全民英檢聽力測驗題型對國中生聽力策略選擇影響之研究 - 政大學術集成

全文

(1)國立政治大學英國語文學系碩士在職專班碩士論文. 指導教授：黃淑真博士 Advisor: Dr. Shu-Chen Huang. 立. 政治大. ‧ 國. 學. 全民英檢聽力測驗題型對國中生聽立策略選擇影響之研究 A Study of Effects of Question Types on Junior High Students’ Listening. ‧. Strategy Choice While Taking GEPT Listening Test. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. 研究生：何佩融撰 Name: Pei-Jung Ho 中華民國一百零二年七月 July, 2013.

(2) A Study of the Effects of Question Types on Junior High Students’ Listening Strategy Choice While Taking GEPT Listening Test. A Master Thesis Presented to. 政治大. Department of English,. National Chengchi University. 學. ‧ 國. 立. ‧ er. io. a. sit. y. Nat In Partial Fulfillment. n. v l of the Requirements for the Degree n iof Ch. U i e h n c g Master of Arts by Pei-Jung Ho June, 2013.

(3) Acknowledgements My greatest thanks go to Dr. Shu-Chen Huang for her inspiring guidance and encouragement throughout my research. She always reminds me to be a careful researcher and leads me to think meticulously about the study. I also want to thank Dr. Chieh-Yue Yeh and Dr. Wei-Yu Chen, for their reading of the draft and for their helpful suggestions. I really appreciate my classmates in ETMA, and my colleagues in. 政治大 Kai-Syuang Junior High School. I could not finish this work without their assistance. 立. ‧ 國. 學. Finally, I am really grateful to my family. My husband, Ken Chen, always encourages. ‧. me and helps do the house chores. My mother-in-law and father-in-law support me by. sit. y. Nat. taking care of my baby. Special thanks go to my parents, their everlasting love make. n. al. er. io. this work possible.. Ch. engchi. iii. i n U. v.

(4) Table of Contents Acknowledgements.......................................................................................................iii Chinese Abstract .........................................................................................................viii English Abstract.............................................................................................................x Chapter One: Introduction .........................................................................................1 Background and Motivation ......................................................................................1 Purpose of the Study ..................................................................................................4. 政治大. Research Questions....................................................................................................5. 立. Chapter Two: Literature Review................................................................................7. ‧ 國. 學. The Process of Listening Comprehension .................................................................7. ‧. Features of Spoken Language................................................................................9 Variables Influencing Listening Comprehension.....................................................10. y. Nat. sit. Text Factors in Listening Comprehension ........................................................... 11. n. al. er. io. Issues on Listening Strategies..................................................................................17. i n U. v. Strategy Use Under Listening Test-Taking Condition.........................................18 Taxonomy of Listening Comprehension Strategies .............................................19. Ch. engchi. Related Studies of Listening Comprehension Test and Listening Strategies...........21 Concluding Remarks of Chapter Two......................................................................25 Chapter Three: Methodology....................................................................................27 Participants...............................................................................................................27 Instruments...............................................................................................................27 Procedures................................................................................................................32 Data Analysis ...........................................................................................................35 Chapter Four: Results ...............................................................................................37 iv.

(5) Research Question 1 ................................................................................................37 Research Question 2 ................................................................................................40 Research Question 3 ................................................................................................46 Chapter Five: Discussion and Conclusion ...............................................................57 Students’ Performance on Different Question Types...............................................57 Strategy Choice among Four Question Types..........................................................63 Language Proficiency Question Types and Stratey Use ..........................................67. 政治大. Pedagogical Implications .........................................................................................68. 立. Limitation of the Present Study and Suggestion for Future Research .....................70. ‧ 國. 學. Reference .....................................................................................................................73. ‧. Appendix......................................................................................................................81. y. Nat. APPENDIX A..............................................................................................................81. n. al. er. io. sit. Qusetionnaire of Listening Strategy Use .............................................................81. Ch. engchi. v. i n U. v.

