An Altmetric As an Indicator of a Publication’s Scientific Impact

Resumo

The results of an empirical pilot project focused on the association between classical bibliometrics—publication, citation index, and journal cited half-life—and an altmetric—the assessment of an article’s impact—are discussed. The analysis included an array of 37 200 domestic articles indexed in SCI-E in 2015. Two altmetrics are used: usage counts for the last 180 days, U1, and usage counts since February 1, 2013, U2. A significant Kendall rank correlation has been identified between citation indices and article-level metrics. A stronger correlation has been observed for long usage counts, U2. The relationship between usage metrics and traditional journal-level metrics (cited half-life) has been analyzed. A rather weak negative correlation between cited half-life and U1 (U2) has been revealed, which is described by an inverse logarithmic dependence. In the authors’ opinion, altmetrics should not be opposed to classical bibliometrics; they should be used as additional metrics to assess an article’s impact.