The Altmetrics Puzzle

Une partie de l’enseignement à l’enssib consiste à appréhender les (nouvelles) technologies de communication en y sautant nous-mêmes à pieds joints. Twitter, Netvibes,, blogs etc. doivent devenir un environnement naturel pour nous, futurs veilleurs/documentalistes/éditeurs/dompteurs d’oursins (ne rayez pas la mention inutile). Alors occasionnellement je me servirai de ce blog pour y déposer des articles que j’ai eus à rédiger dans le cadre du master ! Et on commence tout de suite avec un bref aperçu de ce que sont les altmetrics, ou comment mesurer l’impact en ligne de la recherche. Attention il y a quelques mots en anglais.

buchmesse-frankfurt-2014The Frankfurt Book Fair is in full swing and a number of interesting topics have already been discussed. A few established academic publishers participated yesterday to the Beyond Open Access panel, debating around the post Open Access era challenges. One word was especially interesting to them: data. Traditionally, publishers worry about citation count (the number of time something was cited in other works) and the related metrics, such as the Journal Impact Factor. It tells them directly how influential a paper is and how much others researchers have used it. However, open access content is generating another kind of data: forum discussions, news coverage, social network traffic etc. Articles and papers get discussed more and more on blogs, Twitter and comment sections from many websites. All of this does not result in direct citations but it still represents an impact. Therefore altmetrics have appeared: new ways to collect and correlate data in order to further evaluate the impact, importance or influence of science. Altmetrics provide insight on a paper’s diffusion across the Net by analysing page views, number of links, tweets, YouTube videos and many more.

What are altmetrics used for? The question was best formulated by David Colquhoun, a sharp critic of research metrics:

So far altmetrics have been used to answer the following questions: is the relevant community aware of this new information? Does it have an effect on other people’s work? Does it interest the public? They are fast-working tools, describing how a paper is shared almost in real time when citation counts take years to build up. But another question arises: what do we do with the answers? This is were interests divide. Researchers want to know if their work is discussed, by whom, what is being said. Most of them deeply care that their research found some usefulness by contributing even slightly to other people’s work. But there is also a commercial interest in performance: number of viewers, impact on the public’s opinion, the author’s and the publisher’s reputation. This is more often the publisher’s point of view and explains why Altmetrics scores are beginning to appear alongside the Journal Impact Factor. But it also interests some scientists who have decided to take some time to monitor the digital footprint of their work and include it in their CVs and dossiers.

Here is a 2 min video from the Altmetric company showing how to use their software to « monitor the online impact of digital scholarship »:

Altmetrics raise a lot of criticism. They are based on intrinsically sparse data. They are used to measure a paper’s performance on different websites, different communities or different media, which can lead to unfair comparisons. Some indicators (number of tweets) only show numbers while the content remains a mystery (were the tweets favorable or negative?). Researchers have panned altmetrics for giving too much place to comedy, controversial or shock-value articles and even pseudoscience. Indeed, a paper titled ‘An In-Depth Analysis of a Piece of Shit: Distribution of Schistosoma mansoni and Hookworm Eggs in Human Stool‘ will probably generate a lot of comments on social media, but do we learn anything from that? It does not help that paying attention to altmetrics could become a competitive factor amongst researchers, leading to less actual work and more self-promotion.6795008004_8046829553_z

Altmetrics are a puzzle because nobody really knows what to use it for and what it truly represents. But gradually people are getting their heads around it and try to question these altmetrics not to bring them down but to really learn from them. Open Access advocate Cameron Neylon has recently wrote an insightful piece about the future use of these new metrics:

As he points out, the applications of altmetrics we can currently think of do not relate to the importance of a paper but rather gives us clues about its nature. How it is shared and used tells us who is interested in it. It helps us map the new information pathways unfolding with the development of information technologies and the influence of the open access movement. There is much more to altmetrics than popularity measurements; if they have a chance to reveal to us how research spreads and becomes knowledge in the Internet age, they should make anyone who cares enthusiastic.



Laisser un commentaire

Entrez vos coordonnées ci-dessous ou cliquez sur une icône pour vous connecter:


Vous commentez à l'aide de votre compte Déconnexion /  Changer )

Photo Google+

Vous commentez à l'aide de votre compte Google+. Déconnexion /  Changer )

Image Twitter

Vous commentez à l'aide de votre compte Twitter. Déconnexion /  Changer )

Photo Facebook

Vous commentez à l'aide de votre compte Facebook. Déconnexion /  Changer )


Connexion à %s