(6) List of Tables Table 2.1. Vandergrift’s Classification of Listening Comprehension Strategies.20. Table 3.1. The Structure of GEPT Listening Comprehension Test .....................28. Table 3.2 Speech Rate of Four Sections in GEPT Elemantary Level Listening Test.............................................................................................................................29 Table 3.3 The Classification of the Strategy Items in the Listening Comprehension Strategy Questionnaire ................................................................30 Table 3.4. Description of Items ...............................................................................31. Table 4.1.1 Students’ Scores on Listening Comprhension Test..............................38. 政治大 Table 4.1.3 Scheffe Post Hoc 立 Comparison of Listening Performance among. Table 4.1.2 Students’ Scores among the Four Question Types...............................39. Question Types .........................................................................................................39. ‧ 國. 學. Table 4.2.1 Most Frequently Used Strategies in the Section of Picture Description ....................................................................................................................................41. ‧. sit. y. Nat. Table 4.2.2 Most Frequently Used Strategies in the Section of Answering Questions...................................................................................................................42 Table 4.2.3 Most Frequently Used Strategies in the Section of Conversations.....43. io. n. al. er. Table 4.2.4 Most Frequently Used Strategies in the Seciton of Short Talks .........44. Ch. i n U. v. Table 4.2.5 Frequently Used Strategies in the Four Sections .................................45. engchi. Table 4.2.6 The Effect of Question Types on Strategy Choice................................46 Table 4.3.1 The Interaction Effect of Sections and Proficiency Level...................47 Table 4.3.2 Strategies with Significant Difference for Students with High/Low Levels in Section One...............................................................................................48 Table 4.3.3 Strategies with Significant Difference for Students with High/Low Levels in Section Two...............................................................................................49 Table 4.3.4 Strategies with Significant Difference for Students with High/Low Levels in Section Three............................................................................................50 Table 4.3.5 Strategies with Significant Difference for Students with High/Low Levels in Section Four .............................................................................................52 Table 4.3.6 Strategies with Significant Difference for Students with High/Low vi.

(7) Levels in Four Sections ............................................................................................53 Table 4.3.7 Strategies that High/Low Achievers Used Differently in Some of the Four Sections ............................................................................................................55 Table 5.1 Features of Each Question Types ..........................................................59. List of Figure A Flowchart of the Research Procedures ...........................................33. 立. 政治大. 學 ‧. ‧ 國 io. sit. y. Nat. n. al. er. Figure 3.1. Ch. engchi. vii. i n U. v.

(8) 國立政治大學英國語文學系碩士在職專班碩士論文提要論文名稱：全民英檢聽力測驗題型對國中生聽力策略選擇影響之研究指導教授：黃淑真博士研究生：何佩融論文提要內容：本研究旨在比較國中學生在全民英檢聽力測驗中不同聽力題型之測驗結. 政治大本研究對象為宜蘭一所公立國中八年級六個班級共 177 位學生。研究工具為立. 果，並探究聽力題型是否造成學生聽力策略選擇之差異。. 全民英檢初級聽力測驗以及聽力策略問卷。問卷內容以 Vandergrift(1997)的聽. ‧ 國. 學. 力策略分類為架構，改編自蘇曉雯(2007)自編問卷。問卷回收後資料以統計軟體. io. y. sit. 國中學生在全民英檢聽力測驗四種題型中的聽力表現有所差異。四種題型的. er. 1.. Nat. 簡述如下：. ‧. SPSS 17.0 進行敘述性統計、單因子變異數分析及雙因子變異數分析，獲得結果. 不同因素包括了聽力內容中參與談話者的人數、聽力內容所包含的難字以及測驗. n. al. 題型是否提供圖片輔助。 2.. Ch. engchi. i n U. v. 整體來說，在聽力測驗作答時，國中學生較常使用後設認知策略。例如:理. 解監測(monitoring)，選擇性專注(selective attention)和前導組織(advance organizers)。然而，若細究不同的聽力題型，學生則較常使用不同認知策略。例如:當遇到較長的聽力內容，學生較常使用摘要(summarization)策略。而遇到較難的聽力內容，學生則慣用由下自上(bottom-up)策略。 3.. 綜觀不同語言能力之學生在不同題型中使用聽力策略的差異，本研究發現學. 生之語言能力與題型對聽力策略使用上並無交互作用。整體而言，高低成就學生. viii.

(9) 在四種不同題型中使用相似的聽力策略。然而，若細究各項不同聽力策略的使用頻率，則高低成就者中所慣用的聽力策略有所不同。根據統計及問卷調查結果，本研究最後提供教學上相關建議以供參考。第一，教師可協助學生分析試題題型，以幫助作答。第二，教師應提供學生不同類型之聽力練習，讓學生嘗試使用不同聽力策略。第三，教師可請高成就學生分享聽力理解過程，以幫助低成就學生能有更好的聽力表現。. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. ix. i n U. v.

