英語線上影音學習平台的可讀性及字彙涵蓋量

Abstract

在語言教學與學習中，影片教材被廣泛運用，因其能提供有意義的情境，讓EFL學習者能接收真實的語言輸入及文化資訊。近年來許多線上平台提供影片作為教學與自學的題材，在這其中，VoiceTube 與 Videos in Levels 各自都為英語學習者提供了六種不同難易度的影片。然而目前就這兩個平台所要求的語言能力及各等級在詞彙及句法複雜度上的差異仍所知有限。本研究探討 VoiceTube 與 Videos in Levels的語言能力要求，並使用語料庫分析法就以下三個面向進行探究：(1)兩個平台影片逐字稿的可讀性分析；(2)兩個平台的字彙覆蓋量分析；(3)兩個平台影片對應之CEFR（歐洲共同語言參考架構）等級分析，研究工具使用Lexile Analyzer來分析可讀性，針對詞彙覆蓋率使用軟體AntWordProfiler並搭配英國國家語料庫(BNC)/美國當代英語語料庫(COCA)詞彙表進行分析，最後以Text Inspector探討影片對應之 CEFR 等級。研究結果如下：(1)可讀性分析顯示VoiceTube各等級間的語言複雜度呈現較穩定的上升趨勢， Videos in Levels的語言複雜度則並不總是隨著影片程度而上升，(2)詞彙涵蓋量分析顯示VoiceTube的字彙涵蓋量有較穩定的成長趨勢，若以95%字彙涵蓋量作為標準，VoiceTube 影片涵蓋約3,000到8,000字族(word family)，同時並包含專有名詞(proper nouns)及邊際詞彙(marginal words)，Videos in Levels的詞彙涵蓋量則並非總是隨著等級而擴大，英語學習者只需約2,000至4,000字族加上專有名詞及邊際詞彙即可理解大部分的影片，另外若以98%字彙涵蓋量作為目標，學習者則須要6,000～16,000字族以及4,000～8,000字族加上專有名詞和邊際詞彙來理解VoiceTube以及Videos in Levels的影片，(3) CEFR難度分析表示VoiceTube的影片難度約落在B1~C1之間，Videos in Levels則是在A2~B2之間，兩個平台的影片實際難度都比平台所標示的難度來的高。以上結果顯示VoiceTube影片難度高於Videos in Levels，且就語言難度及複雜度來看，VoiceTube的等級分類優於Videos in Levels，在語言學習上VoiceTube能提供符合學生程度的影片素材，本研究最後針對教學實務以及影片難度分類提出建議，並指出未來的研究方向。In language teaching and learning, audio-materials have been widely used. It provides meaning context where EFL can receive authentic input and cultural information. There are many online resources that provide access to videos as materials for teaching and self-learning materials. VoiceTube and Videos in Levels (VIL) offer videos at six levels of difficulty for students with different language proficiency. However, little is known about the language demand of both websites and how each level differs from each other in terms of vocabulary and sentence complexity.Aiming at exploring the language requirement of the two platforms, the present research adopted corpus-based study to examine the following aspects: (a) the readability of VoiceTube and Videos in Levels (b) lexical coverage of both platforms (c) corresponding CEFR levels of videos at both platforms. Lexile Analyzer was used for the readability of transcripts. Software AntWordProfiler was adopted with BNC/COCA wordlists for lexical coverage. Lastly, Text Inspector was employed to examine how videos from both platforms correspond to CEFR levels.Results show that (1) VoiceTube shows a more stable growth in readability according to Lexile measures while the readability for VIL does not always become higher as the level increases. (2) Lexical coverages of VoiceTube expand more steadily than VIL. Vocabulary coverages of VoiceTube expand from the most frequent 3,000 to 8,000 word families with proper nouns (PN) and marginal words (MW) for 95% coverage. However, lexical coverage does not always become broader as the level increases on VIL. EFL students need to recognize the most frequent 2,000 to 4,000 word families with PN and MW for most of the videos on VIL. The results also suggest that learners must expand their vocabulary size of 95% by two times more to reach 98% of optimal comprehension. (3) Text Inspector shows that videos of VoiceTube are between B1~C1 while those of VIL are between among A2, B1 and B2 according to CEFR, which shows that the actual language complexity of videos are higher than the assigned levels by the two platforms.Based on the findings of present study, VoiceTube is more difficult than VIL. VoiceTube has better video classification than VIL in terms of language complexity and difficulty. It is believed that VoiceTube can provide videos that are more appropriate for language learning. Further pedagogical implications and suggestions for future research are addressed in the last chapter.