The performance issues of state-owned enterprises have been concerned by common people. In order to improve state-owned enterprises¡¦ performance effectively, the government should understand their regular operation efficiency adequately and then establish an appropriate evaluation system. However, the roles and functions of state-owned enterprises have changed significantly since the environment changed and economics progressed in Taiwan. As a result, the views and means toward performance assessment are kind of imbalanced. Therefore, this study researched Taiwan state-owned enterprises¡¦ performance assessment system to assess and review the related problems about existing evaluation methods. The aspects include indicators of evaluation, functions of evaluation and executing the system practically. According to document analysis, this study reviewed literature about performance evaluation, and analyzed rules, methods, indicators and overall report of performance assessment from RDEC (Research, Development and Evaluation Commission, Executive Yuan) over the years to realize the meaning of existing performance assessment system and the challenges of implementing the system. Three comments toward this evaluation system are as follows. First, regulate the relationship between the government and each state-owned enterprise appropriately to avoid or reduce government organizations¡¦ improper interference with state-owned enterprises. Then, there will be more objective and more correct evaluation outcomes. Second, the existing grading mechanism may let importance of evaluation indicators be diluted by subordinate ones. The grading mechanism should be established based on operation features of state-owned enterprises and their budget plans. Third, the existing evaluation emphasizes annual short-term performance and lacks of consideration of long-term development and innovation ability. So, the study suggests that government may apply BSC (Balanced Score Card) to adjust the contents of evaluation standards and items. Key words: State-owned enterprises, Performance assessment system, Performance evaluation

No description available.

No description available.

En dépit de ses qualités, le Rorschach en Système Intégré présente des problèmes psychométriques substantiels que le nouveau système Rorschach Performance Assessment System (R-PAS) ne permet pas de résoudre. Aussi, l’objectif principal de la thèse que nous défendons est de développer une approche psychométrique du Rorschach qui soit satisfaisante sur les plans statistique et clinique. Une Analyse Parallèle et Une Analyse en Composantes Principales réalisée sur un échantillon normatif de 695 participants Belges, Français et Finlandais permet de définir un modèle du Rorschach en 12 Composantes. Si ces dimensions sont cohérentes avec les principes de cotation et les connaissances empiriques sur le Rorschach, elles font également apparaître certaines difficultés de cotation, des aspects du Rorschach qui ont été négligés dans la recherche, de même qu’elles remettent en question certains indices du Système Intégré. Ce modèle est appliqué à deux problèmes empiriques : l’expression de particularités culturelles au Rorschach et la capacité du test à identifier des stratégies de résolution de problème (qui est un aspect central du Système Intégré). Les normes américaines du Système Intégré ainsi que les normes internationales du R-PAS ne peuvent s’appliquer de manière universelle et il faut donc recourir à des valeurs de références nationales. De plus, le Rorschach ne permet pas d’identifier des stratégies de résolution de problème de manière fiable. Ces études montrent que certains principes fondamentaux du Système Intégré et du R-PAS sont à remettre en question et qu’il est nécessaire de développer un nouveau système d’interprétation du Rorschach selon une approche psychométrique. / In spite of its quailties, the Rorschach Comprehensive System (RCS) presents substantial psychometric issues which are not solved by the new Rorschach Performance Assessment System (R-PAS). Thus, the main objective of this dissertation is to develop a satisfying psychometrical approach of the Rorschach test. A Parallel Analysis and a Principal Component Analysis performed on a normative sample of 695 Belgian, French and Fins participants define a 12 Components model of the Rorschach. If these dimensions are coherent with scoring principles and empirical evidences for the Rorschach they also reveal some scoring issues, neglected aspects of the Rorschach in the research field as well as they question the validity of some indices of the RCS. This model was applied to two empirical problems : the cross-cultural study of the Rorschach and its ability to assess problem solving strategies (which is a key feature of the RCS). The american norms of the RCS and the international reference values of the R-PAS cannot be used universally and it is necessary to use national norms. Moreover, the Rorschach does not assess problem solving strategies in a satisfying manner. These studies demonstrate that some fondamental principles of the RCS and the R-PAS must be questionned and the necessity to develop a new interpretation system for the Rorschach based on a psychometrical approach.

No description available.