Track research impact, use altmetrics!

Thursday, 18 April 2013 in Science by Astrid Pellieux

Measuring the impact of research is very important to judge a scientific work. Since years, traditional metrics, as citation metrics, provide information about the significance of a research work. However, the effectiveness of those indicators are now questioned. Hence, alternative metrics called altmetrics are now developing. Those new metrics are based on the social web and gather data from online platforms.

Evaluating the impact of research is the main way for the research world to make business decisions. It measures the performance of institutions and scientists, and thus, it allows different decision-making bodies to decide about allocations. For researchers, it’s also a way to find the most significant sources of information for building a personal project.


The evaluation of the impact of research is predominantly based on peer-reviewing and on the number of citations received by an article in a publication. This last indicator is known as the Impact Factor (IF) created in 1955 by Eugene Garfield. Every years, it gives the number of times an article has been cited in a journal within the 2 precedings years. Thus, the major problem with these tools is that the results of the measures won’t be known before a long time after the publication of the article. Moreover, the IF is now very questioned due to the controversial way it may be used nowadays. In fact, some people condemn self-citations and the constraint imposed by some journals to add unneeded citations.

Another issue is that the IF, like a majority of citation metrics, is a journal level metric. Thus, it reflects more the reputation of a journal than the impact of an article in the research world. Consequently, more article level metrics are needed. They will expand the scope of understanding of the research impact. Other citation metrics are indicators at the author level. It’s the case of the h-index which measures the number of time a scientist’s article had been cited in other publications. The h-index had been presented by Jorge E. Hirsch in 2005. However, the h-index isn’t a reliable indicator. It advantages more scientists with long career and certain field of research.

PLoS article level metrics

The development of the social web leads to new ways of communication for researchers. Thus, metric tools based on the web need to be developed. Those tools are added to the ones already used and are known as alternative metrics: the altmetrics.

A wider range of impact

Web development provides new ways for researchers to communicate their works. In fact, a large number of online tools designed specifically for scientists are now available to find, manage and share data. Non-scientific platforms provide powerful tools too.Tracking the impact of the research online is the purpose of altmetrics.

For instance, the open access movement gives a researcher the possibility to store his articles in some repositories such as PubMed Central and/or to submit them to an open access publisher like PLoS. Consequently, many articles are now rapidly available on the web for the scientific community. The way those articles are used and captured online by researchers is what altmetrics are after. For instance, they track how many times an article has been viewed, downloaded or bookmarked.

In addition, Altmetrics can also track citations. It includes the number of times an article is cited in other online journals or repositories, but also the number of times an article is cited by a social platform like Wikipedia.


Finally, the growth of scientific networks and social bookmarking platforms like Mendeley or CiteULike also constitutes an interesting source of information for the altmetrics. Moreover, numerous platforms where the science can be discussed such as social networks or science blogs are also taken into consideration. That particularly includes Twitter, Facebook, LinkedIn, Google+ and Wikipedia.

Altmetrics are not limited to only track citations. Thus, they cover a larger range of information sources than traditional metrics. Moreover, they also search impact on social platforms. Consequently, the impact of research on society can also be measured.

Organizations collecting altmetrics

In 2009, PLoS was one of the first publisher to start using metrics based on the web for its articles. The journal provides a free and open source application known as the PLoS Article-Level Metrics (ALMs). It searches mentions of PLoS articles from many sources such as other publishers and repositories, social networks (like Twitter and Facebook), science blogs, etc. Since then, many organizations have become specialized in this area. Some are displayed below.

Plum Analytics

Altmetric is a start-up that tracks online mentions of scholarly articles from many different sources including publishers, preprint databases, science blogs, Twitter, Facebook, YouTube, LinkedIn, etc. It provides paid service towards publishers or editors for commercial use. As a result, for non-commercial use, Altmetric is a free tool. Altmetric reports the collected data in the form of a donut where the different colors correspond to the data sources. For instance, the color blue represents Twitter. Finally, the Altmetric score of the analyzed article are given in the middle of the donut.

ImpactStory is a free and open source application that tracks research impact online from sources as diverse as some publishers and repositories, Mendeley, Twitter, etc. Moreover, ImpactStory doesn’t only track the impact of a scholarly article, but also searches mentions of other research artifacts such as software. The data collected are showed as a permalinked report.

Plum Analytics is a company that provides author and group (laboratory, department and institution) level metrics. The company collects data from many sources such as some repositories and publishers, Mendeley, Facebook, Twitter, Slideshare, etc. The data are displayed into a graph.

ScienceCard logo

ScienceCard is a free and open source service based on the PLoS Article-Level Metrics code. It collects mentions of scholarly article from Twitter, Mendeley, PubMed Central, CiteULike, Wikipedia and CrossRef. Other services should be added in the future.

To conclude, the altmetrics are very powerful tools to measure the impact of the research. They provide real-time indicators. Consequently numerous organizations become specialized on altmetrics. Moreover, more online publishers are now using altmetrics to give information about an article. It’s the case for example of PLoS, the Nature Publishing Group (NPG) and BioMed Central.

If you want more information, get it here:
Altmetrics: a manifesto