Altmetrics – highlights attention on research results
About 180 dissertations were published in the KI Open Archive database in the first half of 2018. The database is maintained by KIB. All of these were so-called compilation theses, which means that the doctoral student writes a general introductory thesis frame (“kappa”) that summarises and discusses a number of constituent papers (scientific articles). At least two of these papers must be accepted for publication in peer-reviewed journals.
We identified the three constituent papers that currently have the highest Altmetric values. These three articles are part of the following dissertations.
- Ingela Hasselqvist-Ax: Dual dispatch and the importance of bystander CPR in out-of-hospital cardiac arrest (Altmetric score = 653)
- Erik Benson: Design and analysis of wireframe DNA nanostructures (Altmetric score = 335)
- Andrei Pyko: Long-term exposure to transportation noise in relation to metabolic and cardiovascular outcomes (Altmetric score = 280)
The altmetric score are 653, 345 and 280, respectively. But what constitute those scores of and how should we interpret to them?
If we, for example, take the paper in Ingela Hasselqvist-Ax dissertation, the score 653 is the sum of many "signs of attention". The article appears (are referred to/discussed) in at least fourteen news stories, nine scientific blogs, one policy document, 650 tweets, 163 Mendeley libraries, 47 facebook posts as well as in other web contexts. These "signs of attention" are weighted differently. Where a news story gets eight points, a facebook post gets 0.25. The sum of all these tracks is 653.
We can also see that the article is among the top 5% of all research outputs ever tracked in the database produced by the company Altmetric. We can also compare the score in other contexts at the altmetric record.
There are, of course, many reasons for the high altmetric score. One is the subject being dealt with. The article discusses how early cardiopulmonary resuscitation can reduce mortality at cardiac arrest. The conclusion is that the chance of survival more than doubles for patients suffering out-of-hospital cardiac arrest if civilian volunteers starts the resuscitation.
Another is that the article was published three years ago and has had time to draw attention. A third is that the article is published in a journal (NEJM) with high Impact Factor. But we also wondered if Ingela Hasselqvist-Ax and her co-authors did something special to communicate their results. On this question she answers this way:
- We collaborated with KI Press Office and issued a press release. Our hospital, Södersjukhuset, also spread information about the research on the web page and in other ways. One of our research funder, the Heart-Lung Foundation, also highlighted our results. Otherwise, there was presentation at the congress European Resuscitation Councils and presentations at all other meetings, symposia etc! And then Facebook, of course.
How should we interpret this score then? Well, there is a fairly vivid discussion about that. Some say that we can understand it as a complement to the other citation scores that exist around the publication. In comparison to citations, altmetric scores can accumulate faster, and may reflect impact almost instantly. It may also show attention from outside academia, such as media people and policy makers. In addition, this kind of metrics opens to measure other forms of scientific results than the published article, such as dataset, software or research presentations outside the scientific journal. Both types of metrics are needed to get a full picture of the research value.
Others believe that the underlying data is too uncertain and easy to manipulate to tell us anything about "impact". It is believed that altmetric score are not sufficiently reliable to be used as a measure of research quality. This includes the uncertainties that often stem from indicators that are in an early stage of development. To be taken seriously, the data needs to be robust, transparent and capable of handling manipulations, robots and spam.
In summary, you may say that there are signs that say that the article that Ingela Hasselqvist-Ax has participated in has made impressions in different contexts. But to know, how and why, one has to study the specific case more closely. Maybe the altmetric scores can enrich the research assessment by adding new perspectives (e.g.: visibility, societal impact).
The dissertation then? Well, according to our own statistics, it belongs to the top layer of "Views", namely 263.