Metrics of activity in social networks are correlated with traditional metrics of scientific impact in endocrinology journals

Aims: Traditional journal level metrics, as the impact factor or Scimago journal ranking indicator (SJR) measure the impact of research on other researchers, but fail to assess the reach of these publications among wider audiences. This study aims to assess the correlation between SJR and metrics of presence on social networks of endocrinology journals. Materials and methods: Journals within the area of endocrinology, diabetes and metabolism were identified from the electronic portal Scimago Journal and Country Rank, and their metrics based on the number of citation collected. Different metrics of the presence on social networks, such as the number of followers and number of tweets, were assessed. The correlation between both metrics was evaluated by calculating the Spearman correlation coefficient. Results: Of the 232 journals evaluated, 28 (12.1%) had social networks. The median SJR (1.53 vs 0.60, p < 0.01), and H index (58.5 vs 22, p < 0.01) were significantly higher among journals with social networks. The correlation between the number of followers on twitter and the SJR was moderate (r = 0.60, p < 0.05), and it was better in journals with more than 500 publications in the last 3 years (r = 0.85, p < 0.05). The number of tweets also had a moderate correlation with the SJR (r = 0.59, p < 0.05) Conclusions: Our results suggest that the dissemination of research on social networks is moderately related to the traditional metrics of impact. Both types of metrics offer complementary information: while metrics based on social media may not perfectly predict the impact a scientific journal, it does have the advantage of immediacy.