领导者授权与高层决策：高层管理者的授权领导如何以及何时提高公共组织的决策速度和全面性？

公共组织决策的效率是公共行政学者感兴趣的一个关键领域。本研究基于高层梯队理论，探讨了高层管理人员的授权领导如何影响公共组织的决策速度和全面性。通过对中国 133 个公共组织的 1149 名 TMT 成员进行的三轮问卷调查，我们检验了所提出的理论模型和假设。我们的研究结果表明，高层管理者授权领导力可以通过对 TMT 行为整合的影响，对公共组织的决策速度和全面性产生积极和间接的影响。此外，TMT工作风格相似性和TMT认知多样性可以分别增强和削弱中介关系。

The efficiency of decision-making in public organizations is a key area of interest for public administration scholars. Drawing on upper echelons theory, this study examines how top managers' empowering leadership affects decisionmaking speed and comprehensiveness in public organizations. Through a three-wave questionnaire survey involving 1149 TMT members from 133 public organizations in China, we tested the proposed theoretical model and hypotheses. Our findings reveal that top manager empowering leadership positively and indirectly influences decision-making speed and comprehensiveness in public organizations, and through its effects on TMT behavioral integration. Furthermore, TMT work style similarity and TMT cognitive diversity can enhance and weaken the mediating relationships, respectively.

引言

    几十年来，公共行政决策一直是学术界讨论的焦点，被认为是公共组织战略管理的一个重要方面，对公共组织的绩效和目标实现有重大影响。鉴于公共组织决策的复杂性，目标模糊，利益相关者多样，确定实现有效战略决策的途径具有理论和实践意义。然而，公共行政学者很少关注高层管理者的领导风格如何影响公共组织的战略和绩效。尽管高层管理理论表明领导风格通常是组织结果的更可靠预测因素，但这仍然是一个研究不足的领域。

    总之，我们构建了一个有调节的中介模型来研究高层管理者授权领导力通过 TMT行为整合影响公共组织决策速度和全面性的过程；此外，我们还探讨了 TMT 工作风格相似性和 TMT 认知多样性在这些中介关系中的边界条件。

 For decades, decision-making in public administration (PA) has been a focal point of scholarly discussion, and is recognized as a crucial aspect of strategic management in public organizations, significantly impacting their performance and goal achievement. Given the complexity of decision-making in public organizations, marked by ambiguous goals and diverse stakeholders, identifying pathways to achieve effective strategic decision-making is of both theoretical and practical importance. Yet, PA scholars have paid little attention to how top managers' leadership styles affect public organizations' strategies and performance. This remains an understudied area despite upper echelons theory indicating that leadership styles are usually more reliable predictors of organizational outcomes.

  In summary, we constructed a moderated mediation model to examine the processes underlying top manager empowering leadership to affect decision-making speed and comprehensiveness of public organizations through TMT behavioral integration; furthermore, we also explored the boundary conditions of TMT work style similarity and TMT cognitive diversity with regard to these mediated relationships.

研究重点

    具体来说，我们关注高层管理人员和TMT之间动态的两个主要限制。首先，对公共组织高层的研究主要集中在个人高层管理人员如何直接影响组织成果，往往忽视TMT发挥的关键作用。TMT是指一个组织的高层管理人员，他们带来了各种知识、观点和价值观。第二个主要限制在于缺乏对TMT动态的微观影响的了解，特别是高级经理和TMT成员之间对TMT流程和组织成果的互动。

    总之，高层管理理论为我们提供了一个将高层管理者授权领导力与 TMT 行为整合、决策速度和全面性联系起来的理论框架。此外，“高层管理者-TMT 界面”视角可以更有效地解释影响组织战略决策过程的微观机制。

  Specifically, we pay attention to two major limitations regarding the dynamics between top managers and TMTs. First, research on the upper echelons of public organizations has predominantly focused on how individual top managers influence organizational outcomes directly, often neglecting the critical role played by the TMT. TMT refers to an organization's top-level senior executives, who bring a variety of knowledge, perspectives, and values. The second major limitation concerns the lack of understanding of micro-level impacts of TMT dynamics, especially the interactions between the top manager and TMT members, on TMT processes and organizational outcomes. 

