1、 外文翻译 原文 The Big Question for Performance Management: Why Do Managers Use Performance Information? Material Source: University of WisconsinMadison Author: Donald P. Moynihan, Sanjay K. Pandey The most widespread governmental reform in recent decades has been the requirement for agencies to track and
2、 measure strategic goals, targets, and achievements (Brudney, Hebert and Wright 1999; Moynihan 2008). Within our growing state of agents, performance goals underpin contractual forms of accountability, the means by which webs of connected principals and agents allocate responsibility. Citizens, elec
3、ted officials, and public managers have more performance information now than ever. Every year, new rivers flow into the existing sea of data. These trends are unlikely to be reversed. Performance management both preceded and outlived the New Public Management and continues to be viewed as a central
4、 plank in the future of governance (Kettle and Kelman 2007). A MODEL OF PERFORMANCE INFORMATION USE Previous Research There are a number of empirical pieces on performance information use in the public sector, and the majority of such research is recent. Heinrich (1999) noted that most empirical evi
5、dence came from the private sector. Her study of performance standards in job training found that data were used but much depended upon the design of the overall performance system and that there was little to guide designers beyond neoclassical economic arguments for financial incentives. Since the
6、n, more research has emerged to offer alternatives to neoclassical models, usually relying on self-reported survey data. It appears fair to assert that this previous work has not resulted in a common or overarching theory of performance information 850 Journal of Public Administration Research and T
7、heory use. De Lancer Julnes and Holzer (2001) theorized a basic distinction between rational/technocratic and political/cultural factors. Other research has focused more on explicitly political variables. Boudreaux and Chicot (2008) examined the role of the governor, the legislature, and the state p
8、olitical context. Makers and Willoughby (2005) categorized variables in terms of community characteristics, respondent characteristics, organizational culture, and performance measurement characteristics. Moynihan and Ingra ham (2004) and Dull (2009) both placed the role of leadership as central. Th
9、e study by A Skim, Johnson, and Christophersen (2008) of municipal benchmarking networks in Norway examined network characteristics, administrative factors, political variables, task characteristics, and history dependence. Moynihan and Landuyt (2009) characterized performance information use as an
10、element of organizational learning and identified structural and cultural variables that predict such use. Conceptualizing Performance Information Use as Organizational Behavior We conceptualize performance information use as a form of organizational behavior. Like other forms of organizational beha
11、vior, employees have discretion about whether and the degree to which they engage in it but are influenced by the social context and formal systems in which they work. We test categories of variables consistent with this conceptualization, incorporating individual beliefs, job attributes, organizati
12、onal factors, and environmental influences. Our model draws up on some of the variables used in previous research, although we seek to use alternative measures of the underlying concept. For example, rather than measure leadership support for performance management, we measure whether the leader is
13、in a generalist of specialist position. Rather than measure public participation in performance management routines, we measure if more general forms of participation matter. The model also includes variables whose relationship with the dependent variable has been previously untested, including publ
14、ic service motivation (PSM), reward expectations, role clarity, task-specific experience, the role of budget officials, and the influence of professional organizations. Individual Beliefs There has, thus far, been surprisingly little investigation into how PSM might affect organizational decision ma
15、king. There are two reasons to assume that PSM might foster performance information use. First, performance information use involves costs for the employees. Performance information use is a behavior that imposes costs on the employee. It displaces traditional modes of decision making and heuristics
16、, while adding another decision criterion, making decision processes more rather than less complex. Although using performance data might generate organizational benefits, individual benefits are unlikely or uncertain. It thus resembles a form of extra-role behavior where employees make gifts of the
17、ir time and effort to the organization without the expectation of individual reward. Such behavior is likely to be exhibited by employees driven by prosaically or altruistic motives. Second, it is plausible that high PSM employees identify with and care about the achievement of organizational goals.