(10) Abstract The major purpose of the study is to examine whether question types influence junior high students’ listening performance and their listening strategy choice. The study is mainly concerned with three aspects: (1) whether students performed differently among the four different question types of GEPT listening section; (2) what kind of strategies did junior high students tend to use in the four types of questions; (3) whether proficiency level played a significant role in listening strategy. 政治大 GEPT elementary level of listening comprehension test and a questionnaire were 立. choice among four question types.. ‧ 國. 學. used to collect quantitative data from 177 students in one public junior high school in Yilan Area. The items in the questionnaire were mainly adapted from Su (2007),. ‧. which followed Vandergrift’s (1997) classification of strategy use. In this study,. sit. y. Nat. statistical analysis, including descriptive statistics, one-way ANOVA with Scheffe. io. al. The major findings are summarized as follows:. v i n Junior high students performed C differently in the four U different types of listening heng chi n. 1.. er. post- hoc test and two-way ANOVA were used to analyze the collected data.. questions. Factors that differentiate the four question types include the number of speaker, the content itself and the presence of pictures. 2. Generally speaking, junior high students tended to use metacognitive strategies, such as monitoring, selective attention and advance organizers more often in different question types. However, in specific question types, students appeared to have preference in using certain cognitive strategies. For example, students were. x.

(11) inclined to use summarization when encountering question type with longer listening text. They also tended to use bottom-up strategies to tackle with more difficulty text. 3.. There was no significant interaction effect between question types and. proficiency level. Overall on the four question types, students with high and low levels used same strategies. However, when examining strategy items specifically, high and low achievers had different preference for using strategy in different question types.. 政治大 were presented in the conclusion of the paper. First, teachers could help students 立. Finally, based on the findings in this study, several suggestions and implications. analyze question types before listening test. Second, teachers could give students. ‧ 國. 學. more kinds of listening activities to make students experience the effectiveness of. ‧. using different kinds of listening strategies. Third, teachers could invite high level. y. sit. io. n. al. er. them.. Nat. listeners to share their listening process so that low level listeners might learn from. Ch. engchi. xi. i n U. v.

(12) 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v.

(13) Chapter One Introduction Background and Motivation Listening, a commonly used language skill, plays an important role in daily life. According to Rankin(1928), when people are awake, they spend 80% of their. 政治大. time communicating, and listening composes 45% of which. Also, Feyten (1991). 立. found that listening ability is an indicator of foreign language proficiency. Vandergrift. ‧ 國. 學. (1997) also asserted that listening is probably the most crucial language skills among the four at the beginning stage. It facilitates the emergence of other language skills. ‧. Nat. sit. learning, language teachers should put more emphasis on it.. y. such as speaking, reading and writing. Since listening is so critical in language. n. al. er. io. However, listening is often neglected in language teaching because it is a. i n U. v. receptive skill. Detecting the improvement and achievement of listening ability is. Ch. engchi. difficult. Language teachers usually follow the patterns of reading instruction to teach listening (Rost, 1994). Another reason of underestimating listening skill might be due to its fleeting characteristic. Lund (1991) once pointed out that listening process exists in time instead of in space. Because listening is ephemeral in nature, it becomes a problem for second language learners. Listeners could only retrieve what they have heard from their memory. And memory might not be reliable because it intertwines listening input and listeners’ interpretation together. Hasan (2000) made a comparison between listening and listening comprehension. He addressed that listening is a process of just listening to the input without interpreting but listening 1.

(14) comprehension is a process which contains the meaningful interaction between the text and listeners’ background knowledge. It is suggested that listening is an active, rather than passive process because listeners need to build up meanings by connecting what they hear and what they know (Berne, 2004; Mendelsohn, 1994; Rost, 2002). Knowing the features of listening does not guarantee successful listening comprehension. In fact, EFL listeners encounter great difficulty in listening (Flowerdew & Miller, 2005). They receive insufficient input, noise and unclear. 政治大 which cause them longer time to comprehend listening input than L1 listeners. There 立 sounds that hamper their comprehension. And they rely on controlled processing,. were several studies contributing to factors that influence listening comprehension.. ‧ 國. 學. Boyle (1984) categorized them as listener factors, speaker factors, and factors in. ‧. material and medium. According to Anderson and Lynch (1988), three major factors. sit. y. Nat. influencing listening performance were the types of language, the task of listening,. io. er. and the contextual situation of listening. In Rubin’s review (1994), there were five major factors, which were text characteristics, interlocutor characteristics, task. al. n. v i n C hand process characteristic. characteristics, listener characteristics, These elements engchi U might interfere in listening process and thus might cause failure in listening. comprehension. To deal with listening difficulties, listeners need to use some tactics to compensate for insufficient listening skills. Hence, several studies were contributed to the benefit of using listening strategies. The concept of listening strategies originates from studies in learning strategies. Some researchers compared the use of successful and unsuccessful learners and found that some strategies help learners to boost their learning outcome (O’Malley & Chamot, 1990; Purpura, 1997). The arising attention on learning strategy inspires more research on specific 2.

