影響科技大學學生流失因素之初探研究－以北區某科技大學為例

本研究旨在瞭解科技大學學生個人因素、校園經驗對於持續就讀或流失的關係。除探討學生個人因素對於校園經驗的關聯外，更進一步探討校園經驗對於學生持續就讀或流失之預測力。本研究對象為北區某科技大學103學年度持續就讀或流失學生進行調查，做出結論並提出建議，以提供學校相關單位、教師作為日後協助學生持續就讀及相關政策之參考。本研究共發出300份問卷，回收249份，有效回收率為83%。所得資料以IBM SPSS Statistics 20.0 for Windows進行問卷信效度、描述統計、相關分析、逐步迴以及邏輯斯迴歸分析。本研究結果歸納如下：一、學生的高中（職）三年的學業總成績對於滿意度、目標發展與承諾沒有關聯程度，但對於學術投與人際投入有正向的關聯程度；而社經地位對於各變項間沒相關性，亦即家庭社經地位並不會影響到學生的校園經驗。二、滿意度與學術投入與人際投入之間有著正向中度的相關性，因此學生的學術投入與人際投入確實會影響到學生的滿意度；而學生目標發展與承諾、學術投入和人際投入有著正向中度的相關性。三、學術投入、人際投入、學生目標發展與承諾和滿意度等變項與學生就讀意願達顯著水準，共可以有效解釋就讀意願變項67%的變異量。四、學術投入、人際投入、學生目標發展與承諾和學生滿意度等變項，所建立的邏輯迴歸模式之整體適配達顯著水準，並可以有效預測學生持續就讀或流失。

關鍵字

科技大學；學術投入；人際投入；校園經驗；持續就讀

並列摘要

This paper aims at exploring how students’ personal reasons and their campus experience affect them to continue studying or to become dropouts. In addition to the relevance between students’ personal reasons and their campus experience, we further discuss the predictive power of campus experience. Our samples are students of the school year 103 from one University of Technology in northern Taiwan. We collected the data of both persistent and dropout students, and then provided results and suggestions. A total number of 300 questionnaires were released; 249 of them were collected afterwards. The effective response rate reached 83%. IBM SPSS Statistics 20.0 for Windows was applied on the collected data for questionnaire reliability and validity, descriptive statistics, correlation analysis, stepwise regression analysis and logistic regression analysis. Research results are summarized as the following: 1. There is no correlation between students’ high school (vocation school) three-year scores and their satisfaction degree, goal development and commitment. However, there is positive correlation toward academic and human relationship input. Furthermore, students’ family socioeconomic status has no correlations against all variables, which reveals that family and socioeconomic status have no influence on students’ campus experience. 2. Students’ satisfaction degree shows positive middle correlation toward their academic and human relationship input; thus, there will be influence from above inputs to students’ satisfaction degree. Also, there is positive middle correlation between students’ goal development and commitment, academic and human relationship input. 3. Variables like academic input, human relationship input, students’ goal development, goal commitment and satisfaction degree reach a significant level toward students’ willingness to continue studying. This can explain 67% variance for students’ willingness to continue studying variable. 4. The overall fitness of logistic regression model built up by variables including academic input, human relationship input, students’ goal development, goal commitment and satisfaction degree -- all has reached a significant level. As a result, it can effectively predict students’ choice to continue studying or to become dropouts.