Abstract: PO1378
An Analysis of Scientometrics and Social Media in Nephrology
Session Information
- Educational Research
October 22, 2020 | Location: On-Demand
Abstract Time: 10:00 AM - 12:00 PM
Category: Educational Research
- 800 Educational Research
Authors
- Vaghjiani, Nilan Ghanshyam, Virginia Commonwealth University School of Medicine, Richmond, Virginia, United States
- Shah, Nishi, Virginia Commonwealth University School of Medicine, Richmond, Virginia, United States
- Lal, Vatsal, Virginia Commonwealth University School of Medicine, Richmond, Virginia, United States
- Desai, Tejas, NOD Analytics, Charlotte, North Carolina, United States
- Vinnikova, Anna K., Virginia Commonwealth University, Richmond, Virginia, United States
Background
In the past decade, the use of social media to disseminate scientific literature, particularly in the Nephrology community, has exponentially increased to educate, network, mentor. The hallmark of scientometrics has traditionally been based on Journal Impact Factor (JIF), calculated from the citations of each article. It has been previously demonstrated that twitter mentions of published works, correlate with citations, and therefore JIF, in the fields of Urology, Biomedical Science, and Ecology. However, this relationship has yet to be established in the field of Nephrology.
Methods
The top 5 journals in Nephrology, based on impact factor (Kidney International, Nature Reviews Nephrology, AJKD, JASN and CJASN), published 76 articles in January of 2018 in print. Altmetrics bookmarklet was used to collect twitter demographics on each article (number of tweets, by whom, and number of followers). Citation data was sourced from Web of Science’s InCites Journal Citation Reports. Articles were categorized as ‘highly cited or tweeted’ when they were > 75th percentile of citations or tweets, and ‘less cited or tweeted’ at < 25th percentile.
Results
Of the article cohort, the most common article types were clinical investigations (42%), followed by basic research (20%), and reviews (10%). The citation mean was 18.29 ± 15.87 citations per article, while the twitter mentions mean was 28.38 ± 68.97 tweets. The Spearman correlation coefficient for Twitter mentions and citations was 0.25 (p = 0.026). The odds ratio of an article being both highly cited and highly tweeted was 3.6 (CI: 0.71 to 18.25). The relative risk showed that highly tweeted articles were 1.87 (CI: 0.83 to 4.19) times more likely to be highly cited than less cited. Finally, the peak tweets (279) occurred in October of 2017 while the peak in citations occurred in 2019.
Conclusion
The preliminary analysis showed a significant but weak correlation of twitter mentions and citations. This suggests that twitter increases an article’s reach, its likelihood of becoming cited, and therefore the JIF. Future directions will include exploring a larger sample size and confounding factors such as word count, number of authors, and number of citations.