【作者】 张光进；

【导师】 廖建桥；

【作者基本信息】 华中科技大学 ， 企业管理， 2008， 博士

【摘要】 知识员工在知识经济时代对于获取组织核心竞争力、推动社会可持续发展起着关键作用,如何管理和激励知识员工成为摆在研究者和实践者面前的重要课题。“没有评价,就没有管理”,由于知识员工工作的复杂性以及绩效表现的多样性,评价知识员工的绩效常常是一件棘手的工作。本文在“知识员工考评的关键是抓住其绩效特征和本质”这一基本观点以及权变管理理论的启发下,重点对知识员工绩效特征的测量和识别、绩效特征与考评方法的匹配规律进行了理论和实证研究,研究的主要结论有:(1)对知识员工下定义应主要抓住工作内容和工作性质两个方面,而对各类职业是否属知识工作进行判别,可以为厘清知识员工涵盖的对象提供操作化的界定。工作的非程序化程度和工作所需知识的“硬度”是知识员工分类的两个较好的维度。绩效是工作行为和工作结果的综合体观点更容易让人接受,但当工作具有某些特点时,绩效强调工作行为和工作结果的一个方面也是成立的。二维绩效模型广为学者们所引用,关系绩效考评是必要的,但由于其本质、考评的难度、以及中国文化等原因,本文的研究重点是任务绩效的考评。(2)对知识员工的绩效特征进行集中研究的文献非常少,通过对有关研究成果的分析可以发现,如果把宽泛定义下的知识员工作为一个不言自明的总体,其绩效特征几乎没有,但如果对知识员工进行分类,各类知识员工的绩效特征是比较明显的,这些绩效特征表现在绩效属性上的“显性”或“弱性”。总结已有的关于绩效特征的论述,同时分析考评方法设计过程中回答4w问题时需要考虑的绩效特征,可以发现绩效特征聚焦在五个绩效属性的差异上,进一步对访谈案例的内容分析表明,理论构建的绩效属性要素具有良好的信度和效度。这五个绩效属性要素分别是绩效行为是否可评价、绩效结果是否具体、绩效结果是否具有专业性、绩效结果是否具团队性、绩效结果取得是否需要长周期性。绩效特征则表现为在这五个绩效属性上“是”与“否”的性质,也即“显性”与“弱性”表征。(3)在文献研究和对访谈资料整理的基础上,编制初始的绩效特征问卷,经预测试、项目筛选、正式测试、信度、效度检验,形成最终的绩效特征量表。结果表明,五个绩效属性具有出色的辨别效度,绩效行为可评价性进一步可区分为绩效行为可视性、绩效行为标准化、绩效行为结构化三个维度,绩效结果具体性进一步可区分为结果可量化、内在确定性、环境稳定性三个维度,而绩效结果专业性、绩效结果团队性、绩效结果长周期性都是单维结构。正式的绩效特征量表为辨识绩效特征提供了操作化工具,而对绩效属性测量值的高低分组为绩效特征的识别提供了较为客观的标准,对所调查岗位在不同绩效特征中的分布可以看出,测量值高低分组确定的绩效特征标准具有较好的涵盖性,不同岗位的知识员工具有不同的绩效特征。(4)绩效考评方法的分类视角除了考评内容之外,还有考评者、考评对象、以及考评周期等角度,这些分类视角是相互独立的,多视角的分类体系既利于人们对绩效考评方法有全面的认识,也可以给管理者在选择和设计考评方法时提示应考虑的因素。各种考评方法没有绝对的优劣之分,权变管理理论对于本研究的启示,就是不存在普遍适用的考评方法,只有在特定的绩效特征条件下最为合适的考评方法。(5)知识员工对现行考评方法的接受程度并不理想,不同类型组织中的知识员工对现行考评方法的接受程度存在较大差异。对于不同岗位的知识员工而言,他们所期望的考评方法有明显的差别,而这些差别与他们具有不同的绩效特征有密切关系。(6)基于绩效特征的考评方法权变选择模型就是绩效特征和考评方法相匹配的系列研究假设,假设检验表明,实践中并不存在普遍适用的考评方法,只有在特定绩效特征下最为合适考评方法,具体而言:对于绩效行为可评价性和绩效结果具体性都比较低的知识员工,员工特征-工作行为-工作结果综合考评法比其他考评方法的可接受程度更高;对于绩效行为可评价性较低、绩效结果具体性较高的知识员工,工作结果类考评法比其他考评方法的可接受程度更高;对于绩效行为可评价性和绩效结果具体性都比较高的知识员工,结果类考评方法,或者行为类考评方法,或者结果-行为类考评方法都较为适宜,它们比其他考评方法的可接受程度更高;对于绩效行为可评价性较高、绩效结果的具体性较低的知识员工,工作行为类考评方法比其他考评方法的可接受程度更高。对于绩效结果专业性程度较高知识员工而言,专家评价法比非专家评价法、专家-非专家共同评价法的可接受性更高;对于绩效结果专业性程度较低知识员工,专家评价法、非专家评价法、专家-非专家共同评价法在被接受程度上没有明显区别。对于绩效结果团队性程度较高的知识员工,团队-个人相结合的绩效评价法比团队性评价法、个体评价法更容易被接受;对于绩效结果团队性程度较低的知识员工,个体评价法比团队性评价法、团队-个人相结合的绩效评价法比更容易被接受。对于绩效结果取得需要长周期性的知识员工,年度考评比季度考评、月度考评更容易被接受;对于绩效结果取得短周期性的知识员工,季度考评、月度考评比年度考评更容易被接受。系列研究假设的检验结果还表明,在考虑影响考评方法选择的次要因素时,最优的考评方法仍然都属于匹配于绩效特征的方法,这说明绩效特征是选择考评方法的首要决定因素。尽管对不同绩效特征下考评方法权变选择综合模型中的有些研究假设没有进行检验,但从已得到的检验结果来看,绩效属性对考评方法选择的主效应起决定作用,这就为管理者提供了选择考评方法的参考体系。论文最后还指出了本研究存在的不足以及后续研究方向。更多还原

【Abstract】 Knowledge workers play a vital role to gain organizational core competition ability and push society sustainable development in knowledge-base economy times. It is an important mission that how to manage and motivate knowledge worker for researchers and practitioners. No evaluating, no management. Appraising knowledge worker’s performance becomes a great challenge because knowledge workers’work complexity and performance diversity. Being enlightened by both one basic viewpoint that the key of appraising knowledge workers’performance is grasping their performance characteristics and its nature and contingency management theory, this dissertation carried out a theoretical and empirical study on measuring and identifying knowledge workers’performance characteristics, and the match rule between performance characteristics and appraisal methods. The main conclusion as followed:(1) Defining the‘knowledge workers’should concentrate on two aspects of both work content and work nature, moreover, to judge every kind of occupation whether or not belongs to knowledge work can provide an operational boundary for knowledge workers covering employee scope. The no-programmable degree and knowledge hardness within work are the two appropriate dimensions for classifying knowledge workers. It is easier to be accepted that performance is an integrative body of work behavior and work outcome, however, when a piece of work has some of characters, it is right that performance emphasizing either work behavior or work outcome. The task performance-context performance model is quoted extensively by scholars, appraising context performance is important, however, since its nature, the relative appraisal easiness, and Chinese culture, and so on; task performance appraisal is the importance in this dissertation.