(15) language skills, such as listening. Generally speaking, listening strategies may be divided into three main types: cognitive, metacognitive, and social/affective strategies (Rost, 2002 ; Vandergrift, 1997). Listeners do not use all the strategies at one time. Instead, they choose what strategies they need according to different situations. Furthermore, studies were conducted to explore factors influencing strategy use. Some of them focused on the factors relating to task requirement. O’Malley and Chamot (1990) suggested that text factors were related to strategy choice. Bacon (1992) addressed that listening strategy was influenced by passage. 政治大 comprehension. However, the classification of task and text is inconclusive and thus 立 difficulty. It is proved that text factor plays an important role in listening. more research is needed.. ‧ 國. 學. According to Lewis (1999), the listening process is not visible; only the result. ‧. of the listening is visible. Therefore, using listening comprehension test to explore. sit. y. Nat. strategy use is necessary for it shows the outcome of listening process. And. io. er. researchers could further speculate on how listeners process aural input by questionnaires of listening strategy use. Most of the studies conducted to date were. al. n. v i n Cbetween mainly about the relationship useU and text type. Usually text type is h e n strategy i h gc defined by the number of speakers involved in the listening text. However, there are. still other elements to be considered in the listening text. Under the circumstances of listening comprehension test, there might be factors such as visual support, text length, familiarity of words and so on. Therefore, the present study aims to explore a broader aspect of listening text and listening strategy use. The listening text would be chosen from listening comprehension test. Since the listening text presented in listening test is in the form of questions, and the questions presented vary in type, listening text in the present study would be called as question types. The reason to call for a study in the relationship between question types and 3.

(16) strategy use is because of the great difficulty encountered by EFL learners in Taiwan. Chuang (2009) conducted a research to explore listening problems from students’ and teachers’ perspectives and found that students and teachers perceive different listening problems while listening. In Chuang’s (2009) study, students reported that the text factor was the hugest barrier to their listening, whereas teachers indicated that their students encountered listening difficulties in the listener factor most often. The results showed that teachers seemed not to understand why students failed in listening comprehension. But helping students respond correctly in listening exercise. 政治大 sources of listening input in students' learning experience. If students successfully 立. or listening test is important because listening exercise and test might be the major. answer listening comprehension questions, they might feel more confident in their. ‧ 國. 學. English ability and be more motivated in learning English. To answer listening. ‧. comprehension questions correctly, teachers could help students use the clues in the. sit. y. Nat. questions and instruct them to apply useful tactics. The present study probes into the. io. er. relationship between the question types and listening strategy use. It is hoped that teachers could select appropriate listening exercise and instruct students to use. al. n. v i n Ch facilitative listening strategy when encountering different question types in listening engchi U comprehension test.. The Purpose of the Study Studies related to relationship between question type and strategy use were mostly concerned with senior or college students (Luo, 2004; Su 2007). Little study is conducted to examine the effect of listening text on junior high school students’ listening strategy use. The purpose of the study was first to examine whether question types have effect on junior high school students’ listening comprehension performance. And, second, to investigate what listening strategy junior high school students use between different question types. Finally, the relationship between 4.

(17) question types, language level and strategy choice was discussed. The Research Questions 1. Do junior high students perform differently for different question types in GEPT listening comprehension test? 2. Are there differences between the strategies students use for different question types? 3. Do students with high/low proficiency levels demonstrate different patterns of strategy?. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 5. i n U. v.

(18) 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 6. i n U. v.

(19) Chapter Two Literature Review This chapter comprised four parts: first the process of listening comprehension were introduced. The second part portrayed the variables influencing listening comprehension. The issues in listening strategies were reviewed thirdly.. 政治大. Finally, related studies on the relationship between listening comprehension tests. 立. and listening strategy choice would be discussed.. ‧ 國. 學. The Process of Listening Comprehension. Listening used to be considered as a receptive skill. In traditional view, listeners. ‧. passively accept information like a tape-recorder. However, Anderson and Lynch. y. Nat. io. sit. (1988) deemed listening an interactive process. It requires listeners to apply. n. al. er. background knowledge to oral input. Listeners might hear the same information but. i n U. v. generate different interpretations. According to Anderson (1995), the first step of. Ch. engchi. listening is called perception. In this phase, listeners pay attention to input and a large number of information keep coming to be stored in the short-term memory, where specific details can only be retained for a short time. Because of limited capacity in the short-term memory, the information undergoes the process of being decoded. In the decoding process, listeners select important cues, such as pauses and hesitation, to help them comprehend meanings. Listeners also concentrate on task and are influenced by text types in this phase. The second stage of listening comprehension process is called parsing. Listeners apply existing knowledge in the long-term memory to represent the words just received in the short-term memory. 7.

