Various social networks aimed at researchers, academics and students have appeared in recent years. The most well known of these include ResearchGate, Academia and MyScienceWork.
How do they work? What are the risks involved in using them? What is their position with regards to Open Access?
Access
Registration
Registering on ResearchGate requires a university email address. You can link it to the ORCID ID.
Registering on Academia is done through a Facebook or Google account. A profile will be created with information that is taken automatically from these services.
Authors are asked to fill in their profile, university, experience, CV and list of publications. This data is not obligatory and will not be verified.
Risk of identity theft
Anyone can claim to be the author of a publication without their real identity being verified. As a result, the risk of abuse or identity theft is real.
ResearchGate and Academia are known for sending automatic invitation emails that use the names of registered members, without informing these members that these messages are being sent to their colleagues/contacts/relatives. Many consider identity theft to be an issue.
Access to documents
Registration is often necessary in order to access certain publications. This can lead to the automatic creation of a profile.
If you are already logged in to access a document, you may have a profile without actually having entered your details.
Differences compared to open archives
Features of open archives
The purpose of open archives is to disseminate publications, conserve them, centralise the results of scientific research and promote Open Access, which is supported by teaching and research institutions.
Open archives disseminate researchers’ publications as broadly as possible, whether via other Open-Access platforms, search engines, specialised platforms, etc.
Features of academic social networks
Academic social networks make it possible to establish links and exchanges between scientific communities, and communication networks and connections between researchers. They are more like the ‘Facebook’ or ‘LinkedIn’ of research than open archives.
Social networks care about populating their databases and establishing networks of connections between people more than anything, rather than disseminating the publications that are uploaded by researchers. This is why they do not share their data.
The permanence of data on these academic networks is not at all guaranteed. From one day to the next, these companies may go bankrupt, be sold to third parties, change their business model or even be attacked by publishers who force them to remove content en masse.
Requirements for uploading to open archives.
An increasing number of scientific-research funds require that publications from the research they fund be uploaded to open archives (See EU and F.R.S.-FNRS policy).
Academic social networks are not open archives and uploading publications to these tools does not satisfy the obligations of funding organisations. Publications must be uploaded to open archives or institutional directories like ORBi, PubMed, HAL, Zenodo, etc.
Copyright and authorisation to upload
Belgian copyright law
Belgian law contains an exemption to copyright that enables one to disseminate publications within a context that is limited to teaching and research. You can therefore upload files with restricted access to institutional directories (such as ORBi) that fall under this remit.
As ResearchGate, Academia and MyScienceWork are not institutional directories, nor are they platforms linked to education and research institutions, the exemption under Belgian copyright law does not apply to them.
Publishers’ policies
Authorisations granted by publishers specifically in regards to uploading to open archives or institutional directories do not necessarily apply to academic social networks.
You must systematically check whether the publisher's policy, or the periodical’s policy, explicitly authorises online publishing to academic social networks, even if the documents are not made openly accessible to everyone and only remain accessible to registered users.
As a result, a publisher may, at any moment, assert their rights over these platforms and require that your texts be immediately removed in their entirety. They may also initiate other action against these platforms and you as author/uploader.
In 2013, the publisher Elsevier made Academia remove 2,800 full texts from its interface, due to an infringement of dissemination law.
A solution: only publish links on academic social networks
To take advantage of academic social networks, without taking any risks, upload links to references that are already present on an institutional repository or any other open archive.
Retrieving your data
Retrieving your references
Academic social networks retrieve the data from publications that have been uploaded to open archives such as ORBi, PubMed, HAL, arXiv.org or Zenodo, or which have been published directly via Open Access. This is because some of your publications are in open archives and have already been disseminated on online, meaning they are also on these scientific networks.
If your publications are not already accessible online, you must add them manually or import them. However, they will not appear on academic social networks without having previously been referenced elsewhere.
Retrieving your full texts
Although the references can easily be retrieved online, the full texts must be uploaded by the authors.
These platforms sometimes automatically retrieve and disseminate the full texts of publications when they are made available by the source platforms, or when they fall under Creative Commons licences.
