Method And Application Of Government Management Target Performance Evaluation

As an important composing part of government public management, government management target performance evaluation (GMTPE) is an effective rule tool for evaluating and improving government work, and its level and efficiency will directively influence the district development quantity and velocity. In the globle scope, competition among varied cities becomes more and more fury, the governments of varied cities in varied nations have observed that, GMTPE is an important cycles to improve government public management capality, and it has become one of the most vital elements in city core competition force.The GMTPE work is taken as the research background, based on the summary of foreign and domestic literatures, the multiattribute comprehensive evaluation is introduced as main theory basis, and the GMTPE work in Shenyang city is taken as practical application proving. Several issues of GMTPE system are studied according to the thought that is from theories to applications. The detailed work is as follows:1) Features and targets of GMTPE are analyzed, the commonness problems in GMTPE from equity degree, scientific degree and economic degree, and the inner reasons for the problems are analyzed.2) On the basis of analyzing present problems and reasons in index system of GMTPE. The target location and principle for improving index system of GMTPE are proposed. The two level index system of GMTPE for local government is established from economic construction, culture construction, society construction, party construction and soft environment construction six aspects.3) Feature reduction is a key technique for selecting features of performance evaluation, to improve the evaluation efficiency of government task performance, the grey predicting and rough set based feature reduction method is studied. For the vacancy values of performance evaluation features in past year, the grey theory based disposing method for imperfection information is given to repair the vacancy values. An approximation precision based continuous feature dispersing method for time series is presented to obtain varied dispersed spans for each continuous feature. Based on the differentiating matrix, a variational precision rough set based feature reduction is designed, and the anti-noise capability is improved. The above method is applied to the practice of features selection for Shenyang city task performance evaluation. It shows that the obtained evaluation by above reduced feature set is almost identical with that by all features, and the high error tolerance capability.4) The value pretreatment methods of GMTPE are studied, which includes the consistency treatment of single attribute value, integrating of the absolute quantity and the relative quantity, dimension eliminating and so on. The disposing method for single index value of economic and noneconomic category index are respectively introduced, the combination technique of absolute value and relative value is especially studied.5) In the integrating cycle of multiattribute values, the rank correlation analysis based weight parameters endowing method for level indexes is given; the conventional scatter degree method is adapted and employed to endow weight parameters for level two indexes. To validate the validity of the GMTPE method for local government, the GMTPE work in Shenyang city is taken as the research’ background, and above method is employed to the WPE practice in Shenyang city. It is showed that by above method presented in this paper, the development quality difference among the evaluated districts and the collectivized development trend are enlarged.6) According to the practical condition of GMTPE in China, several suggestions are proposed on GMTPE from environment constructing, mechanism perfecting, technique innovating, operation standardizing and cost controlling.Finally, the problems and directions of the further research in future are given in this dissertation.