基于大数据的h指数及其衍生指数的探索性因子分析研究

Exploratory Factor Analysis on H-Index and its Derivative Indexes Based on Big Data

h指数自2005年被Jorge E.Hirsch提出后,因其具有计算简单和适用范围广等特点而引发了学术界的广泛关注。许多研究者在h指数的基础上,对h指数进行修正,并提出了多种衍生指数。为深入了解这些同源性衍生指数间的关系,评估它们在不同层次人才评价中的应用效果,选择15种主要类h指数,以2008-2011年SCIE收录的农业科学领域114643篇论文为研究对象,采用探索性因子分析方法对15种指数进行相关性分析,结果表明在大数据情况下A指数和m指数与作者论文质量密切相关;g指数、h_w指数和R指数能更好地评价高产高被引作者;h_w指数和χ指数在评价高产低被引作者时效果较好;π指数、χ指数、m指数、A指数、h_w指数、e指数对低产高被引作者评价效果较好。

Since Jorge E. Hirsch proposed the h index in 2005, this indietor has been focused on by academia because of its characteristic of simple calculation and wide application. Many researchers made a lot of work to revise it, and proposed many types of derivative h indexes. In order to find out clearly the homology relationship among these indictors, this article selected 114 643 articles of ESI Agrieultural Scienee from SCIE database during 2008 to 2011 , to analyze their correlation by Exploratory Factor Analysis. The results showed that： in the ease of big data, A Index and m Index are closely related to the quality of articles; g Index, h Index and R Index are better to evaluate higher production and higher citation authors; h Index and X Index are better for higher production and lower citation authors; π Index, X Index, m Index, A Index, hπ Index, e Index are hetter for lower produc, tion and higher citation authors.