论文部分内容阅读
为了研究期刊特征因子与其他文献计量指标间的关系,克服简单相关系数和普通回归分析的缺陷,本文基于汤森路透JCR纯数学期刊数据,采用面板数据(Panel Data)研究特征因子分值(Eigenfactor Score)及论文影响分值(Article Influence Score)与其他文献计量指标的关系。研究发现:特征因子分值和论文影响分值互为高度相关。即年指标及影响因子与特征因子分值无关,但和论文影响分值正相关。被引半衰期及5年影响因子与特征因子分值负相关,与论文影响分值正相关。总被引频次及期刊论文数量与特征因子分值正相关,与论文影响分值负相关。研究认为,特征因子的推出有利于期刊重视提高学术质量而不是提高文献计量指标值,对文献计量指标的相关关系的研究方法要重新审视,对于其他学科特征因子与文献计量指标的关系有待进一步研究。
In order to study the relationship between periodical characteristics and other bibliometric indexes and to overcome the shortcomings of simple correlation coefficient and ordinary regression analysis, panel data (Panel Data) was used to study the eigenfactor Score and Article Influence Score and other literatures. The study found that: the score of eigenvalue and essay influence scores are highly correlated with each other. The annual index and influencing factor have nothing to do with the score of eigenvalue, but they are positively correlated with the score of essay. Cited half-life and five-year impact factors were negatively correlated with the scores of the characteristic factors, and positively correlated with the scores of the papers. The total number of citations and the number of journal articles were positively correlated with the scores of characteristic factors, and negatively correlated with the scores of papers. The study suggests that the introduction of the characteristic factors is conducive to journals emphasis on improving academic quality rather than increasing the value of bibliometric indicators, the research methods of the correlation between the bibliometric indicators to be re-examined, the relationship between the other characteristic factors and bibliometric indicators needs further study .