台灣公共行政的發展迄今僅有近50 年，學門的研究重點與發展趨勢一直是值得關注的課題。本研究選擇以碩士論文為研究客體，期能從不同視角深入地檢視公共行政學門的發展現況。本文的主要目的包括：（1）藉由分析碩士論文的研究領域與研究方法，瞭解此一學門的發展以及檢視學門的教育現況；（2）在時間面向，藉由分析近12 年來碩士論文的研究領域與研究方法的變化，瞭解此一學門在不同時期的演變趨勢；（3）在空間面向，藉由分析系所所在地區，對研究領域與研究方法的應用是否有差異，以瞭解此一學門的地方發展特色。本研究獲得幾個重要且值得討論的研究發現：（1）論文篇數呈現增長的趨勢，反映出公共行政學門的蓬勃發展，不過，超過半數以上的碩士論文集中在北部地區，東部地區的比例不及一成。（2）在研究領域方面，公共政策、公共管理與政治學為主要研究主題，但有關法制與政府預算方面的研究則仍嫌不足；另一方面，研究主題選擇在地理區域出現較大的差異性。（3）在研究方法方面，有半數左右屬於質性論文，使用量化研究方法的論文比例偏低。（4）在抽樣設計方面，多數論文採取非隨機抽樣方法；統計分析方法則以單變量分析和雙變量分析居多，仍有進一步提升的空間。

Compare to United State, the study of public administration developedrelative late during past 50 years in Taiwan, and it’s worthwhile to track the focusand developmental trend of research conducted in this field. This researchfocus on the graduate theses, look at development of public administration fromdifferent angles. The subject of research includes: (1)Understand the developmentand present state of public administration education through analysis ofresearch interest and method in graduate theses. (2)The analysis of changes inresearch interest and method in graduate theses over past 12 years in order tounderstand development of public administration faculties over that period.(3)The analysis of location of different schools to determine the divergence inresearch interest and method, in order to understand the characteristics of publicadministration development in different locations.This study concluded several important findings: (1)There are increases innumber of theses published, which reflected the interest in public administrationstudies. However over half of the graduate theses take place in northernTaiwan, less than ten percent in eastern Taiwan. (2)Research interested focuson public policy, public administration and political science, but research onlaw and government budget are still limited. The choice of research topicsshows significant divergence in geographic location. (3)Over half of the thesesare qualitative, only small percentage of theses are quantitative. (4)In sampledesigning, most theses choose non random sampling, statistical analysis methodfocus on univariate and multivariate analysis, still a lot of room for improvement.