18、 It has been proposed that PSM fosters higher individual and organizational performance, although without definitive evidence (Brewer 2008). If high PSM employees see performance information use as a means of achieving organizational goals, they may also see it as a means to fulfill their desire to
19、serve. Job Attributes Job attributes represent the interplay between the individual and organizational level (Wright 2001). In contrast to the intrinsic approach represented by PSM, other approaches assume self-interest. From this perspective, extrinsic motivators and incentives may foster performan
20、ce information use. Organizations do not reward performance information use but do reward individualized performance to varying degrees. If individuals perceive that their organization links pay and promotion with goal achievement, employees have an extrinsic incentive to use performance information
21、 (Jennings and Heist 2006). Another factor that affects the types of decisions individuals make, and by extension their tendency to use performance data, is their organizational role. Previous research tends to treat role differences in terms of membership of different organizations and of seniority
22、 or hierarchical level. Both qualitative (Radon 2006) and quantitative (A skim, Johnson, and Christophersen 2008; Dull 2009) research report that variation in the use of performance data depends upon the type of program an employee worked in.2 Other research tests the influence of seniority, finding
23、 no relationship (de Lancer Julnes and Holzer 2001; Dull 2009) or a positive relationship (Moynihan and Landuyt 2009). A final role factor is the amount of experience that an individual has with a specific position. Organizational learning theory points to the importance of task knowledge to learnin
24、g (Moynihan 2005). Managers rarely learn directly from quantitative numbers, but from interpreting these numbers, making sense of what they mean given their knowledge of the context in which they work. Individuals with a deep knowledge of task are therefore advantaged in the ability to apply perform
25、ance data. Lewis (2008) finds that bureau- specific experience was a significant predictor of why some managers received higher program evaluation scores than others. One measure of task-specific knowledge is time in a particular position. Organizational Factors All levels of government have devoted
26、 considerable effort into creating and disseminating performance information. This effort reflects a supply-side approach to performance information, assuming that provision of performance data is the key to its use (Jennings and Heist 2006). There is some research support for this view. Various mea
27、sures that track the availability of performance information have been found to be positively associated with use (Boudreaux and Chicot 2008; de Lancer Jules and Holzer 2001; Moynihan and Ingra ham 2004; Moynihan and Landuyt 2009). By contrast, Makers and Willoughby (2005) find that the availability
28、 of measures in budget documents actually reduces the influence of these measures in the budget process and suggest that this may be the result of information overload. Performance management systems, reflecting a doctrinal connection with the New Public Management, were often presented as part of a