    In conclusion, upper echelons theory provides us with a theoretical framework linking top manager empowering leadership with TMT behavioral integration, decision-making speed, and comprehensiveness. Additionally, the “top manager–TMT interface” perspective allows for a more effective explanation of the micro-level mechanisms influencing organizational strategic decision-making processes. 

研究假设

H1：高层管理者授权领导与TMT行为整合呈正相关。

H2a：TMT行为整合与决策速度呈正相关。

H2b：TMT行为整合与决策全面性呈正相关。

H3：高层管理者授权领导通过TMT行为整合的中介作用间接正向影响（a）决策速度和（b）决策全面性。

H4：TMT工作风格相似性调节了高层管理者授权领导对TMT行为整合的正向影响。具体而言，当TMT工作风格相似性较高时，高层管理者授权领导对TMT行为整合的正向影响增强。

H5：TMT工作风格相似性通过TMT行为整合调节了高层管理者授权领导对（a）决策速度和（b）决策全面性的间接正向影响。具体而言，当TMT工作风格相似性较高时，这些间接正向影响增强。

H6：TMT 认知多样性调节了高层管理者授权领导对 TMT 行为整合的正向影响。具体而言，当 TMT 认知多样性较高时，高层管理者授权领导对 TMT 行为整合的正向影响减弱。

H7：TMT 认知多样性通过 TMT 行为整合调节了高层管理者授权领导对 (a) 决策速度和 (b) 决策全面性的间接正向影响。具体而言，当 TMT 认知多样性较高时，这些间接正向影响减弱。

H1. Top manager empowering leadership is positively related to TMT behavioral integration.

H2a. TMT behavioral integration is positively related to decision-making speed.

H2b. TMT behavioral integration is positively related to decision-making comprehensiveness.

H3. Top manager empowering leadership indirectly and positively affects (a) decisionmaking speed and (b) decision-making comprehensiveness through the mediation of TMT behavioral integration. 

H4. TMT work style similarity moderates the positive effect of top manager empowering leadership on TMT behavioral integration. Specifically, when TMT work style similarity is high, the positive effect of top manager empowering leadership on TMT behavioral integration is strengthened.

H5. TMT work style similarity moderates the indirect positive effects of top manager empowering leadership on (a) decision-making speed and (b) decision-making comprehensiveness through TMT behavioral integration. Specifically, when TMT work style similarity is high, these indirect positive effects are strengthened.

H6. TMT cognitive diversity moderates the positive effect of top manager empowering leadership on TMT behavioral integration. Specifically, when TMT cognitive diversity is high, the positive effect of top manager empowering leadership on TMT behavioral integration is weakened.

H7. TMT cognitive diversity moderates the indirect positive effects of top manager empowering leadership on (a) decision-making speed and (b) decision-making comprehensiveness through TMT behavioral integration. Specifically, when TMT cognitive diversity is high, these indirect positive effects are weakened.

研究方法

    本研究以中国县级（区级）教育局为分析单位，数据收集时间为2018年至2019年，与中国北方多个省（包括直辖市）的地方人民政府和380个县级（区级）教育局合作进行了时滞三轮调查数据收集。在发放问卷之前，各教育局的高层领导确认了参与的TMT成员，同时，研究团队将名单与公开的人员信息进行交叉对照，最终共有2863名TMT成员参与了研究，其中包括380名高层管理者。随后，研究团队将纸质问卷分发给负责参与调查的县（区）人民政府内的指定联系人，联系人再将问卷分发给教育局的高层管理者（即主要局长）。

