Author-level Altmetrics for the Evaluation of Chinese Scholars

Guo Ying,Xiao Xiantao
DOI: https://doi.org/10.1007/s11192-021-04228-y
IF: 3.801
2022-01-01
Scientometrics
Abstract:Altmetrics has been widely recognized in the evaluation of articles, books, and journals. However, only a few of scholars have focused on author-level metrics which limits the wide application of altmetrics. Moreover, most of the commonly used altmetrics indicators are selected from international platforms, such as Google + , Facebook, and Twitter, which are not widely available to Chinese scholars. It will be biased if these indicators are directly used to evaluate Chinese scholars. This paper aims to construct an indicator system suitable for the evaluation of Chinese scholars based on altmetrics. We investigate representative platforms which are popular in China and similar with the frequently used altmetrics platforms and extract measurements from them. The principal component analysis is applied to reduce the multicollinearity among the selected indicators and objectively calculate their weights in the altmetrics system. Furthermore, the receiver operating characteristic curve is adopted to give evidence that the proposed altmetrics system is non-specific to the Social Sciences and the Natural Sciences, and the correlation analysis between the altmetrics scores and the traditional bibliometric indicators is implemented to verify its applicability to evaluate Chinese scholars.
What problem does this paper attempt to address?