公務人員受委託行使行政權，享有身分保障和穩定收入，因而被視為社會體系的中上階層。基於此等特性，學者曾經提出代表性官僚理論與多元代表性等理論，希望能夠以此調和民主與效率兩大價值，達成政策執行的實質公平。然過去相關的實證研究，多半屬於「事後檢測」。但是我國公民除了通過公務人員考試之外，幾乎沒有其他擠身常任文官的管道，因此有關代表性或多元性的討論應該包括考試階段。換言之，考試錄取人員的人口特性值得深入的分析。本研究以2003至2007年的高等考試參與者為主要研究對象，輔以司法官三等考試的資料以利比較。接著，引用社會學「地位取得」的分析架構，採取指數比較分析與邏輯迴歸分析，探索性別、年齡、受教育時間、畢業學校特性、出身地域與錄取與否的關係。 　　本研究結果顯示，女性、非傳統公立綜合大學的畢業生、鄉鎮地區出身者以及41歲以上的考生在公務人員考試中處於較為不利的地位，但是與優勢團體之間的差距隨著考試等級和種類有所不同。例如女性在行政類與司法官考試中，與男性的表現平分秋色甚至猶有過之；剛完成高等教育的25歲以下人口在司法官考試裡最具優勢，但高考三級則有利於26至35歲的青年；傳統公立大學的文憑與都市出身的背景，在技術類考試能發揮的正效果比行政類考試為弱。基於上述的研究發現，為消除各種團體在公務人員考試裡的地位差距，本研究對於未來的考選政策提出下列建議：1.加強命題與口試委員的多元化；2.強化試題的研究發展；3.配合政府再造鬆綁人事法規；4.營造一個落實多元平等的大環境。 / Civil servants, with administrative power in hand, are commonly regarded as part of upper class. Therefore, researchers, to integrate democracy and efficiency in civil service system and to realize the genuine equality, propose representative bureaucracy and team diversity theory. Senior Civil Service Examination, the main approach, if not only, for citizens in Taiwan to enter the bureaucracy affects the representativeness and diversity of state apparatus substantially. By adopting the research approach of “status attainment” from sociology, this study gives an account of the relation between the examination result and the demographics in terms of gender, education, school characteristic, and region. All the data of this research is based on participators’ personnel information cards collected by the Ministry of Examination from 2003 to 2007. The result indicates that four types of participators are inferior in the examination, including women, graduates from private untraditional technological colleges, participators from rural areas, and adults above forty-one years old. The differences between superior and inferior groups, however, vary with the level and subject of examination. To prevent demographic differences in Senior Civil Service Examination, the study suggests the following factors be considered: the diversity of the composition of examiners committee should be ensured, the enhancement of development research of test questions, the deregulation of public personnel rules and, last but not least, the construction of a diversity-respecting society.

過去研究雖然發現團隊情感氛圍 (group affective tone)能有效預測團隊成員的助人行為、團隊創造力、團隊缺席率與離職等重要變數，但在概念上或是理論上仍有許多未釐清之處。因此，本研究擬針對下列的文獻缺口加以探討：(1) 探討影響正/負向團隊情感氛圍形成的前因變數，釐清有哪些變數會影響團隊情感氛圍的形成；(2) 探討不同時間點正/負向團隊情感氛圍的動態改變及情感事件的干擾效果；(3) 探討正/負向團隊情感氛圍與團隊績效的因果關係；由於單一研究不易同時兼顧上述三個研究目的，因此本研究將以三個研究分別針對上述目的加以探討。 在研究一，本研究以101個學生團隊做為樣本 (523位成員)，路徑分析的結果發現，團隊成員的年齡、教育背景多元化，會透過負向影響群體認同、進而強化負向團隊情感氛圍。此外，知覺團隊間競爭、成員的迎合、自我表現印象管理策略，則會透過正向影響群體認同、進而強化正向團隊情感氛圍。在研究二，本研究則以44個大學部、碩士班學生組成的團隊做為樣本 (和研究一的樣本獨立，成員為215人)，階層線性模式分析的結果指出，正、負向團隊情感氛圍在跨時間點確實會產生動態改變，且前期的正/負向團隊情感氛圍均能顯著影響後期的正/負向團隊情感氛圍。且前後期間發生的負向情感事件，亦會減弱前、後期正向團隊情感氛圍之間的正向關係。最後，在研究三，本研究以81個高科技產業研發團隊做為樣本 (包括259位團隊成員及81位主管)，交叉延遲貫時性分析 (cross-lagged panel analysis, CLPA)的結果指出，前期的正向團隊情感氛圍對後期團隊績效有顯著正向影響，而前期團隊績效對於後期正向團隊情感氛圍，也有顯著的正向影響。整體而言，本研究之假設大多符合理論預期，而研究結果對團隊情感氛圍、團隊多元化、以及印象管理研究的理論意涵與實務貢獻，亦在文中一併加以討論。 / Extending previous research on group affective tone (GAT), three studies were conducted to examine the antecedents, dynamics, and consequences of GAT. In study 1, 101 student teams (523 members) were chosen as the sample. The results of path-analysis showed that team age and educational background diversity decrease group identification, which in turn increases negative group affective tone (NGAT). In addition, perceived intergroup competition, members’ self-promotion and ingratiation impression management tactics increase group identification, which in turn increases positive group affective tone (PGAT). In study 2, 44 student teams (215 members) were selected as the sample, and the experience-sampling method was used to capture the dynamics of GAT (i.e., 6-time repeated measure). The results of hierarchical linear modeling analysis revealed that the T-1 PGAT positively predict T0 PGAT over time; while T-1 NGAT positively predict T0 NGAT over time. Moreover, the negative affective event occurred during T-1 and T0 attenuates the positive relationship between T-1 PGAT and T0 PGAT. In study 3, 81 research and develop teams (259 members and 81 supervisors) were chosen as the sample, and the cross-lagged panel analysis (CLPA) was used to examine the causal association between GAT and team performance. The results of CLPA showed that T1 PGAT has a positive effect on T2 team performance. Additionally, T1 team performance has a positive effect on T2 PGAT. These result suggest that the presence of a reciprocal relationship between PGAT and team performance. All three studies support the proposed theoretical framework.