(2) There were a few literatures that concentrate on studying performance characteristics of knowledge workers, based on reviewing the relative literature, we can find that there were little performance characteristics if regarding all knowledge workers as one collectivity and not classifying them, however, if classifying knowledge workers in advance, then every type of knowledge workers has unambiguous performance characteristics which embody the‘explicit’or‘implicit’on some performance attributes. Summarizing the existed discussion about performance characteristics, as well as analyzing the performance characteristics needed to be considered when answer 4W questions in process of designing performance appraisal methods, we can find that these performance characteristics of every kind of knowledge workers embody the significant difference on five performance attributes, further content analysis about interview cases indicates that the five performance attributes theoretically constructed has fine reliability and validity. These five performances attributes are that whether work behaviors can be evaluated or not, whether work outcome is definite or not, whether work outcome is special or not, whether work outcome is team oriented or not, whether performance acquired need long time or not. Performance characteristics represent‘yes’or‘no’on these performance attributes, namely‘explicit’or‘implicit’embodiment above mentioned.(3) Initial performance characteristics questionnaire was compiled on the basis of studying literature and processing interviewing cases text, the final edition one was formed by preliminary test, items filter, formal test, reliability and validity test. The data analysis indicates that these five performance attributes have outstanding differentiating validity, further exploring factor analysis shows that the one work behaviors to be evaluable can be divided into three dimensions that work behaviors to be observable, work behaviors to be standardized, and work behaviors to be structural, the one work outcome to be definite can be divided into three dimensions that work outcome to be quantitative, work scheme to be mature, work environment to be stable, the other three ones that work outcome is special, work outcome is team oriented , and performance acquired need long time are all single dimension. The final edition questionnaire provide a operational tools to identify performance characteristics, and the high and low grouping about surveying value of performance attributes provide an objective standard to judge performance characteristics, through descriptive statistics on distribution of knowledge workers investigated at different performance characteristics, we can find that the high and low grouping about surveying value of performance attributes has extensive covering, knowledge workers in different positions have different performance characteristics.(4) As to the criterion for classifying performance appraisal methods, besides appraisal content, it is adopted that appraiser, appraising unit, and appraisal cycle, these criterion are independent separately, multiple criterion system of classifying performance appraisal methods not only to be help for understanding all kinds of appraisal methods, but also offer a series of directing factors to managers who select and design appraisal methods. Every kind of performance appraisal methods has no absolute superiority; contingency management theory tells us that there were no performance appraisal methods which can be applied universally; there was only one which is the best for knowledge workers whit a certain performance characteristics.