    在问卷收集方面，第一轮我们获得可用组织答案337份，第二轮263份，第三轮176份。我们对所有问卷进行单独匹配和筛选，以选取合适的适合研究的样本。在样本选取上，我们优先考虑在三轮调查中高管回复完整的组织，排除高管发生变更的组织；我们还关注在整个研究过程中始终参与的TMT成员，排除期间加入或离开TMT的成员的数据。本文所用量表均为已建立并验证过的量表。英文原文由PA领域的专家采用回译法译成中文，以确保内容和含义的准确性。借鉴前人对TMT的研究，我们控制了五个变量：TMT规模（TMT成员数量）和TMT特征差异，包括性别、年龄、教育程度和工作经验。具体来说，我们计算每个TMT内每个变量的标准差，来表示该变量在TMT内变异的水平。

    本研究纳入的所有量表的Cronbach's alpha系数均超过0.7，表明其具有较高的信度。进一步计算了各量表的结构信度（CR），结果表明所有测量量表的CR值均超过了推荐阈值0.70。然后，采用平均方差提取法（AVE）评估量表的收敛效度。结果表明所有量表的AVE值均达到了推荐阈值0.50，表明所有量表均具有良好的收敛效度。

    由于我们的理论模型是团队级模型，其中每个变量都需要在TMT级别进行分析，并由多名TMT成员进行评估，因此数据汇总对于假设分析是必不可少的。在汇总数据之前，重要的是考虑数据是否可以汇总。为了实现这一目标，我们计算了理论模型中每个变量的 Rwg（即组内信度）、ICC (1) 和 ICC (2)（即组内相关系数）值的平均值。

    虽然我们采取了诸如使用时间滞后调查方法、避免单项和社会期望指标、确保受访者匿名等程序性技术来减少共同方法偏差 (CMB) 对研究结果的影响，但仍需要进一步检验。即可以将共同方法方差作为潜在变量，将所有研究变量加载到其上，通过比较有和没有共同方法方差因子两种情况下的模型拟合度来检验 CMB。当在假设的六因子模型中加入共同方法方差因子时，加入共同方法方差因子前后两个模型的拟合指标（CFI、TLI、RMSEA、SRMR）的变化幅度在0.00~0.02之间，且变化不显著，因此本研究中不存在潜在的共同方法偏差。

   This study focuses on county-level (district-level) education bureaus in China as the unit of analysis. Data was gathered during 2018 to 2019. Time-lagged three-wave survey data collection was conducted in collaboration with local People's Governments and 380 county-level (district-level) education bureaus located in several provinces (including municipalities that were directly under the administration of the State Council) in northern China. Prior to distributing questionnaires, the top manager of each education bureau confirmed the participating TMT members. Simultaneously, the research team cross-referenced the list with publicly available personnel information. This process resulted in a total of 2863 TMT members, including 380 top managers, participating in the study. The research team then distributed paper questionnaires to designated contact persons within the People's Government responsible for the participating counties (districts), who in turn distributed them to top managers (i.e., chief directors) of education bureaus.

    With regard to the questionnaire collection, we obtained 337 usable organizational responses in the first round, 263 usable organizational responses in the second round, and 176 usable organizational responses in the third round. We individually matched and screened the questionnaires to select appropriate suitable samples for the study. In selecting our sample, we prioritized organizations with complete responses from top managers across all three waves of the survey and excluded those with changes in top managers. We also focused on TMT members consistently involved throughout the study, excluding data from those who joined or left TMTs during the period. All scales used in this article were established and validated scales. The original English scales were translated into Chinese using the back-translation method by experts in the field of PA to ensure of content and meaning accuracy. Following previous research on TMTs, we controlled for five variables: TMT size (number of TMT members) and TMT characteristics differences, including gender, age, education level and working experience. Specifically, we calculated the standard deviation of each variable within each TMT to represent the level of variation of this variable within the TMT.

