摘要：

【目的】分析Web of Science（WoS）数据库近十年撤销论文的引证特征,探究撤销行为对学术界的净化效果。【方法】分析撤销论文数量、被引频次等信息和期刊影响因子（impact factor,IF）的相关性,撤销时滞与论文被引频次的相关性;计算高被引撤销论文对IF的贡献率;评价撤销时滞对撤销前、后被引频次的影响。【结果】76种期刊（撤销论文数量≥5篇）IF和撤销论文数量、总被引频次、撤销后被引频次、篇均被引频次相关（均为P<0.05）。撤销时滞和撤销论文总被引频次相关（P=0.000）。随撤销时滞延长,撤销前被引频次逐渐增大。高被引撤销论文对期刊IF的贡献率为0.00%~2.00%。【结论】高IF期刊发表的、撤销时滞长的撤销论文对学术界的负影响较大;肯定期刊进行自我净化效果的同时,建议科研管理部门、期刊编辑和科研人员等加强对撤销论文及其不良影响的关注。

关键词: Web of Science, 撤销论文, 被引频次, 影响因子, 撤销时滞

Abstract:

[Purposes]This paper aims to analyze the citation characteristics of retracted articles which are obtained from the Web of Science（WoS）in nearly ten years and to explore the effect of self-correction of journals.[Methods]We made use of citation statistical analysis viaWoS from four aspects： the correlation among journals'impact factor（IF）and the number of retracted articles,citation of retracted articles; the correlation between time-to-retraction and the total citation of retracted articles; the contribution rates of highly cited retracted articles to IF;the effect of the time-to-retraction on the citation of pre-retraction or post-retraction.[Findings]As for 76 journals in which the number of retracted articles are more than five,there is positive correlation among IF and the number of retracted articles, total citation, citation of post-retraction, per paper citation （P<0.05）.Time-to-retraction and total citation have aweak association （P=0.000）.W ith the increase of time-to-retraction, the citation of pre-retraction is becom ing larger and larger.Contribution rates of highly cited retracted articles to IF are 0.00%~2.00%.[Conclusions]Retracted articles which have published in the higher IF journals or possess long time-to-retraction have an increasing negative impact on the academ ic community.The effect of journals'self-purification is sure.At the same time,it is recommended that the research management departments,journal editors and scientific researchers should pay more attention to retracted articles.

Key words: Web of Science, Retracted article, Cited times, Impact factor, Time-to-retraction