線上語料庫檢索查詢對臺灣高中生搭配詞與字彙定義學習之效能研究

The Effectiveness of Online Corpus Consultation on Developing Taiwanese High School Students’ Collocational Knowledge and Enhancing Vocabulary Learning In Terms of Multiple Definitions

陳怡倩(Yi-chien Chen)

指導教授 ： 林至誠

國立臺灣師範大學/文學院/英語學系/碩士(2011年)

摘要

本研究的目的是了解線上協語器教學對台灣高中生在學習英語搭配詞與字彙定義學習之效能。七十六名台北市某所市立高中高一學生參加此項研究。首先，他們被區分為實驗組與控制組，實驗組接受密集的線上協語器使用訓練之後，展開為期十週的線上協語器教學：他們每週登入Moodle網頁，下載教學投影片，並完成線上練習。另外一方面，控制組接受傳統的搭配詞教學—每週獲得講義，其教學內容及呈現方式與實驗組相同，他們也需登入專屬的Moodle網頁，完成線上練習。本研究使用四種方式來檢測所有參與學生的英語字詞搭配能力；分別是三份相同題目的考題（其中包含填充、翻譯及選擇題）、十次的線上學習記錄和成績、一份問卷、及半開放性訪談。其中，這三份考題將分別使用於前測、後測及延遲後測（十日後），並以二因子變異數分析（Two-way ANOVA）教學方法與三次考試對學生成績的影響。
研究結果分量化和質化分析。首先，數據顯示，經過十週的搭配詞教學，實驗組和對照組在搭配詞的學習成效上有顯著差異，且教學方法和考試之間無交互作用。兩組的搭配詞知識在分項探討中均有顯著增強：這五種不同類型的搭配詞包括同義詞（synonyms）、上位詞和下位詞(hypernym and troponyms)、中英文互不對照之動詞（verb-noun collocation that are not congruent in Chinese and English）、虛化動詞（delexicalized verbs）與文化特定搭配詞（culture-specific collocations）；其中受試者在學習同義詞詞組時有顯著的差異。第二，在測驗的三種題型中（填充題、引導式翻譯題、選擇題），填充題為受試學生答題情形最差的題目，然而，卻也是在延遲後測中為唯一有顯著進展的。第三，實驗組在詞彙知識方面亦有顯著的進步，研究發現實驗組獲得了部分沒有包含在課程中的搭配詞和字義的知識。除了量化分析之外，本研究亦採用質化分析。首先，實驗組在線上的表現（較高的出席率和較好的週測驗成績），與他們在後測的表現相符。其次，在問卷和面試中，受試者給予正面和負面的反饋。前者包括他們願意撥出時間在線上學習，儘管過程繁瑣、他們渴望好成績、以及他們會將以前的知識應用到新獲得的信息。後者則包括承認自己難以專心於課程和無法分析從網路上獲得的大量訊息。在面試中，也有人指出，使用協與器和搭配詞查詢工具是一個全新的詞彙學習方法，也因此，對於高一生而言，過程中遭遇到的問題和挑戰往往會影瞎們在選擇未來使用此學習方法的意願。也就是說，覺得這些工具有所幫助並開始增加自己的字彙量的人願意持續使用，而那些認為過程繁瑣無趣的則採保留的態度。
本研究認為，若學生能經過長期訓練、分析能力培養和習慣養成，雙語協語器對學生的搭配詞學習與字彙定義學習是有效的。高中教師可以將協語器和搭配詞查詢工具融入於字彙教學中，而藉由明確的運用中文，並使學生有更多接觸這些例子的機會，學生便能成功地加強字彙和搭配詞學習。

