程涂媛 Tuyuan Cheng

以資料驅動途徑選取老化相關詞彙應用於護理專業英文課程之教學與評量
A Data-driven Approach to Select Aging-related Vocabulary for a Nursing ESP Course in Instruction and Evaluation

The past decades have witnessed rapid advances in the field of corpus studies. Among them, in particular, investigating the effectiveness of data-driven teaching and learning (DDTL) has achieved increasing prominence, through which teachers can discover and adopt the data derived for students to learn. This study focuses on the use of scripts accessed from TED spoken corpora to select Aging-related vocabularies and on the integration of that lexis into a Nursing English for Specific Purposes (N-ESP) course.
TED talks have comprised an emerging genre, being hailed as a "powerful way to convey an innovative idea to a giant global audience". In designing the N-ESP course for EFL (English as a Foreign Language) instruction, the access to such internet databases of words (language corpora) in TEDs has provided a limitless number of various types of vocabulary used in authentic texts for teachers. In this data-driven study, the scripts of 50 TED talks, selected by key word search on topics AGING, ALZHEIMER, and BRAIN, consisting of 104,000 running words were analyzed using the RANGE program (Heatley et al. 2002) to determine the number of encounters with high frequency words, resulting in three base-word lists by default. The compilation of these word lists were then calculated for its vocabulary load using Nation’s (2012) BNC/COCA lists with the AntWord Program. This tool allows users to view an individual user file and highlight the different levels of vocabulary in the file using a color coding. It also shows the overall coverage of different vocabulary levels . The target 20 words for five TED talks constitute a vital 100-high-frequency word set of vocabulary for the N-ESP nursing students at the beginning of their studies. Then, the instructor uses these corpus-based search into the N-ESP course for the Pre-Test, Instruction, and Post-test. Integrating these procedures into an ESP course for nursing students by using Nation’s (2013) framework of planning, strategy training, testing, and teaching vocabulary is further discussed.
Above all, how to realize these corpus-based techniques into N-ESP courses is an important task. Empirical evidence on this approach can be obtained by investigating whether DDTL is effective in vocabulary learning compared to traditional instruction (TI). The experimental procedures are as follows. (1) The five TED talks related to aging healthcare are decided. Twenty words selected from the data to be learned and tested were embedded in the pre-/posttest. (2) The five talks are uploaded to the online platform, Moodle, for the learners to preview in advance for the Flipped classroom discussion. (3) Before the classroom discussion, a pretest is held to ask students to make a judgement on a sentence containing the first half 10 words (4) After discussion, a comprehension and word choice test (posttest) on the second half 10 words are held immediately. Explicit instructions on the talk and the vocabularies are conducted as well.
The Pre-/Posttest analyses reveal that, although both DDTL and TI groups performed in the posttests better than the pretest, the DDTL group achieves significantly higher scores than the TI group on the immediate posttest. The results suggest that the potential for vocabulary learning through data-driven approach on analyzing TED scripts in the contexts with conveyed meanings can be used for explicit/implicit vocabulary teaching/learning and effective evaluation in a nursing ESP classroom.