advantage.xlsx

VAN NOORDEN, R. Online collaboration: Scientists and the social network. Nature. 2014, vol. 512, nº 7513. pp. 126-129. DOI: 10.1038/512126a

External links

Academia.edu – <http://www.academia.edu/>

Polynumeral – <http://www.polynumeral.com/>

About Ernesto Spinak

Collaborator on the SciELO program, a Systems Engineer with a Bachelor’s degree in Library Science, and a Diploma of Advanced Studies from the Universitat Oberta de Catalunya (Barcelona, Spain) and a Master’s in “Sociedad de la Información” (Information Society) from the same university. Currently has a consulting company that provides services in information projects to 14 government institutions and universities in Uruguay.

Translated from the original in Spanish by Lilian Nassi-Calò.