闡釋多模態單字教學設計：多模態分析與滯後序列分析

Abstract

多模態在語言教育中扮演著至關重要的角色，可協助學習者傳遞知識或獲取新訊息。研究顯示，學習者對各種不同模態的使用抱持正面評價，但也擔憂此種多模態創作無益於語言發展。根據理論（Krashen, 1982; Mayer, 2001; Paivio, 1986; Schmidt, 1990），語言和非語言模態的結合其實有助學習，特別是在字彙發展方面，英語教室內也常見多模態學習資源，但很少相關研究分析著重於英語教師的多模態課程設計。因此，本研究採用深度式的單案例研究方法來分析一位教師的多模態字彙教學。選定的研究對象在台灣北部的一所高職任教，其課堂錄影、課程包和半結構化訪談結果為此深度式研究的資料集。多模態分析的語料庫包含49個字彙的教學錄影，本研究參考七種模態（Cope& Kalantzis, 2020）、多媒體學習的常見學習資源（Mayer, 2001）和單字學習內容（Nation, 2001）進行影片編碼，辨別出教師所使用的多模態種類後，進一步利用滯後序列分析法（Bakeman & Gottman, 1997）來分析模態之間的轉換顯著性、並繪製事件轉移圖以呈現單字教學的序列。 將統計結果與教師的陳述進行了交叉分析後，本研究發現，教師傾向（1）改變文本的字體和佈局來減輕單字學習負擔，（2）使用圖形或實物例子來則增進學生對於單字意思的掌握，以及（3）提供多種聲音資源來協助學生熟悉單字發音。此外，教師教學中兩個最顯著的序列，即在展示圖片後重複定義和在播放錄音後重複發音，可以提高學習者對於多模態理解性。而兩個教師提及的序列未能顯著，原因在於教師會考量單字種類、多變課堂環境，彈性調整教學順序。總而言之，希望本研究能提供多模態設計參考方向也希望研究結果能強調一些潛在影響因素。

Multimodality plays a crucial role in language education. Modes canserve either as learners' output for knowledge representation or input for information acquisition. Researchers of multimodal composing have revealed learners' positive feedback toward using various modes, while concerns about linguistic development have been raised. According to many theories (Krashen, 1982; Mayer, 2001; Paivio, 1986; Schmidt, 1990), combining verbal and non-verbal modes facilitates learning, especially vocabulary development. English teachers tend to integrate multimodality in the classroom, but theexamination of their lesson designs is scant. Therefore, this study adopts an embedded single-case study to analyze a teacher's multimodal vocabulary teaching. The selected teacher works at a vocational high school in northern Taiwan. The study collected three datasets: classroom recordings, lesson packages, and semi-structured interview results. Video recordings of 49 vocabulary lessons were collected as a corpus for Multimodal Analysis. The presented modes were encoded based on seven meaning forms in Transpositional Grammar (Cope & Kalantzis, 2020), common input for multimedia learning (Mayer, 2001), and Nation's (2001) framework of vocabulary knowledge. Lag Sequential Analysis (Bakeman& Gottman, 1997) further analyzed and visualized the transitional patterns between modes. We triangulated the statistical results with the teacher's statements. We found that the teacher tended to (1) change the text's fonts and layouts to alleviate learning burdens, (2) use graphics or physical examples to enhance meaning acquisition, and (3) provide multiple audio resources to familiarize learners with spoken forms. In addition, the two most significant sequences, repeating the definition after presenting pictures and repeating the pronunciation after playing the recordings, could increase the comprehensibility of input available to learners. Two teacher-reported sequences failed to be significant due to dynamic classroom conditions and the nature of the target word. To conclude, this study is important in providing teachers with a reference for multimodal designs and in highlighting some latent influential factors.