[目的/意义] 分析JCR最新公布的"影响因子百分位"指标的特点以及其对期刊评价的影响。[方法/过程] 从影响因子百分位的内涵分析入手,比较其与影响因子的统计学特征,在因子分析的基础上,采用多元回归和分位数回归研究影响因子百分位与其他文献计量指标之间的关系,同时采用Spearman相关系数分析其与其他文献计量指标的关系。[结果/结论] 研究结果表明,影响因子百分位改变了影响因子的统计学特征;影响因子较低的一些期刊,转换成影响因子百分位后相对值更低,其他期刊转换成影响因子百分位后相对值更高;影响因子百分位与其他文献计量指标相关程度中等;当影响因子百分位较低时,影响因子相关指标的弹性系数较高,当影响因子百分位处于中等时,总量指标及特征因子指标的回归系数较低;影响因子百分位并不适宜作为其中一个指标用于期刊多属性评价。
[Purpose/significance] This paper aims to analyze the characteristic of journal impact factor percentile indicator issued by JCR and its impact on journals.[Method/process] Starting with analysis on implication of journal impact factor percentile, this paper compares the statistical characteristic between it and journal impact factor. On the basis of factor analysis, it studies the relationship between journal impact factor percentile and other bibliometric indicators with multiple regression and quantile regression, and related coefficient of Spearman respectly.[Result/conclusion] The results show that:Journal impact factor percentile has changed the statistical characteristic of journal impact factor; Journals with low impact factor have lower relative values when converting into impact factor percentile, yet other journals have comparatively higher relative values; It has medium correlation between impact factor percentile and other bibliometric indicators; When the percentile of journal impact factor is low, the regression coefficient of its index is high, while the percentile is medium, the regression coefficient of aggregate indicator and eigenfactor indicator is low; Journal impact factor percentile is not an appropriate indicator to be used in multi-attribute evaluation of journals.
[1] [EB/OL].[2016-03-19].http://ipscience-help.thomsonreuters.com/incitesLiveJCR/glossaryAZgroup/g8/9586-TRS.html.
[2] BRADFORD B C. Source of information on specific subjects[J].Journal of information science, 1985, 10(4):173-180.
[3] HINTZE J L,Nelson R D. Violin plots:a box plot-density trace synergism[J]. The American statistician,1998, 52(2):181-184.
[4] LEYDESDORFF L.Alternatives to the journal impact factor:I3 and the top-10% (or top-25%?) of the most-highly cited papers[J]. Scientometrics, 2012, 92(2):355-365.
[5] BORNMANN L,MUTZ R. Further steps towards an ideal method of measuring citation performance:the avoidance of citation averages in field-normalization[J]. Journal of informetrics, 2011,5(1):228-230.
[6] CAMPANARIO J M. Analysis of the distribution of cited journals according to their positions in the h-core of citing journal listed in Journal Citation Reports[J]. Journal of informetrics, 2014,8(3):534-545.
[7] VINKLER P.Introducing the current contribution index for characterizing the recent, relevant impact of journals[J]. Scientometrics,2008,79(2):409-420.
[8] DAVIS P M. Eigenfactor:does the principle of repeated improvement result in better estimates than raw citation counts[J].Journal of the American Society for Information Science and Technology,2008,59(13):2186-2188.
[9] ADLER R, EWING J,TAYLOR P. Citation statistics[J]. Statistical science, 2009,24(1):1-26.
[10] 赵星. 期刊引文评价新指标Eigenfactor的特性研究——基于我国期刊的实证[J]. 情报理论与实践, 2009, 32(8):53-56.
[11] 黄贺方,孙建军, 李江. 期刊影响力评价指标之间的相关性研究[J]. 情报科学, 2011, 29(9):1322-1326.
[12] 俞立平, 潘云涛, 武夷山. 基于因子分析的学术期刊评价指标分类研究[J]. 图书情报工作, 2009, 53(8):146-149.
[13] 俞立平,武夷山. 学术期刊评价中标准分与原始分的比较研究[J]. 情报学报,2011,30(11):1187-1193.