29、 package of reforms that also included the provision of greater autonomy to managers, in particular greater discretion with financial and human resources (Moynihan 2008). For example, Schick (2001) argues that performance measurement should be presaged by wider organizational changes, including grea
30、ter managerial flexibility, if it is to succeed. Studies have found that flexibility and its flipside, centralization are antecedents to organizational learning (Moynihan and Landuyt 2009; Schulz 2001) and performance information use (Willis, Mastrofski, and Weisburd 2007). EXTERNAL FACTORS To exami
31、ne the role of external factors, we test how the influence of citizens and external professional organizations affects performance information use. It is possible that the use of performance information is a form of bureaucratic behavior that is disconnected from the public. Managers may believe tha
32、t the public may evince strong support for performance but actually care little about the details. On the other hand, results-based reforms draw from popular sentiment that governments were not as effective as they could be. There is significant evidence that perceived citizen support for or involve
33、ment in performance management processes facilitates use. Research suggests that perceived citizen demand for performance-based accountability encourages performance information use (Moynihan and Ingra ham 2004; Poster and Streib 1999).De Lancer Jules and Holzer (2001) find that support among extern
34、al interest groups (in the form of elected official/citizens) for performance management also fostered use. Yang and Hsieh (2006) find that stakeholder participation is a positive predictor of the perceived effectiveness of performance measures, whereas Ho (2006) finds that citizen involvement in pe
35、rformance measurement practices increases the perceived usefulness of data in the eyes of elected officials. RESULTS AND DISCUSSION In terms of our individual-level variable, we find that PSM is positively correlated with reported performance information use. A one-unit increases on the 30-point PSM
36、 scale results in a .04 increase in the 6-point performance information use scale. The finding also contributes additional circumstantial evidence to the question of whether PSM fosters higher performance. Brewer (2008, 146) notes there is little empirical evidence on the PSMperformance relationship
37、. The results here find that PSM is positively associated with a form of behavior that is a logical contributor to both higher individual and organizational performance. The findings on external factors are not definitive. Both professional influence and participation are positively related to perfo
38、rmance information use but are significant only at the 10% level. The findings therefore do not imply a clear rejection of these variables but instead suggest the need for additional testing and imply the need for additional specification of environmental variables that might affect performance info
39、rmation use. Although none of the controls proved significant, there are additional environmental factorssuch as social capitalthat have plausible influences on performance information use (Tavits 2006) but where suitable data at the municipal level is lacking. CONCLUSION This article argues for exa