(20) Therefore, what have already been saved in the long-term memory would affect listening comprehension. The last phase of listening comprehension is utilization. Listeners associate the meaning of listening text with declarative knowledge, which is stored in the long-term memory. Declarative knowledge could be divided into real world knowledge and linguistic knowledge. Using real world knowledge, which is related to the content itself, is a top-down process while using linguistic knowledge is a bottom-up process. The three phases of listening process mentioned above can be recursive and overlapping. Therefore, listening comprehension requires intricate. 政治大 interpretations by different strategies. 立. process in each step. Listeners passively receive input but actively produce personal. One study about listening processes was conducted by O’Malley, Chamot,. ‧ 國. 學. and Küpper (1989). They investigated 11 high-school ESL native speakers of Spanish.. ‧. Eight of them were grouped as effective listeners and three were ineffective ones. All. sit. y. Nat. subjects listened to academic lectures in a tape, and the lectures were imposed with. io. er. pauses. During each pause, the researchers then interviewed the subjects and asked them what came to their mind while listening. Unfortunately, ineffective listeners. al. n. v i n C h for them. Data collected quitted because the task was too difficult from the effective engchi U listeners showed that listeners used different strategies in each phase of listening. process. Effective listeners were conscious of attention problems and intended to deal with them during the perceptual stage. In the second stage, parsing, effective listeners used more top-down strategies than bottom-up strategies. During the phase of utilization, effective listeners related listening input to both personal experience and knowledge of the world. Main differences between effective and ineffective listeners were self-monitoring, elaboration and inferencing. Goh (2000) also investigated listeners’ comprehension problems, and analyzed listening problems according to the three comprehension phases developed by 8.

(21) Anderson (1995). The purpose of this study was mainly to probe into real-time problems that listeners faced during the process. Goh collected data from a group of EFL college students by having them keep diaries. Participants wrote down how they tried to understand the listening text and the problems they encountered. In addition to diaries, 17 students from the same group took semi-structured interviews to explain how much they knew about the listening task. Twenty-three students also took part in immediate retrospective verbalization to show their use of processing strategies. The study showed that among the ten real-time listening problems, half of them were. 政治大 ineffective attention. Three were problems with parsing and two were with utilization. 立 perceptual processing problems relating to unsuccessful word recognition and. The study helped us to recognize the detailed situation when ESL learners listened.. ‧ 國. 學. Language teachers can thus help listeners develop strategies that might be needed to. ‧. overcome listening difficulties.. sit. y. Nat. Both of the studies above seemed to agree with Thompson’s (1995) opinion that. io. er. aural medium has special effect on listening comprehension. Listeners must understand the text while listening, retain message in memory, coordinate listening. al. n. v i n Cand text with the prior knowledge check comprehension by sifting U h econtinually i h ngc incoming information. Listeners shoulder a heavy cognitive load in order to. successfully obtain the whole meaning of information. To further explore the factors that influence listening process, we need to examine the source of aural input, the spoken text. Features of Spoken Language Reading and listening both receive input; however, the features of their information carrier, the written text and the spoken text, are totally different. Buck (2001) indicated three main features of speech that are specifically crucial in building up listening comprehension. First is that the medium of sound might generate ambiguity 9.

(22) because of the indistinct acoustic input. During fast speech, speakers often unconsciously use phonological modifications such as assimilation, elision and intrusion. Listening to every word clearly might be difficult for listeners. Second, spoken language is real-time. Listeners need to grasp as much information as they could because the spoken text won’t be accessible in the next moment unless the speakers are required to repeat the text again. This feature might add difficulty to listening process because listeners can’t control the speech rate. The speed of talking is determined by speakers, and listeners can only use their limited short-term memory. 政治大 fours with what is heard. Third, speakers seldom use sentences to convey their 立. to store listening input. Therefore, what is remained in the memory couldn’t be on all. thoughts, they use idea units instead. Idea units are short phrases or clauses which. ‧ 國. 學. strung loosely together. Other than that, the vocabulary and the grammar used in. ‧. spoken text seem to be more colloquial and less formal. There are many expressions. sit. y. Nat. that are only presented in spoken text, not the written text. According to the above. io. er. three characteristics of spoken text, we might infer that listeners might hear acoustic input but don’t really understand what was said. However, listening input is not. al. n. v i n always presented as oral text; someC might be script as well. h e n g c h i UOral text and script are at two ends of a continuum. We need to be very careful not to attribute listening. difficulty solely to the features of spoken text. Other features, such as the speakers and the listeners, both determine whether the listening process is successful or not. Since the research aims to investigate listeners’ performance on different question types, and the classification of question types is related to the text variables, the present study would focus on the variables of listening text in the following literature review.. Variables Influencing Listening Comprehension 10.

