Analysis of Alternative Metrics of Research Impact: A Correlation Comparison between Altmetric Attention Scores and Traditional Bibliometrics among Plastic Surgery Research.

Abstract

Scholarly output has typically been measured by citation-based metrics such as the Hirsch index (h-index). The Altmetric Attention Score has emerged as a substitute to measure digital attention given to a project. This study aims to determine whether there is any correlation between h-index and the Altmetric score in the plastic surgery literature. Article metrics (full-text views, abstract views, PDF downloads, times e-mailed, Altmetric Attention Score, times tweeted, and number of citations by posts) were extracted from articles published in Plastic and Reconstructive Surgery over a 2-year period. Author metrics, including h5-index, were also collected. Pairwise correlations were performed using Spearman's rank correlation coefficient (r). A total of 1668 articles were published, with 971 included. Altmetric Attention Scores showed strong correlation with other article metrics (r = 0.48 to 0.97; p < 0.001) but weak correlation with h5-index (r = 0.14; p < 0.001) and sum of times cited without self-citation (r = 0.14; p < 0.001). It did not correlate with total publications, average citations per item, or sum of times cited. The h5-indexes showed strong positive correlation with other author bibliometrics (r = 0.66 to 0.97; p < 0.001); moderate correlation with times e-mailed (r = 0.41; p < 0.001); weak correlation with number of citations by posts (r = 0.10; p = 0.002); and no correlation with full-text views, abstract views, PDF downloads, and times tweeted. The Altmetric Attention Score and conventional senior author bibliometrics have weak positive correlation at best and appear to have distinct but complementary roles in measuring scholarly output.