The Essential Science Indicators (ESI) database of Thomson Reuters is widely used to evaluate and identify important, high-impact papers, but few researches have looked at attributes of the ESI database. Using the category of "chemical engineering" as an example, characteristics of the ESI database were described, and distributions of document type, language of the paper, and journal of publication were reported. Five indicators, total number of papers, first-author papers, corresponding-author papers, independent papers, and collaborative papers, were applied to evaluate publications by country, institution, and author. In addition, the number of authors cited, number of institutions cited, number of countries cited, number of subject areas cited, and number of journals cited were also used to evaluate highly-cited ESI papers. Results showed that journals with a higher impact factor did not necessarily to have more papers in the ESI. The most highly cited ESI papers had fewer authors, were more likely to be single-country papers, and in general had not yet reached a citation peak, or had an extended citation peak over several years. Self-citation does not appear to be an issue among them. The ESI database only includes papers that were published within the last ten years, and likely excludes some top-cited papers even before their citation peak is reached. We suggest that ESI criteria should be amended to include all papers but only consider citation frequencies within the last ten years.
|頁（從 - 到）||47-63|
|期刊||Malaysian Journal of Library and Information Science|
|出版狀態||已發佈 - 2013|
ASJC Scopus subject areas
- Library and Information Sciences