Interactions between ORBi and academic social networks
ORBi only allows social networks to retrieve bibliographic references and does not grant the right to disseminate the full text on other platforms without the authors’ consent.
Submitting your publications to ORBi makes it possible, if you choose, to have them included on ResearchGate, Academia, etc. as well. However, the reverse is not permitted.
In order to avoid the need to upload them several times, it is a good idea to first upload your publications to an open archive, then reference the link on academic networks.
Financial model
Commercial companies, first and foremost
ResearchGate, Academia and MyScienceWork have all raised funds worth several millions of dollars or euros from private investors. Although the creators of these platforms are known, the identity of the companies that finance them is relatively obscure and their interests are unknown.
Even though not all the services on offer are paid, these are commercial companies and not research or teaching institutions, or scientific or publishing circles.
No link with teaching
There are no links between academic social networks and teaching.
Usually, the .edu extension used by websites is reserved for teaching institutions (education). However, Academia (academia.edu) registered its domain name before these restrictions on the use of extensions came into effect. This can cause confusion, since Academia falsely appears to be a site linked to teaching institutions.
Moving towards a closed, paid service?
There is a fear that the actual aim of these platforms will be overridden by commercial companies, such as large publishers, for example, once they have reached a critical size, bringing them significant added value.
Another risk is that, once researchers become reliant on them, the services that are currently free will become paid, without it being possible to escape.
One may also question the wisdom of allowing a commercial intermediary to take control of the dissemination and conservation of scientific publications, without its financial model being clear and transparent or the permanent nature of the data protection being guaranteed.
Quality of the statistics
Unreliable figures
You must always be careful about downloading and viewing statistics that are displayed by certain websites. Many do not apply any filters, which means that the access figures do not always represent access by people, but by robots (bots, crawlers, spiders, etc.).
It is estimated that only 1/3 of total access is by actual people, with the remaining 2/3 coming from robots, which must therefore be eliminated.
On ORBi, only access/downloads by individuals are considered. Access by tens of thousands of robots is eliminated. If new robots are detected, the statistics are corrected and updated.
Opaque metrics
In 2012, ResearchGate introduced its own bibliometric index: the RG score. According to ResearchGate, the purpose of this index is to position a researcher within the scientific community based on how this author’s research is appraised by their colleagues.
We know some elements of this index: activity on ResearchGate, the number of questions/answers handled on the platform, the number of file downloads and profile views, as well as the impact factors for journals.
However, ResearchGate does not provide any specific information about how the RG score is calculated and refuses to explain or outline its algorithm. It has therefore been defined as ‘non-transparent and non-reproducible’. It has also been observed that the calculation has changed over time, though no clarifications about the changes have been made.
Although we may already be questioning the benefit of taking the impact of a researcher’s popularity into account, the mixture of data included in the calculation of the RG Score gives it no credibility or scientific validity.
Choose altmetrics
In order to measure the impact of research outside the world of academia, it is better to use tools such as altmetrics. This index is not linked to a specific platform or a researcher’s popularity, but instead collects data from various tools, globally, with a visible and transparent breakdown of how scientific publications are shared.
Do you have any questions?
If you have any other questions, or would like to share your experience with this topic, please feel free to contact us.
Learn more
- Benech, C., “Academia.edu : le réseau social scientifique préféré des SHS“, 12 avril 2013. [Online]
- Benech, C., “Protection et propriété des données sur Academia.edu et ResearchGate" , 14 mars 2014. [Online]
- Bouchard. A., “Pour une utilisation critique des réseaux académiques“, 14 février 2014. [Online]
- Contat, O. “Réseaux sociaux pour les chercheurs – A lire ou à relire“, 18 novembre 2013 [Online]
- Tellier-Becquart, N., “Academia.edu et ResearchGate : quelles conditions d’utilisation ?“, 25 mars 2014. [Online]
- Fortney, K. and Gonder, J. “A social networking site is not an open access repository“, 1 décembre 2015 [Online]
- Kraker, P. & Lex, E.: "A Critical Look at the ResearchGate Score as a Measure of Scientific Reputation", 26 mai 2015 [Online] 10.5281/zenodo.35401