    The Cronbach's alpha coefficients of all scales included in this study exceeded 0.7, thus indicating a high-level reliability. Furthermore, we calculated the construct reliability (CR) of each scale, and the results showed that the CR values of all the measurement scales exceeded the recommended threshold of 0.70. Then, we used average variance extraction (AVE) to evaluate the convergent validity of the scales. The results showed that the AVE values of all scales reached the recommended threshold of 0.50, thus indicating that all scales exhibited good convergent validity.

   Since our theoretical model is a team-level model, in which each variable needs to be analyzed at the TMT level and is evaluated by several TMT members, data aggregation is necessary for hypotheses analysis. Before aggregating the data, it is important to consider whether the data can be aggregated. To accomplish this goal, we calculated the mean value of Rwg (i.e., within-group interrater reliability), ICC (1) and ICC (2) (i.e., intraclass correlation coefficients) values for each variable included in the theoretical model.

     Although we have taken procedural techniques such as using a time-lagged survey method, avoiding single-item and socially desirable measures, and ensuring respondent anonymity to reduce the impact of common method bias (CMB) on the research results, further testing remains necessary. That is, the common methods variance can be used as a latent variable and all study variables can be loaded on it, and the CMB can be tested by comparing the degree of model fit in the two situations with and without the common methods variance factor. When a common method variance factor was added to the hypothesized six-factor model, the changes in the fitting indicators (i.e., CFI, TLI, RMSEA, and SRMR) of the two models before and after the inclusion of the common method variance factor ranged from 0.00 to 0.02, and the changes were not significant. Therefore, there is no potential common method bias in our study.

研究结果

假设一至假设七均得到证实。

Hypotheses 1 to 7 are all confirmed.

研究结论

     在公共组织中，战略决策过程更加苛刻且官僚化，这就要求决策者考虑如何协调以提高决策过程的有效性。借鉴高层管理者理论，我们基于有调节的中介模型，对中国公共组织高管团队的问卷调查数据进行了分析。研究结果表明，高管授权领导力通过与高管团队行为整合相关的关键高管团队过程对公共组织决策速度和全面性产生正向影响。此外，高管团队工作风格相似性可以强化这些中介关系，而高管团队认知多样性可以削弱这些中介关系。以下讨论了本研究与现有的公共部门高层管理研究和领导力文献相关的理论意义。

    首先，我们丰富了有关高管领导风格和公共组织决策过程的研究。探讨高管授权领导力如何影响公共组织决策速度和全面性可以更准确地揭示高管团队的微观过程，并进一步丰富授权领导力文献的范围。

    其次，我们确定了TMT行为整合在这些关系中的中介作用，从而丰富了高层管理理论的研究，拓展了公共部门研究中“高层管理者—TMT界面”的视角。从“高层管理者—TMT界面”视角进行的高层管理研究强调了高层管理者与TMT成员互动以触发TMT过程并影响组织结果的事实。

    第三，我们探讨了TMT工作风格相似性和TMT认知多样性作为高层管理者授权领导影响决策速度和全面性过程的边界条件的作用。通过捕捉公共TMT的关键认知和行为特征，我们为公共组织“齐心协力思考和工作”提供了进一步的经验证据。我们的理论假设和实证结果表明，认知多样性会导致冲突和分歧，从而潜在地损害决策的速度和全面性；考虑到团队内部冲突在不同实证情境中对组织决策的影响可能有所不同，这可能为我们假设的模型提供进一步的理解。实践中有效的冲突管理可以促进视角的多元化，从而帮助组织做出更全面的决策。