關鍵字

語料庫檢索 ； 搭配詞學習

並列摘要

The study aims to investigate on the effectiveness of online corpus consultation on developing Taiwanese high school students’ collocation knowledge and enhancing vocabulary learning in terms of its multiple definitions. There were 76 Taipei municipal high school freshmen recruited in the study. They were divided into two groups: the experimental group and the control group. The experimental group first received intensive training on corpus consultation through a bilingual concordancer, TotalRecall and a collocation retriever tool, Tango. Afterwards, they took a ten-week web-based course, in which they logged in a Moodle webpage, downloaded instructional powerpoint files, performed assigned queries via the two tools and completed online practices. On the other hand, the control group received normal collocation instruction by means of handouts; the content and the way of presentation were identical with that of the powerpoint files. They were also required to complete the same online practices on Moodle. To examine their growth, the present study adopted four kinds of instrument: three tests, ten online quizzes, a questionnaire and a semi-structured interview. The three tests, the pretest, the posttest and the delayed posttest (10 days after the immediate posttest), were identical in the test items, but items were rearranged in a different order. The data were later analyzed through two-way repeated measure ANOVA.
Results were presented both quantitatively and qualitatively. The quantitative results showed that after ten weeks of treatment, there was significant difference between the two groups and that no interaction effect was observed between the two variables: teaching approach and time of measurement. Findings of the test results were listed as follows. First, both groups were discovered to have enhanced their collocational knowledge in every collocation type categorized in the study, including the synonym verb pairs, hypernymy and troponymy verb pairs, delexicalized verbs, the V-N collocations that were non-congruent in Chinese and English, and culture-specific collocations. The results showed that the students had successful collocation learning in synonym verb pairs. Secondly, the participants responded differently to different task types, filling-in-the-blank, guided translation and multiple choice. Among the three task types, filling-in-the-blank questions were considered the most daunting as the scores were the lowest; yet statistics suggested the most prominent improvement in the delayed posttest. Thirdly, analysis of the students’ item responses also revealed that the experimental group implicitly deepened their vocabulary knowledge in terms of its multiple meanings. Significant improvement was discovered in certain word meanings that were designed in the tests, but not instructed as the course content. From their significant improvement in these items, the students’ word knowledge gains were observed. In addition to the quantitative data, qualitative data were under scrutiny along with the participants’ online learning behavior. First, the experimental group had a slightly higher attendance rate and better performance in the online quizzes, which corresponded to the results of the final performance. Secondly, in the questionnaire and the semi-structured interview, the participants gave both positive feedback, such as willingness to allocate time for the online course despite the tedious practices, desire for good grades, and application of prior knowledge to the newly-gained information, and negative feedback, such as the struggle to concentrate on the online course and inability to analyze the large amount of data. In the interview, it was also observed that collocation consultation was a brand new experience of vocabulary learning for the high school freshmen and that problems and challenges encountered in the process easily influenced the participants’ application to future vocabulary learning. That is, those who enjoyed the benefits and started to build their own vocabulary bank responded positively while those who suffered from the tedious work held a reserved attitude.
The results of the present study supported that with constant training, analytical skills cultivation and habit formation, bilingual corpus consultation were effective to assist EFL students in learning collocation and deepening vocabulary knowledge in terms of its multiple meanings. High school teachers can incorporate concordancers and collocation retriever tools into regular vocabulary lessons for their long-term effects. It is also suggested that future collocation learning should first draw students’ attention to L1 explicitly and then implicitly expose them to as many examples as possible. In this way, L1 can scaffold their learning of collocations and word meanings successfully.

並列關鍵字

Corpus consultation ； Collocational Knowledge

參考文獻

Huang, H. C., Chern, C. L., & Lin. C. C. (2006). EFL Learners Online Reading Strategies: A Comparison Between High and Low EFL Proficient Readers. English Teaching & Learning, 30(4), p.1-22.

Yo, Y. T. (2005). Effects of automatic essay grading system and bilingual concordancer on EFL college students’ writing. National Tsing-Hua University, Hsin-Chu, Taiwan.

Barlow, M. (1996). Corpora for theory and practice. International Journal of Corpus Linguistics, 1(1), 1-37.

Benson, M., Benson, E., & Ilson, R. (1999). The BBI Dictionary of English Word Combinations. Taipei: Bookman.

Boechler, P. M. (2001). How spatial is hyperspace? Interacting with hypertext documents: cognitive processes and concepts. Cyber Psychology and Behavior, 4 (1), 23–46.