(23) For many years, studies on listening were influenced by research of reading. Factors influencing listening comprehension were based on parallels in reading research. Hence, Rubin (1994) devoted much effort to sort out five major categories of variables which affect listening. In his review, these five factors include text characteristics, interlocutor characteristics, task characteristics, listener characteristics and process characteristics. Text characteristics depict variation in a listening text or associated visual support. Interlocutor characteristics focus on variation in speakers’ personal features. Task characteristics are related to the purpose for listening and. 政治大 Process characteristics are mainly about listeners’ cognitive activities and the 立. types of responses. Listener characteristics portray listeners’ personal attributes.. interaction between speakers and listeners. The purpose of the present study is to. ‧ 國. 學. explore the relationship between question types and strategy use in a listening. y. sit. io. Text Factors in Listening Comprehension. er. Nat. text in the listening comprehension process.. ‧. comprehension test. Therefore, the following passages would focus on the factors of. al. n. v i n Language teachers needCto choose appropriateUlistening text for listeners. hengchi. According to Brown and Yule (1983), listening comprehension process is influenced by the speaker, the listener, the content and the support. Thompson (1995) also made similar conclusion, i.e., to select a proper listening text, we need to consider the feature of oral extent, visual support, length, content familiarity, speech rate and pauses. Shang (2008) made some suggestions on how to enhance the effectiveness of teaching and learning of listening comprehension by analyzing in detail the nature of listening, listening process and listening skills. One of the methods to improve listening comprehension is by carefully selecting listening materials. Listening materials could be surveyed in different ways. First, there are different types of input, 11.

(24) such as English songs, everyday conversations, and announcement. Second, they could be presented with visual aids such as pictures and diagrams. Third, listening material should be chosen in accordance with listeners’ proficiency level. Fourth, other than vocabulary, if listeners acquire the knowledge of structure before listening, it is possible that they could use this knowledge to comprehend the target language more efficiently. In the present study, students with different ability need to receive the same test. Also, to make the study impartial, listening instruction is not allowed. Therefore, the following passage would focus on the elements of oral extent and visual aids. Oral Extent in Listening Text. 立. 政治大. ‧ 國. 學. Oral extent, according to Thompson (1995), is a continuum which includes. ‧. conversation at one end and lectures at the other. It is believed to influence listening. sit. y. Nat. comprehension. She also suggested that when the listening text is closer to the oral. io. er. end of continuum, the higher the listening comprehension test scores would be. The idea of oral extent could be enlarged to include text types by adding elements such as. al. n. v i n the interaction between listeners andCspeakers. Berne (1992) h e n g c h i U gave text type a general. definition, which is the format and genre to which they belong. She concluded that in Shohamy and Inbar’s study (1991), the degree of interaction between speaker and listener differentiated text types. News broadcast is at one end of the continuum and dialogue at the other. The more interaction between speaker and listener, the easier the text is. Overall, the most commonly seen classification in text type is by the number of speakers involved (Luo, 2004). Texts with only one speaker contain more explicit language and complicated sentence structure. They are presented in the forms of statements, narration, speech, lectures, news broadcast, instruction, and announcement, which are often termed as monologue. As for texts involving more than one speaker, 12.

(25) they contain more implicit information and oral features. Conversations, interviews, movies, dramas, debates and discussions all involve many speakers, and they are termed as dialogues. The influence of text type differed in several studies. In the study of Shohamy and Inbar (1991), text type was determined by its location along the oral and literate continuum. These researchers used three main text types. One is a pre-written, edited news broadcast. Another is a lecturette in which a speaker interacted with an addressee. The other is a consultative dialogue. The study found that subjects performed better in dialogue because it contained more pauses, false. 政治大 written feature had effects on listening comprehension. 立. starts and other oral features. It confirmed that whether the text contained the oral or. However, the results of Read (2002) were contrary to Shohamy and Inbar’s study.. ‧ 國. 學. He investigated the influence of applying different text types in EAP (English for. ‧. academic purpose) condition. First, he chose an appropriate topic and wrote a script.. sit. y. Nat. The monologue was spoken by one speaker with the written script, and the researcher. io. er. developed the test item by the written text. Then, three speakers read the written text and planned what to say in a discussion. The discussion was spontaneous without. al. n. v i n Cdiscussion script. The result showed that moreU difficult than monologue. The h e n gwas i h c. researcher concluded that the result may have something to do with the fast speech rate in discussion. Though the experiment had some flaws in test item, the results still indicated that the oral extent of listening text should be examined along with its speech rate and vocabulary in a listening comprehension test situation. Visual aids in Listening Text Listening in daily life usually occurs along with non-verbal information, such as gesture and facial expression of the speaker. Non-verbal information provides listeners with clues to comprehend the meaning. In the context of a listening test, one 13.