    最后，我们的研究也存在一定的局限性。首先，虽然我们采用宽泛的方法选取研究样本，保证了样本来源的多样性，但我们忽略了不同类型组织、不同文化背景下的高层管理者和TMT特征的差异。未来的研究应该调查我们的研究结果是否可以推广到不同的组织和国家情境中。其次，除了TMT的行为和认知特征外，TMT的一些微观心理特征也可能影响高层管理者和TMT成员之间的互动。此外，在本文中，我们借鉴了一些关于私营部门的理论和文献，讨论了高层管理者的授权领导对公共组织决策过程的影响。需要注意的是，“管理”这一概念既适用于公共部门也适用于私营部门，未来的研究可以检验理论模型在私营部门的有效性，并解释可能存在的不同结论，以加深我们对高层理论和授权领导文献在公共部门和私营部门适用性的差异的理解。最后，虽然我们使用了几种方法来提高研究设计的有效性，但我们没有得出任何因果推论。未来的研究应采用纵向研究设计或混合方法设计，以建设性地复制我们的研究结果。

   In public organizations, the strategic decision-making process is more demanding and bureaucratic, a situation which requires decision-makers to consider how to coordinate to improve the effectiveness of the decision-making process. Drawing on upper echelons theory, we analyzed data collected through a questionnaire survey of TMTs from Chinese public organizations based on a moderated mediation model. Our findings show that top manager empowering leadership positively affects decision-making speed and comprehensiveness in public organizations through the critical TMT process associated with TMT behavioral integration. Furthermore, TMT work style similarity can strengthen these mediated relationships, while TMT cognitive diversity can weaken these mediated relationships. The theoretical implications of our study in relation to existing upper echelons research and leadership literature in the public sector are discussed below.

   First, we enrich the literature on top manager leadership style and the research on decision-making processes in public organizations. Exploring how top manager empowering leadership influences decision-making speed and comprehensiveness in public organizations can more accurately reveal the microprocesses of TMTs, and further enrich the scope of empowering leadership literature.

  Second, we identified the mediating effect of TMT behavioral integration on these relationships, thereby enriching the research on upper echelons theory and expanding the perspective of the “top manager–TMT interface” in the public sector research. Upper echelons studies conducted from the perspective of the “top manager–TMT interface” have highlighted the fact that top managers interact with TMT members to trigger TMT processes and influence organizational outcomes.

   Third, we explored the roles of TMT work style similarity and TMT cognitive diversity as boundary conditions of the process by which top manager empowering leadership affects decision-making speed and comprehensiveness. By capturing the key cognitive and behavioral characteristics of public TMTs, we provide further empirical evidence that public organizations can use to “think and work with one mind.” Our theoretical hypotheses and empirical findings reveal that cognitive diversity leads to conflict and disagreement, thereby potentially undermining the speed and comprehensiveness of decision-making; given that conflict within a team may have mixed impacts on organizational decision-making across different empirical contexts, this may provide a further understanding of our hypothesized model. Effective conflict management in practice could promote a greater diversity of perspectives, thereby helping organizations make more comprehensive decisions.

  Finally, our study has certain limitations. First, although we selected the research sample using a broad approach to ensure the diversity of sample sources, we ignored the differences in the characteristics of top managers and TMTs across different types of organizations and different cultural backgrounds. Future research should investigate whether our findings are generalizable to different organizational and national contexts. Second, in addition to the behavioral and cognitive characteristics of the TMT, some micro-level psychological characteristics of the TMT may also affect the interactions between the top manager and TMT members. Furthermore, in this article, we draw on some theories and literature concerning the private sector to discuss the impact of the empowering leadership of top managers on decision-making processes in public organizations. It should be noted that the concept of “management” applies to both the public and private sectors, and future research can test the effectiveness of theoretical models in the private sector and explain the potentially different conclusions to improve our understanding of the differences in the applicability of upper echelons theory and the empowering leadership literature between the public and private sectors. Finally, although we used several methods to increase the validity of the research design, we did not draw any causal inferences. Future research should employ longitudinal research designs or mixed methods designs to constructively replicate the findings of our research.

《公共行政》微信公众号刚刚上线不久，我们期待来自各界的宝贵意见，您可将您的建议发送给后台。如果您对我们推送的文章感兴趣，也可在下方留言评论，与其他读者一起交流。

”