(5) In practice, knowledge workers hold low acceptability to current appraisal methods, and there were great acceptability difference among knowledge workers in different type of organizations. As for knowledge workers in different position, they expect different appraisal methods, which are related to their different performance characteristics.(6) The contingency theory model about appraisal methods selected and performance characteristics is a series of research hypothesizes that a certain appraisal method matchs some of performance characteristics, the test about research hypothesizes shows that there are no performance appraisal methods which can be applied universally; there is only one which is the best for knowledge workers whit a certain performance characteristics, specification as followed:As to knowledge workers whose work behaviors not to be evaluable and work outcome not to be definite, trait-behaviors-outcome comprehensive appraisal method is fit better for them than other ones. As to knowledge workers whose work behaviors not to be evaluable and work outcome to be definite, outcome oriented appraisal method is fit better for them than other ones. As to knowledge workers whose work behaviors to be evaluable and work outcome to be definite, outcome oriented appraisal method, or behaviors oriented appraisal method, or outcome- behaviors comprehensive appraisal method is the best for them. As to knowledge workers whose work behaviors to be evaluable and work outcome not to be definite, behaviors oriented appraisal method is fit better for them than other ones.As to knowledge workers whose work outcome is special, peer review is fit better for them than nopeer review, and peer-nopeer together review. As to knowledge workers whose work outcome is not special; the three appraisal methods above mentioned are the same at being accepted.As to knowledge workers whose work outcome is team oriented, team-individual combinative appraisal method is fit better for them than team oriented appraisal method, and individual oriented appraisal method. As to knowledge workers whose work outcome is individual oriented, individual oriented appraisal method is fit better for them than both team oriented appraisal method and team-individual combinative appraisal method.As to knowledge workers whose performance acquired need long time, performance appraisal once a year is better than do it once a month or once a quarter. As to knowledge workers whose performance acquired is not long time cycle, performance appraisal once a month or once a quarter is better than do it once a year.The test result about research hypothesizes indicates that the best appraisal method is still belong to the ones which fit performance characteristics even if considering the less important factors that influencing appraisal method selected, which imply that performance characteristics is the most important factor that determining appraisal method selected. Although some research hypothesizes in the comprehensive contingency model of appraisal methods selected have not been tested yet because sample shortage, the known test result shows that main effect of performance attributes on appraisal methods selected is crucial. The research results about contingency model provide a reference system to select appraisal method.Finally, this dissertation pointed out the research shortcoming and further research filed.更多还原