(26) of the possible ways to aid non-verbal information is by using visual supplements such as videos, pictures, graphs and tables. Many researchers have suggested that visual aids play a significant role in listening comprehension. As Early (1989) claimed, graphics help listeners to lower the language barrier and decrease short-term memory load by activating both content and rhetorical schemata simultaneously. When listening text can not be completely understood, listeners use visual aids to anticipate, predict, infer and confirm meaning. Some theorists believed that a well-organized picture can be representative in expressing speakers’ meaning, and it might be more. 政治大 deal with information in comprehension, storage and recall. Mayer (1997) addressed 立 effective than a thousand words. Paivio (1971) protested that an image help listeners. similar viewpoint. He contended that listeners are active constructors who select. ‧ 國. 學. crucial words and images from the listening text at the beginning of the listening. ‧. process. They then systematize these words and images and make them coherent. sit. y. Nat. mental representation. The last step is to coordinate the verbal and visual input and to. io. er. build up the meaning. It is noteworthy that the degree of using visual aids depends on the listeners’ cognitive ability and cognitive styles. Despite the positive effects. al. n. v i n C hshowing adverse U demonstrated, there were some research effect of visual aids on engchi. listening comprehension. Gildea et al. (1988) conducted an experiment to find the effects of various amounts of contextual enrichment for native language learning retention. Materials for the experiment were the movie ‘Raiders of the Lost Ark’ and 36 passages adapted from a text of the story, which were similar in length and complexity. They grouped sixty grade-school students into five conditions: the group who read the text only, the group receiving word definition on request, the group receiving definition and sentence using of words on request, the group receiving illustrative pictures from a movie and sentences on request and the last group receiving definition, illustrative pictures and sentences on request. The result showed 14.

(27) that more information did not proportionally guarantee increase in learning. Surprisingly, using pictures decreased vocabulary learning outcome. Thus, Gildea et al. contented that visual information may divert learners’ attention when the pictures presented were “novel” and “salient.” Though the study aims at discovering learning effectiveness on vocabulary, it provides some important viewpoints on using visual aids while learning. First, the amount of visual aids is not proportional to learning performance. It is worth discussing whether the presence of pictures plays an important role on learning. Second, while using visual aids, there are some factors to. 政治大. be considered. The feature of pictures may influence the effectiveness in comprehension.. 立. Arbib et al. (1987) insisted that some elements in a picture decide how helpful it. ‧ 國. 學. would be for learning. These elements include: the size of an image, its centrality in. ‧. the frame, variance from learners’ expectation, and how it portrays movement. What. sit. y. Nat. listeners hear need to correspond with what they see; otherwise, the beneficial effect. io. er. would be diminished. That is to say, whether a picture assists or distracts listeners rely on the presentation of a picture and how listeners use it.. al. n. v i n C hby the researchers U One of the factors noticed is the timing of presenting a picture. engchi. Brandsford and Johnson (1972) once conducted four experiments to examine the. effect of thematic organizers. Thematic Organizer is like a bridge, which connects the background knowledge of learners and the knowledge that needs to be learned. One of the thematic organizers in Brandsford and Johnson’s study was a picture relating to the context. The researchers put 50 high school students into five groups. Group one heard a passage once with no context. Group two heard a passage twice with no context. Group three read a contextual picture for 30 seconds before hearing the content. Group four looked at a contextual picture for 30 seconds after hearing the content. The last group viewed a picture of the objects mentioned in the content 15.

(28) before hearing the passage. The only difference between group three and group five is that for group five, the picture of objects had no meaningful relationship to one another, though the objects in the picture were relative to the content. The result showed that group three had the best performance among the five. The researcher found that contextual visuals help the most when they are presented at the beginning of a story. Therefore, when providing listeners pictures, it is better to consider the sequence of the listening activity. Omaggio’s (1979) study also agreed with the conclusion above. Another. 政治大 less contextual visuals than lower language learners did. The same viewpoint can be 立. interesting finding from the research is that the higher language learners tended to use. sustained by a review of Berne (1998). The researcher examined through the literature. ‧ 國. 學. related to L2 listening. One of the viewpoints is that the effectiveness of different. ‧. types of visual aids differs according to the ability of L2 listeners. From the schematic. sit. y. Nat. perspective, low-achievers might need more visual aids because of their insufficient. io. er. linguistic knowledge. Therefore, they rely more on the cues from the pictures, which help them to predict, perceive and integrate listening input. As for high-achievers,. al. n. v i n visual aids might be redundant, but C they are never hindrance. h e n g c h i U They can use linguistic cues to comprehend listening text and check whether their responses are correct. In addition to the level of listeners, difficulty of listening text might also be a factor influencing the use of visual aids. When listeners hear easy or simple input, they can use the cues from the language itself. On the contrary, when listeners need to pay attention to more difficult listening text, it is possible that they could not cope with the linguistic cues and need the help of visual images. This surmise is supported by Wolff (1987). Wolff carried out a study on the effects of visual images with German secondary ESL students. He suggested that an easier text allowed students to overlook the visual images because bottom-up process was not impeded. On the contrary, when 16.