40、mining performance information use as a key variable if we are to develop systematic knowledge about contemporary governance. The usual caveats and flaws of cross-sectional survey data apply. In particular, the potential that common source bias inflates the relationship between independent variables
41、 and the dependent variable cannot be dismissed. But in testing such relationships, capturing a large number of comparable individual responses on items that are difficult to externally observesuch as PSM or cultureis a key requirement, necessitating survey-based approaches. In addition, the use of
42、previously tested measures of the majority of variables, as well as the relatively low correlation between these measures, suggests discriminate validity among the items tested and reduces concerns about common source bias. 译文 绩效管理的大问题：为什么经理使用绩效信息？ 资料来源： 威斯康星大学麦迪逊分校 作者： 桑杰潘迪 最近几十年来，政府在广泛的改革 中 已经要求各机
43、构跟踪和衡量 其 战略目标 、对象 和成就 （ 伯尼 、 赫伯特 、 赖特 ， 1999； 莫伊尼汉 ， 2008）。随着 代理人的增长， 以合同形式巩固 绩效目标问责制产生， 用这种 方式连接主要的代理责任 分配网 。公民 、 民选官员和公共管理者比以往任何时候使 用更多的绩效信息。就如每一年 新的江河入海流量 一样 ，这些趋势不太可能逆转。绩效管理 不同 于 之前寿终正寝的 管理，是 新 的 公共管理 ， 被 继续 看作未来 管理 的 中心政策（ 凯特 、 凯尔曼 ， 2007）。 绩效信息使用模型 以往的研究 有一组 表现公共部门 使用 绩效信息的 实证研究数据，多数是最近的 研究。海因
44、里希 （ 1999）指出， 大 多数实证研究证据来自私营部门。她在 工作 绩效标准 培训 的研究发现，使用的数据 很大程度上取决于该系统的设计和整体性能 ，很少有引导设计师超越新古典主义的经济论点的财务激励。 从那时起， 更多 可以提供替代新古典模型 的研究应运而生 ，这 通常依赖自我报告的调查数据。 公平而言，以前的 研究 并没有 得出 关于使用绩效信息的总体理论。 德蓝 瑟 朱尼斯 和 霍尔泽 （ 2001）理论化地划分技术专家 、 政治 及 文化因素的基本区别。 其他的研究更 专注 于明确的政治变量。 包德留克斯和 奇科特（ 2008）研究了 绩效信息在 政府职能、立法机关和国家政治 中
45、的作用 。 马克 和威洛比 （ 2005） 对群体的特性、 答辩人的特点 、 组织文化 、 绩效评估的特点和条件进行归类。 莫伊尼汉 、 英格拉哈姆 （ 2004）和 杜尔 （ 2009）都立足于中央领导作用 研究绩效管理 。 阿脱脂，约翰逊 和 克里斯托弗 （ 2008） 研究 挪威 市政网络基准的特点、行政因素、政治因素、任务特点和历史关系。 莫伊尼汉，朗迪 （ 2009） 认为 绩效信息的使用特点表现为组 织学习 元素 ，并确定了结构和文化的变量 来 预测这种信息的使用。 组织行为学绩效信息使用的概念化 表现 我们将绩效信息的使用作为一个概念化表现形式 的 组织行为。 其他形式的组织行为
46、，员工除了受社会环境 和 工作制度的影响，可自行决定是否且 在何种程度 上使用绩效信息 。我们 将 与此概念 相 一致的变量 进行测试 ，纳入个人信仰、工作属性、组织因素，受环境的影响。 我们的模式借鉴了以往的研究中所使用的一些变量 ， 虽然我们设法利用 基本概念替代 。 例如， 比起 衡量领导支持 的 绩效管理 ， 我们衡量 的是 领导者是否能全面实施绩效管理 。比起衡量公众参与绩效管理程 序，我们更关注的是公众参与度。该模型还包括之前未经测试的因变量，包括公共服务的动力系统（ PSM）、奖励的期望、角色清晰 度 、任务具体经验、预算官员的作用，以及专业组织的影响力。 个人信仰 迄今为止，关
47、于 PSM 影响组织决策的小调差的结果令人惊讶 。 PSM 的表现可能会促进 绩效 信息的使用 有两个原因 。首先，绩效信息的使用涉及员工费用。绩效信息的使用是一种 加强员工 费用 有效使用的行为 。它的决策和传统模式 相比 ，同时加入另一个决策准则，决策过程 会 更加复杂，而不是更少。虽然使用绩效 数据可能 出现 组织利益 和 个人利益 不可能或不确定能达 成一致的现象 。它类似于一个额外角色的行为方式 奖励员工为企业付出的时间和努力 ，无需单独奖励。 这 很可能是由社会或利他主义的动机驱使员工 付出 。 第二，高 PSM 和关怀员工 、关心 组织目标的实现 是合理的 。 有人提出， PSM
48、 可以 培养较高的个人和组织绩效，虽然没有确切的证据 （布鲁尔 ， 2008）。如果高 PSM 的员工看到绩效信息的使用作为一种实现组织目标的手段 ，也可看到它作为一种手段来实现 自己 的愿望。 作业属性 作业属性代表个人和组织 层次之间 的相互作用 （赖特， 2001） 。与此相反的内在办法 以 PSM 为代表 。 从这个角度来看，外在 的 激励 和奖励措施可以促进绩效信息的使用。组织不奖励绩效信息的使用， 除了奖励不同程度的个性化绩效信息 。如果个人觉得他们的薪酬和 晋升与 组织目标 的实现和 成果 的 推广 有联系 ，员工 就 会 有一个 外在的动力去 使用绩效信息（詹宁斯 、 海斯特
49、， 2006）。 另一个因素是影响个人 做出 决定的类型 。 组织促使 他们 的员工 倾向于 使用绩效数据 。 以往的研究往往区别对待不同组织和资历水平方面有差异的成员，既有定性（雷丁 ， 2006）和定量（ 阿脱脂 、 约翰逊 、 克里斯托弗 ， 2008； 杜尔 ，2009)的研究报告， 使用绩效数据变化后的雇员工作标准 和 其他研究试验的影响，和工龄没有关系（ 德蓝瑟朱 、 霍尔泽 ， 2001； 杜尔 ， 2009）或 成 正关系（ 莫伊尼汉 、 朗迪 ， 2009）。 最后一个 主要 因素是 特定职位上 的经验 的积累 。组织学习理论指出了 学习知识 和学习任务 的重要性 (莫伊尼汉 ， 2005)。 管理者很少直接从量变分析数字，但是 会利用工作经验分析这些数字并 做出 决策。渊博的个人知识及经验使其具备得天独厚的应用 绩效数据的 能力 。 刘易斯 （ 2008）发现，具体经验 预示了 为什么有些经理人比其他人获得较高的方案评价分数 ，具体表现是知识 在某个时间用在 特定 的 位置。 组织因素 各级政府 都 投入 了大量的精力在 创造和传播 绩效 信息 上 。假设提供的绩效数据是其使用的关键 ，这方面的努力反映了供应绩效信息的方 法 （ 詹宁斯 、 海斯特 ， 2006
