圖書館是一國文化建設的重要指標，在現今知識經濟時代中，資訊服務競爭日趨激烈且多元，面對少子化，大學圖書館應高度回應學術環境的變遷，運用營運績效來表現其在機構任務中的價值，並發展相關專業知識，以陳述圖書館對於學生學習與成功的影響。在執行績效評鑑時，國家圖書館之〈全國圖書館調查統計系統〉係為各館一致遵循，且定期提供的統計數據系統，應可利各圖書館進行營運績效評鑑之應用，同時亦能提供圖書資訊學專家或社會學家進行學術研究之參用。在績效評鑑模式上，關鍵績效指標（KPI）是用以衡量組織績效最為重要的資訊，也是績效指標範圍當中最核心的部分，可以說是引導圖書館發展方向的儀表板。 本研究目的有四：（1）探討大學圖書館績效評鑑與關鍵績效指標。（2）探討大學圖書館統計資料與績效評鑑指標的關係，以建構適合大學圖書館績效評鑑之模式。（3）探討〈全國圖書館調查統計〉支持本研究建構的大學圖書館統計支持績效評鑑模式的可行性，與績效評鑑實施的有效性。（4）探討圖書館統計是否滿足大學圖書館績效評鑑實施的需求。 為獲研究結果，本研究透過焦點團體訪談法建構成五個評鑑構面、20項重要績效指標，作為國內大學圖書館之關鍵績效指標評鑑模式之評鑑工具。為確保績效指標的可行性，經過研究者取樣國內13所大學圖書館統計調查資料，實際進行個案研究結果，獲得以下結論：（1）大學圖書館實施績效評鑑仍為少數，但各館普遍配合〈全國圖書館調查統計系統〉進行統計並如期填報。（2）本研究提出大學圖書館績效評鑑之五大構面20項重要績效指標具可行性，可做自我評鑑及檢視績效成果之用。（3）運用最佳權重和關鍵績效指標，建構出關鍵績效指標評鑑模式。（4）〈全國圖書館調查統計〉統計項目大致可滿足於大學圖書館20項重要績效評鑑指標項目。 針對所獲得的研究結果，提出三點建議：（1）國家圖書館之〈全國圖書館調查統計系統〉，與績效評鑑指標結合，透過各圖書館上傳各項統計至系統後，系統即可自動進行大數據分析，讓統計資料即時轉換為績效成果，預期將節省實施績效評鑑之時間和人力成本，提高大學圖書館全面實施績效評鑑之意願。（2）圖書館可以運用關鍵績效指標評鑑模式，透過全國大學圖書館的績效評鑑結果，進一步針對各館進行SWOT分析，訂定未來營運發展的具體可行之量化與質化目標，展現圖書館的影響價值。（3）推動大學圖書館實施績效評鑑，建立長期和完整的評鑑資料庫。 / A library is an important indicator of a country's cultural construction. In the present time of knowledge-based economy, information services are becoming increasingly competitive and diverse. In the face of a low-birth rate, university libraries have to demonstrate their value through operating performance and develop related professional knowledge in response to changes in the academic environment in order to state their influence on students’ learning and academic success. When carrying out the operating performance evaluation, university libraries may refer to the data published regularly by the Library Statistical System of National Central Library; experts or sociologists may also refer to these data in their academic study. To measure the performance of an organization, Key Performance Indicator (KPI) is the most important information and the core of all performance indicators; it can be said to be the dashboard that guides the development of a library. The research has four purposes: (1) exploring the performance evaluation and the key performance indicator of university libraries; (2) exploring the relationship between the library statistics and the performance indicator in order to build an appropriate performance evaluation model for university libraries; (3) exploring the feasibility of supporting the performance evaluation model by the university library statistics established with the support of Library Statistical System and exploring the effectiveness of implementation of performance evaluation; and (4) exploring whether the library statistics satisfy the requirements for the implementation of performance evaluation at university libraries. As to the performance evaluation model, the research has established five aspects of performance evaluation of university libraries and 20 key performance indicators through focus group. To ensure the feasibility of these performance indicators, the researcher sampled 13 university libraries’ statistics and carried out the case study. The conclusions are as follows: (1) only few university libraries carried out performance evaluation, but each university library gathered statistics and filed with the Library Statistical System as required; (2) five aspects of performance evaluation of university libraries and 20 key performance indicators proposed by the research were feasible and applicable to self-evaluation and self-review of performance results; (3) the performance evaluation model was established based on the optimum weight and key performance indicators; and (4) Statistical items of the Library Statistical System generally meet 20 key performance indicators of university libraries. Three recommendations are proposed based on the result of the research: (1) With the application of the Library Statistical System of National Central Library and performance evaluation indicators, university libraries upload the statistics to the system where data are analyzed automatically and converted into the performance results. Such application is expected to save time and labor costs of implementation of performance evaluation and increase the willingness of university libraries to implement performance evaluation; (2) Based on the KPI model and the results of performance evaluation of nationwide university libraries, university libraries may be able to carry out the SWOT analysis and set up the feasible quantitative and qualitative goals of future operation and further demonstrate their influence; and (3) university libraries are encouraged to implement performance evaluation and establish an extended and sound evaluation database.