基于学科规范化引文影响力视角的SCI论文评价与管理思考——以某高校附属医院为例

Reflections on the evaluation and management of SCI papers based on the category normalized citation impact:Taking a university hospital as an example

目的从学科规范化引文影响力(CNCI)视角分析某高校附属医院2015—2019年发表的SCI论文基本情况与特征,为SCI论文科研管理政策调整提供参考依据。方法从该医院的科研管理系统中检索2015—2019年以第一或通讯单位完成的SCI收录论文,利用Web of Science、InCites数据库获取该组论文的期刊、影响因子、JCR分区及学科规范化引文影响力等指标数据,运用SPSS 21.0统计软件进行描述性及卡方检验。结果该高校附属医院在2015—2019年发表SCI论文3201篇,数量呈现增长趋势。论文以1<IF<3区间最多,占比44.92%;其次为3<IF<5区间,占比32.27%。论文集中在Q2区期刊,占比31.52%,Q1和Q3区期刊各占比25%。从学科规范化引文影响力角度来看,2015—2017年度论文的篇均CNCI小于1,2018—2019年度论文的篇均CNCI大于1,不同年度之间的CNCI无统计学差异。IF小于3的论文篇均CNCI小于1,IF大于3的论文篇均CNCI大于1,高影响因子论文的高CNCI比例更高;Q1、Q2区论文的篇均CNCI大于1,Q3、Q4区论文的篇均CNCI小于1,Q1区论文的高CNCI比例更高。结论不同科研管理视角下得出的评价结论并不一致,在期刊分区基础上的CNCI分层综合评价更为合理。在新时期不唯论文数量和影响因子等指标的情况下,遴选学科领域内的顶级期刊,同时将学科规范化引文影响力等综合评价指标应用于科技论文管理评价具有一定的现实意义。

Objective Analyze the basic information and characteristics of SCI papers published by an institution in 2015-2019 based on the Category Normalized Citation Impact(CNCI),to provide a reference basis for SCI paper management policy improvement.Methods Retrieve SCI papers completed by the first or correspondent unit from 2015 to 2019 from the scientific research management system of the institution.Obtain index data including journal information,impact factors,and CNCI of this group of papers from Web of science and InCites databases,and use SPSS21.0 statistical software to perform descriptive and chi-square tests.Results The university-affiliated hospital published 3201 SCI papers from 2015-2019,with a growing trend.These papers were most frequently published in the 1<IF<3 zone,accounted for 44.92%,followed by the 3<IF<5 zone,accounted for 32.27%.These papers are mainly in Q2 journals,accounted for 31.52%,while Q1 and Q3 journals each account for 25%.From the perspective of CNCI,the average CNCI of papers in 2015-2017 was less than 1,and the average CNCI of papers in 2018-2019 was greater than 1,with no statistical difference in CNCI values between years.The proportion of high CNCIs was higher for papers with high IF.The average CNCI per page for papers in Q1 and Q2 was greater than 1,and the average CNCI per page for papers in Q3 and Q4 was less than 1.The proportion of high CNCIs was higher for papers in Q1.Conclusions The evaluation conclusions reached under different research management perspectives are in-consistent.A comprehensive evaluation of papers based on journal divisions and CNCI is more reasonable.In a new era where indicators such as number of papers and impact factor are not the only thing that matters,it is relevant to select the top journals in the discipline and to apply comprehensive evaluation indicators such as CNCI to the evaluation of papers.