摘要
为了深入分析非线性评价带来的评价结果实际差距扭曲问题,本文采用回归分析法将非线性评价转换为线性评价并进行深入分析。研究结果表明,不同评价方法评价结果的用途不同,基于排序的评价其结果不能进行差距比较,线性评价在不同分值情况下改进难度不一样,非线性评价会带来评价对象实际差距的数据扭曲;TOPSIS法是一种惩罚落后,略微鼓励先进的评价方法;灰色关联法是一种防止骄傲,帮助落后的评价方法;在竞争中,期刊必须认识到非线性评价所带来的与竞争对手差距扭曲问题,从而明确差距,进行改进。
This paper changes nonlinear evaluation to linear evaluation by regression in order to analyze realistic disparity tortuosity of nonlinear evaluation.The results show that different evaluation methods' usage isn't the same.The results based on ranking can' t be compared.Linear evaluation' s improvement is different at different score and nonlinear evaluation may cause evaluation results' tortuosity.TOPSIS may encourage excellent and discourage lagging behind.Gray relation analysis may discourage excellent and encourage lagging behind.Academic journals should analyze realistic disparity tortuosity of nonlinear evaluation and do something to improve.
出处
《情报学报》
CSSCI
北大核心
2011年第4期435-440,共6页
Journal of the China Society for Scientific and Technical Information
基金
国家自然科学基金资助项目(70973118)
ISTIC-THOMSON科学计量学联合实验室开放基金项目(IT2009001)
关键词
学术期刊
多属性评价
非线性评价
实际差距
academic journal
multiple attribute evaluation
nonlinear evaluation
realistic disparity