Skip to main content
. 2020 Jul 6;22(7):e15607. doi: 10.2196/15607

Table 2.

Characteristics of correlational studies of very good quality (n=7).

Study ID and reference Health research area/unit and period of analysis Metrics reported Results Methodological quality indicatorsa




1 2 3 4
Costas, 2015 [61] Biomedical and health sciences; 217,115 articles in health sciences available from WoSb (2011-2013) Social media: Altmetrics-Bibliometrics: Citations (WoSb) Positive relationship between number of Altmetrics and the average citation impact and citation scores + + + +
Delli, 2017 [64] Dental medicine; 100 articles with highest AASc from Altmetric Explorer and JCRd (2015) Social media: Altmetrics-Bibliometrics: Citations (Scopus) No significant correlation between Altmetrics and citations + + + +
Eysenbach, 2011 [17] Medical informatics; 208 tweets including links to 286 JMIRe articles (2008-2010) Social media: Twitter-Bibliometrics: Citations (Google Scholar and Scopus) Moderate correlations + ++ + ++
Haustein, 2014 [16] Biomedical and health sciences; 1,431,576 biomedical and health sciences articles available on PubMed (2010-2012) Social media: Twitter; Altmetrics-Bibliometrics: Citations (WoSb) Moderate correlations + + + +
Knight, 2014 [35,36] Organ transplantation; 6979 articles with citation data available; 1346 with social media mention (2011-2012) Social media: Altmetrics-Bibliometrics: Citations (Scopus) Significant correlations between social media mentions and citations + + + +
Livas, 2018 [71] Orthodontics; Top 200 articles in orthodontics available from Altmetrics Explorer (2017) Social media: Altmetrics-Bibliometrics: Citations (Scopus) No correlation was observed between Altmetrics score and citations + + + +
Maggio, 2018 [72] Health profession education; 2486 articles with Altmetrics published in health profession education (2013-2015) Social media: Altmetrics-Bibliometrics: Citations (WoSb) Significant correlations between Altmetrics and bibliometrics, but moderate effects + + + +

a1: appropriately adjusting for time of the social media metric (+); 2: appropriately adjusting for confounders such as article type (+) and seasonality/time factors (++); 3: appropriately exploring correlations by including scatterplots (+); 4: appropriately reporting nonlinear correlations tests and statistics (+) as well as log-linear relationship tests (++).

bWoS: Web of Science.

cAAS: Altmetrics attention score.

dJCR: Journal Citation Reports.

eJMIR: Journal of Medical Internet Research.