近年來網際網路的快速發展，線上學習平台掀起了一個浪潮，促使了大規模開放線上課程（Massive Open Online Courses, MOOCs）的興起，也就是「磨課師」，學生只需連上網路，學習則可以不受時間、地點的限制。MOOCs紀錄學生的線上學習操作動作，例如說影片觀看、測驗學習、討論區互動…等，目前這些紀錄只有一般的描述性統計，MOOCs的統計資料還未有更深入的探討與分析，本研究針對這些統計資料作分析，探討學生的學習記錄是否影響他們的學習成效，找到影響學習成效的關鍵性行為，可作為後續課程的學生在影片學習過程中，老師可以給予適當的回饋及建議。 本研究針對MOOCs記錄的影片學習部份研究，探討學生在瀏覽影片時有哪些影片瀏覽模式，以及影片瀏覽模式如何影響學生的學習成效。本研究使用了三種方法論，使用滯後序列分析(Lag-sequential Analysis, LSA)方法，建構出不同學習成效之中學生的影片瀏覽模式；接著使用探索性因素分析(Exploratory Factor Analysis, EFA)方法將學生有哪些影片學習行為面向分類；最後使用多重因素分析(Multiple factor analysis, MFA)方法，找出學生在影片學習中，不同的學習成效在不同的學習行為面向中有哪些關鍵性的學習行為。 ;The rapid development of the Internet has created a wave of online learning platforms that have prompted the rise of Massive Open Online Courses (MOOCs). Students only need to connect to the Internet to learn without being restricted by time or place. MOOCs record students′ online learning operations, such as course video viewing, quiz learning, discussion forum interactions, etc. At present, these records are only general descriptive statistics, and the statistics of MOOCs have not been further explored and analyzed. In this study, we analyzed these statistics to explore whether students′ learning records affect their learning performances, and attempted to find the crucial video viewing patterns that would affect their learning performances so that teachers can give appropriate feedback and suggestions in the following courses. This study used three methodologies. To begin with, the Lag-sequential Analysis was used to extract the viewing behavior from three groups of different learning performances in order to construct the categories of the students′ video viewing patterns. Next, the Exploratory Factor Analysis was employed to classify the diverse aspects of online video learning behavior. Finally, the Multiple Factor Analysis was applied to find out the crucial viewing motifs of the students from the three groups of different learning performances.
