An Altmetric lens on retractions and misinformation

By analyzing the Altmetric Attention Score for retracted publications, two recently published papers offer insights into the potential of Altmetric to understand the makings and evolution of scientific misinformation. Could this be a step toward combating the spread of misinformation and strengthening trust in research?

Altmetrics and the Altmetric Attention Score were born from the need to gauge the attention and impact of academic research in a digital age that has seen an acceleration in information dissemination and a burgeoning of platforms for sharing scientific information. Originally devised to quantify the extent and nature of attention garnered by research outputs, the Altmetric Attention Score seems to have found applications beyond its primary intent. A blog published in June 2023, “Beyond the Doughnut: Five Ways to Use Altmetrics for Academic Success,” featured in the London School of Economics Impact Blog, gave insights on how one can squeeze more out of altmetric scores. 

Now, two recent papers show that the Altmetric Attention Score can have applications pertinent in this era of misinformation in the digital age. The papers Investigating Scientific Misinformation Using Different Modes of Learning (a conference paper presented at the Proceedings of the Workshop on Scientific Document Understanding, AAAI Press, 2023) and Evolution of retracted publications in the medical sciences: Citations analysis, bibliometrics, and altmetrics trends published in Accountability in Research, looked into how the data behind the Altmetric Attention Score can be used to analyze scientific misinformation. Both papers examined the scores for retracted publications.

The authors of Investigating Scientific Misinformation Using Different Modes of Learning were interested in understanding if it was possible to tease out themes of scientific thought that are poor quality and popular from those that are not. The team used the Altmetric Attention Score as a metric for publication popularity to categorize articles into low-popularity and high-popularity subsets. “We use retracted publications as a proxy for identifying publications with a high potential for misinformation and the Altmetric Attention Score as a proxy for publication popularity,” the authors write. The authors analyzed scientific misinformation across different research areas and measured the prevalence of scientific misinformation. 

The second study, Evolution of retracted publications in the medical sciences: Citations analysis, bibliometrics, and altmetrics trends, looked into retracted articles, their presence, and evolution in online platforms. One of the questions the team sought to answer was: To what extent were retracted articles shared on social platforms? To answer this research question, they tapped into the Altmetric Attention Score to gain a comprehensive view of these articles’ online engagement. Of the retracted publications examined, the highest Altmetric Attention Score of 2,972 was associated with a withdrawn paper on COVID-19 research, which garnered 4,808 shares on Twitter. A thought-provoking observation was that certain articles continued to attract attention on social media even after being retracted, ostensibly from individuals without specialized expertise. The authors, however, note that the difficulty in attributing mentions before or after retraction underscores the nuanced approach required for interpreting Altmetric data.

Both papers discussed show that as research ecosystems and the means of information dissemination continue to evolve rapidly, tools such as the Altmetric Attention Score have the potential to find versatile uses, from tracking attention within and outside academic settings and assessing academic impact to analyzing scientific misinformation and possibly more. 
To learn more about the Altmetric Attention Score and how it can be used, visit the donut and Altmetric Attention Score page.