本研究旨在探討中彰投偏遠地區國小教師留任意願之影響因素，以中彰投三縣市偏遠地區326位國小教師為研究對象，並使用描述性統計、獨立樣本t考驗、單因子變異數分析及多元迴歸分析等統計方法進行分析。本研究結果如下：（一）中彰投偏遠地區國小教師留任意願屬中等程度。（二）中彰投偏遠地區51歲以上國小教師之「自願性留任意願」高於其他年齡教師；已婚教師之「自願性留任意願」高於未婚教師；南投縣教師之「自願性留任意願」高於彰化縣與臺中市教師；在校服務11年以上教師之「自願性留任意願」高於服務6－10年及5年以下教師；兼任行政職務教師之「自願性留任意願」高於科任與級任教師；本地教師之「自願性留任意願」高於外地教師。（三）31－40歲教師之「非自願性留任意願」高於其他年齡層教師；在校服務5年以下教師之「非自願性留任意願」高於服務6－10年及11年以上教師；外地教師之「非自願性留任意願」高於本地教師。（四）婚姻狀況與服務縣市及學校認同度對留任意願具有顯著預測力，其中又以學校認同度的預測力最高。最後根據研究發現提出具體建議，以供教育機關決策、偏遠地區國小行政及未來研究者參考。

The purpose of this study was to explore the influence factors of retention of 326 rural area elementary school teachers in Taichung, Changhua, and Nantou. Descriptive statistics, independent sample t-test, one-way ANOVA and multiple regression analysis were applied to analyze data. The findings are as follows. First, teachers perceived a moderate level of retention. Second, in terms of voluntary retention, the scores of teachers over 51 years old were significantly higher than those younger than 50 years old. The scores of married teachers were significantly higher than unmarried teachers. The scores of teachers in Nantou were significantly higher than that of Taichung and Changhua. The scores of those with teaching years of employment over 11 years were significantly higher than those with teaching years of employment less than 11 years. The scores of teachers with an administrative position were significantly higher than those of subject teachers and homeroom teachers. The scores of local resident teachers were significantly higher than of non-resident teachers. Third, in terms of involuntary retention, the scores of teachers between 31 to 40 years old were significantly higher than other age groups. The scores of teachers with teaching years of employment less than 5 years were significantly higher than those with teaching years of employment more than 6 years. The scores of non-resident teachers were significantly higher than those of local resident teachers. Fourth, the variables of material status, different cities, and school identification significantly predict retention, and the variable of school identification was the strongest one. The suggestions of the findings were made for educational authorities, rural area elementary schools, and future researchers.