摘要：

【目的】针对引用的偏态分布,提出修正影响因子失真的方法。【方法】以影响因子不同的7种物理学综合期刊为样本,根据2011—2015年度各学术载文的被引频次,以被引量的分布曲线和洛伦兹曲线揭示期刊被引量在学术载文间的分布特征,引入基尼系数来描述引用分布的不均匀程度。【结果】期刊被引量在学术载文间均为右偏分布,基尼系数均大于0.4。【结论】被引量的非正态分布能导致影响因子出现不同程度的失真,以(1-G)为基尼修正系数,可有效修正影响因子失真。

关键词: 引用分布, 影响因子, 基尼系数, 失真, 基尼修正系数

Abstract:

[Purposes] This study aims to propose a method to correct the distortion of journal impact factor (JIF) due to skewed distribution of journal citation. [Methods] Taking seven physical multidisciplinary journals with different JIFs as samples, this study presented the characteristics of journal citation distributions by distribution curves and Lorenz curves based on the cited times of their scholarly items from 2011 to 2015, and introduced Gini coefficient to describe the unevenness of citation distribution. [Findings] All the distributions of journal citations are positively skewed with Gini coefficients higher than 0.4. [Conclusions] The skewed distribution of citation results in the distortion of JIF in certain degree. Applying (1-G) as the Gini correction factor to JIF can effectively correct the distortion.

Key words: Citation distribution, Journal impact factor, Gini coefficient, Distortion, Gini correction factor