On the interplay between normalisation, bias, and performance of paper impact metrics
作者：
Highlights：
• We evaluate paper-level impact metrics on field bias, time bias, and ranking performance.
• We evaluate mean-based, percentile-based, co-citation-based, and post hoc rescaling approaches to normalise citation scores.
• Percentile-based citation scores are less field and time biased than mean-normalised citation counts.
• No significant difference in ranking performance exists between percentile- and mean-normalised citation scores.
• Citation counts are always less time biased but always more field biased than PageRank.
摘要
•We evaluate paper-level impact metrics on field bias, time bias, and ranking performance.•We evaluate mean-based, percentile-based, co-citation-based, and post hoc rescaling approaches to normalise citation scores.•Percentile-based citation scores are less field and time biased than mean-normalised citation counts.•No significant difference in ranking performance exists between percentile- and mean-normalised citation scores.•Citation counts are always less time biased but always more field biased than PageRank.
论文关键词：Impact indicators,Ranking evaluation,Field normalisation,Field bias,Test data
论文评审过程：Received 14 October 2018, Revised 9 January 2019, Accepted 9 January 2019, Available online 24 January 2019, Version of Record 24 January 2019.
论文官网地址：https://doi.org/10.1016/j.joi.2019.01.003
