[目的/意义]在科技评价中，不同评价指标的重要性和地位是不同的，采用科学合理的方法进行关键指标的测度具有重要意义。[方法/过程]用指标信息熵和离散系数表示信息量，用指标权重或模拟权重与指标信息量的几何平均值表示关键指标系数。以JCR 2015经济学期刊为例，采用TOPSIS进行评价，并计算指标的信息量、模拟权重和关键指标系数，发现特征因子、标准特征因子、总被引频次是关键指标。[结果/结论]研究表明，关键指标系数是一种有效的关键指标筛选方法；不同评价方法关键指标是不一样的；本方法同样适用于采用信息量赋权的评价方法；防止遗漏重要指标，保证评价指标的完备性是本文方法的前提条件；关键指标是双刃剑，既是科技管理工作的重要抓手，也要防止其中的不正当竞争。
[Purpose/significance] Different evaluating indexes have different degrees of importance and various statuses in the evaluation of science and technology; therefore, it's significant to adopt a scientific and reasonable method to evaluate key indexes.[Method/process] This paper applied the index information entropy and discrete coefficient to express the quantity of information, and used the index weight or the geometrical mean of the simulated weight and the index information to express the key index coefficient. By taking JCR2015 economic journals for instance, it could be found that characteristic factors, standard characteristic factors and total cited frequency were the key indexes when taking TOPSIS for an example to evaluate and calculate the information quantity, simulated weight and the key index coefficient of the indexes.[Result/conclusion] The research shows that the key index coefficient is an effective measure to screen the key indexes. The key indexes of various evaluation methods are different. This method is applicative to the evaluation methods which apply information quantity to give the weight. The key index is a double-edged sword which is an important handle in science and technology management, but we should avoid its illicit competition.
[1] HARTLEY R V L. Transmission of information[J]. Bell system technical journal,1928(7):535-563.
[2] SHANNON C E, WEAVER W. The mathematical theory of communication[J]. Mobile computing and communications reviews,1948,5(1):3-55.
[3] CRISTIAN C S, MONICA D. Entropic measures, Markov information sources and complexity[J]. Applied mathematics and computation,2002,132(2):369-384.
[4] DADPAY A, SOOFI E S, SOYER R. Information measures for generalized gamma family[J]. Journal of econometrics, 2007, 138(2):568-585.
[5] PAWLAK Z. Rough sets[J]. International journal of computer and information sciences,1982,11(5):341-356.
[6] 秦敏. 基于信息量与理想点法的图书馆读者满意度评价模型[J]. 农业图书情报学刊,2010(10):91-93.
[7] 王晓勇,楼佩煌,唐敦兵. 基于信息量的不确定型多属性决策方法[J]. 运筹与管理,2012(1):64-69.
[8] 左淑霞,席建锋,肖殿良,等. 特色交通标志设计机器信息量度量方法研究[J]. 中国安全科学学报,2010(11):139-144.
[9] FRAGKIADAKI E G, EVANGELIDIS G,Samaras N, et al. F-Value:measuring an article's scientific impact[J]. Scientometrics,2011,86 (3):671 -686.
[10] FRANCESCHET M. The difference between popularity and prestige in the sciences and in the social sciences:a bibliometric analysis[J]. Journal of informetrics,2010,4(1):55-63.
[11] 苏新宁.构建人文社会科学学术期刊评价体系[J]. 东岳论丛,2008(1):35-42.
[12] 叶继元. 图书馆学期刊质量"全评价"探讨及启示[J]. 中国图书馆学报,2013(7):83-91.
[13] 俞立平, 潘云涛,武夷山. 科技评价中不同客观评价方法权重的比较研究[J]. 科技管理研究,2009(7):148-150.
[14] 俞立平, 宋夏云. 期刊评价中非线性评价方法选取的检验研究[J]. 中国科技期刊研究, 2014, 25(8):1063-1067.
