本研究旨在探討主管悖論領導行為對部屬工作效能之影響，並進一步檢驗 不確定性規避之調節作用。悖論領導係指領導者能同時展現看似對立但實際互 補之行為特徵，例如在控制與授權、規範與彈性、效率與創新之間進行動態平 衡，以回應組織內外部複雜且多變之情境需求。在當前充滿多變性、不確定 性、複雜性與模糊性之 VUCA 環境下，悖論領導已逐漸成為值得重視之領導 能力。 本研究採兩階段主管—部屬配對問卷調查法蒐集資料。第一階段向台灣地 區各產業主管發放主管問卷，並同步蒐集部分部屬問卷；第二階段則針對同意 協助之主管及其直屬部屬發放配對問卷。經問卷回收、配對與篩選後，本研究 取得172組有效配對樣本。其中，量表信效度分析與測量模型檢驗以172組樣 本為基礎；另因部分樣本於人口統計變項上存在缺漏，故階層迴歸分析與研究 假設檢定係以115組可用樣本進行。 研究結果顯示：第一，悖論領導對部屬工作效能之影響具有差異化效果， 僅對知識分享具有顯著正向影響(β=.280， p=.003)，對團隊學習與創新行為則 未達顯著水準。第二，不確定性規避未能顯著調節悖論領導與團隊學習、知識 分享及創新行為之關係，三項交互作用效果皆未達顯著。整體而言，本研究發 現悖論領導對不同工作效能構面之影響並不一致，其中對知識分享之促進效果 較為明確，而不確定性規避之調節作用則未獲支持。;This study aimed to examine the effects of supervisors’ paradoxical leadership on subordinates’ work effectiveness and to further test the moderating role of uncertainty avoidance. Paradoxical leadership refers to a leader’s ability to simultaneously exhibit seemingly contradictory yet complementary behaviors， such as balancing control and empowerment， structure and flexibility， and efficiency and innovation， in order to address complex and dynamic organizational demands. In today’s VUCA environment characterized by volatility， uncertainty， complexity， and ambiguity， paradoxical leadership has become an increasingly important leadership capability. This study adopted a two-stage supervisor–subordinate matched-pair questionnaire survey. In the first stage， questionnaires were distributed to supervisors across various industries in Taiwan， along with a portion of subordinate questionnaires. In the second stage， matched questionnaires were distributed to supervisors who agreed to participate and their direct subordinates. After data collection， matching， and screening procedures， 172 valid matched pairs were retained. Among them， the analyses of reliability， validity， and measurement model fit were based on 172 matched pairs. Because some cases contained missing demographic data， hierarchical regression analyses and hypothesis testing were conducted using 115 usable matched pairs. The results showed that， first， paradoxical leadership had differential effects on subordinates’ work effectiveness. Specifically， it had a significant positive effect only on knowledge sharing (β=.280， p=.003)， while its effects on team learning and innovative behavior were not statistically significant. Second， uncertainty avoidance did not significantly moderate the relationships between paradoxical leadership and team learning， knowledge sharing， or innovative behavior， as none of the interaction terms reached statistical significance. Overall， the findings suggest that the effects of iii paradoxical leadership vary across different dimensions of work effectiveness， with its facilitating effect being most evident for knowledge sharing， whereas the hypothesized moderating role of uncertainty avoidance was not supported.
