期刊文献+

偏最小二乘法在公共部门绩效多元评估中的应用 被引量:8

Application of partial least squares to performance multi-evaluation for Dublic sectors
原文传递
导出
摘要 将偏最小二乘法(PLS)应用于公共部门绩效的多元评估研究,首先,回顾总结了PLS建模的特点及建模步骤,并就公共部门绩效评估与普通最小二乘法(OLS)进行了对比;其次,以上海市为例,建构了上海市公共部门绩效评估PLS方程模型;最后,对模型展开讨论认为:1.PLS建模方法是公共部门绩效多元评估的有效的、可行的方法;2.对于上海市公共部门绩效多元评估实践来看,评价指标的重要性次序,由大到小依次是:政务公开、办事效率、执法公正、履行职责、清正廉洁和服务态度;而综合来看,评价指标体系对各评价主体评价结果贡献的大小有所不同,影响程度由大到小依次是:上级评价、同级评价、专业评价和公众评价.这些结论,对于开展公共部门绩效评估的方法论研究,对于公共部门绩效多元评估的理论与实践都具有一定的参考价值. The paper applied the Partial Least Squares (PLS) approach to multi-evaluate the performance of public sectors. First, the PLS approach was reviewed, the PLS modeling process was interpreted and distinguished with Ordinary Least Square(OLS) on performance assessment of public sectors. Second, the PLS method was utilized to analyze public sector performance evaluated by four different bodies in Shanghai as an example, through analysis it was found that the different evaluated sector had one accord with the weighted order of different evaluation index from the four evaluators, the weighted order from larger to smaller was government openness, the management efficiency, executive fairness, fulfilled responsibility, incorruptibly honest, and the service attitude. Meanwhile, in the midst of the four evaluators, there were different contributions from integrated index of every sectors, the influenced via more to less was followed by the superior evaluator, parallel evaluator, professional evaluator and public evaluator. Finally, it was testified that the PLS was suitable method to multi-evaluate public sectors. These conclusions were meaningful upon researching and conducting performance multi-evaluation to public sectors.
作者 陈永国
出处 《系统工程理论与实践》 EI CSCD 北大核心 2009年第1期89-96,共8页 Systems Engineering-Theory & Practice
关键词 偏最小二乘法 公共部门 绩效评估 多元评估 partial least square (PLS) public sectors performance evaluation multi-evaluation
  • 相关文献

参考文献9

  • 1Joreskog K G, SSrbom D. LISREL 8 User's Reference Guide[M]. Chicago: Scientific Software International, 1996.
  • 2Chin W W. The partial least squares approach to structural equation modeling[C]// Marcoulides G A. Modern Methods for Business Research. N J: Lawrence Erlbaum, 1998, 237-246.
  • 3LohmLoller J B. Latent Variables Path Modeling with Partial Least Square[M]. Physica-Verlag, Heildelberg, 1989.
  • 4Chin W W. PLS-Graph User's Guide[M]. C T Bauer College of Business, University of Houston, USA, 2001.
  • 5Chou C P, Bentler P M, Satorrra A. Scaled test statistics and robust standard errors for non-normal data in covariance structural analysis: A Monte Carlo study[J]. British Journal of Mathematical and Statistical Psychology, 1991, 44: 347-357.
  • 6West S G, Finch J F, Curran P J. Structural equation models with non-normal variables: Problems and remedies[C]// Hoyle R H. Structural Equation Modeling. Thousand Oaks, CA, 1995:56 75.
  • 7Chen F, Bollen K A, Paxton P, et al. Improper solutions in structural equation models: Causes, consequences, and strategies[J]. Sociological Methods and Research, 2001, 29: 468-508.
  • 8McDonald R P, Ho M R. Principles and practices in reporting structural equation analysis[J]. Psychological Methods, 2002, 7: 64-82.
  • 9王惠文,付凌晖.PLS路径模型在建立综合评价指数中的应用[J].系统工程理论与实践,2004,24(10):80-85. 被引量:49

二级参考文献4

  • 1Guinot C, Latreille J. Tenenhaus M. PLS path modelling and multiple table analysis. Application to the cosmetic habits of women in Ile-de-France [J]. Chemometrics and Intelligent Laboratory Systems. 2001, 58: 247-259.
  • 2Lohm?ller J -B. Latent Variables Path Modeling with Partial Least Squares[M]. Heildelberg: Physica-Verlag, 1989.
  • 3Wold H. Soft modeling: The basic design and some extensions[A]. K G J reskog H Wold, Systems under Indirect Observation[C], North-Holland: Amsterdam, 1982. 1-54.
  • 4Wold H. Partial least squares[A]. Kotz S Johnson N L. Encyclopedia of Statistical Sciences[M], New York: John Wiley & Sons, 1985, 6:581-591

共引文献48

同被引文献105

引证文献8

二级引证文献69

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部