论文部分内容阅读
【目的】分析学术期刊来源指标与目标指标之间的关系以及由此导致的对期刊评价的影响。【方法】以JCR2015年数学期刊为例,采用典型相关分析对期刊来源指标与影响力指标的关系进行研究。【结果】实验结果表明:特征因子是期刊影响力的主要指标;期刊来源指标与影响力指标相关关系显著,以特征因子为主的影响力指标与载文量的相关系数最高,其次是引用半衰期,最后是文献选出率;总被引频次、被引半衰期、影响因子、影响因子百分位对特征因子的贡献较大,而其他诸如他引影响因子、5年影响因子、即年指标的贡献较小;影响力指标比来源指标包含更多信息量。【局限】来源指标与影响力指标的关系尚需进一步检验。【结论】从期刊多属性评价角度,影响力指标的权重应大于来源指标的权重;在影响力指标中,有必要增加特征因子分值与标准特征因子的权重,并合理分配总被引频次、被引半衰期、影响因子、影响因子百分位的权重,另一方面要适当降低他引影响因子、5年影响因子、即年指标的权重。
【Objective】 To analyze the relationship between source index and target index of academic journals and the consequential impact on journal evaluation. [Methods] Based on the JCR2015 academic yearbook, the author used canonical correlation analysis to study the relationship between periodical source index and influence index. 【Result】 The experimental results show that the eigenfactor is the main index of journals’ influence; the correlation between periodical source index and influence index is significant; the correlation coefficient of eigenvector mainly has the highest correlation coefficient with the number of articles; secondly, the reference half-life , And finally the selection rate of the literature. The total cited frequency, the cited half-life, the impact factor and the impact of the percentile of the impact factor have a greater contribution to the feature factor, while others such as the impact factor, 5-year impact factor, Contributions are smaller; impact indicators contain more information than source indicators. [Limitations] The relationship between source indicators and influence indicators needs further examination. 【Conclusion】 From the perspective of multi-attribute evaluation of journals, the weight of influence index should be greater than the weight of source index. In the influence index, it is necessary to increase the weight of the feature factor score and the standard feature factor, Cited half-life, the impact factor, the weight of the percentile of percentile, on the other hand, it is necessary to appropriately reduce the influence factor of him, the impact factor of 5 years, that is, the weight of the annual index.