Post-publication expert recommendations in faculty opinions (F1000Prime): Recommended articles and citations

Abstract

• Experts recommended published articles were cited more than other articles of the same biomedical journals except for cell. • Although correlations between recommendation level and citedness were observed, a linear relationship was not found between the two measurements. • The recommended articles were mostly classified as new finding, interesting hypothesis, and technical advance. • Experts’ star ratings correlate with their sentiment comments for top recommended articles. • Experts used hedged sentences to praise or criticize the articles in their comments. This exploratory study of the post-publication expert recommendations (PPER) of biomedical articles in Faculty Opinions observed whether the recommended articles were cited differently from other articles in the same journal. The collected data include 830 research articles published in Cell, Nature Genetics, Nature Medicine , and PLoS Biology in 2010 and their 205,976 citations in Web of Science (WoS) from 2010 to 2019. Of the 830 articles, 417 were recommended in Faculty Opinions. A recommendation made by a Faculty Member (FM) includes a star rating and optional classification and commentary. For Nature Genetics, Nature Medicine , and PLoS Biology , the recommended articles (dataset.FM) were cited significantly more than other articles (dataset.other). Certain correlations were found between recommendation level and citedness , but a scaled mapping showed no linear relationship between the two measurements. The majority of the articles reached a citation peak two years after publication. The most assigned classification tags are New Finding, Interesting Hypothesis, Technical Advance, and Novel Drug Target. Sentiment analysis of the 118 recommendations of the 30 top articles found that FM ratings were correlated with sentiment intensity level. The repeated measures ANOVA did not show the Matthew effect of citations. Suggestions include refining Faculty Opinions’ rating schema.