(29) encountering more difficult text, they relied on visual support to build up the meaning. When seeing a picture, learners act differently in different levels of text difficulty. When they hear the text with and without visual aids, they use different cognitive process. Duquette and Pinchaud (1996) found that when listeners could see a picture and hear a passage simultaneously, they seemed to pay less attention to linguist cues. While hearing a passage without seeing a picture, they tended to pay more attention to linguist cues. Therefore, to facilitate comprehension, learners adjust their strategies. 政治大 research, study and theories and studies of strategy use in listening comprehension. 立 when facing different kinds of task. The following passages would present the. ‧ 國. 學 Issues on Listening Strategies. ‧. sit. y. Nat. The main role of English teaching has been shifted from teachers to students. io. er. recently. Boyle (1984) once examined factors that influenced listening performance from the perspective of both students and teachers in secondary and tertiary levels.. al. n. v i n C hfactors that students Surprisingly, there were several and teachers perceived engchi U. differently. For students, they conceived vocabulary an important elements in listening process more than teachers did. They thought that the more words one knows, the more possible one might understand listening texts. Three other factors mentioned by students but barely mentioned by teachers were memory, attention and the influence of reading habit. The teachers, on the other hand, focused much more on linguistic elements, such as the ability to cope with stressed syllables, intonation and complex syntactic structures. The study suggested that teachers and students had different concepts in listening strategy use. Therefore, teachers need to examine students’ listening process by constantly checking their strategy use. Only by doing so 17.

(30) can teachers know what strategies students need and how to help them. Strategy Use under Listening Test-taking Condition Learning strategies might help learners acquire language knowledge in langue learning situations. However, in langue test situations, strategy use is related to the ongoing working memory to solve task problems (Gagne et al., 1993). When inspecting variation in language test performance, the strategies used during tests could not be neglected. The concept of test-taking strategies was first started with the. 政治大 formats and/or test-taking situations to raise test scores (Ritter & Idol-Maestas, 1986, 立 word “test-wiseness.” It depicts “one’s capacity for using test characteristics and. p.350). What test takers do to solve test problems is worth discussing. Phakiti (2003). ‧ 國. 學. examined test-taking strategies on multiple-choice reading comprehension tests.. ‧. Subjects of the research are 384 Thai university students enrolled in an English course.. sit. y. Nat. They took an 85-item achievement test and finished a cognitive-metacognitive. io. er. strategies questionnaire. Eight of the subjects were further interviewed to describe their thinking process. The result showed that the use of cognitive and metacognitive. al. n. v i n Cthe strategies had a positive influence on Also, high achievers U h etestnperformance. i h gc. tended to use more metacognitve strategy than low achievers. Oxford et al. (2004) used tests to examine test-takers’ reading strategy use. In the study, it was found that reading strategies were not affected by task difficulty or proficiency alone. However, when examining specific items, high achievers used more top-down strategies while low achievers tended to use more bottom-up strategies. Most of the research before mainly focused on the reading task-based situation; studies on listening task-based situation are comparatively rare. Vogely (1995) investigated how English learners of Spanish used and perceived listening comprehension strategies on an authentic task. In this study, subjects tended to use the 18.

(31) following strategies. By the order of importance, they were getting the overall meaning of the text, relating the background knowledge with the information in the text, understanding the meaning of each word, focusing details, and mentally sounding out the words and phrases. Interestingly, the strategies often used above don’t correspond with strategies subjects perceived. For example, in the questionnaire, subjects thought that understanding the gist of the text is important, but they didn’t report using it often in their strategy use questionnaire. That is to say, listeners might know what strategies could be helpful for comprehension, but in the listening task,. 政治大 help listeners compensate for the gap between “knowing what” and “knowing how”, 立 they do not know how and when to use these strategies successfully. Therefore, to. is important for facilitating listening comprehension.. ‧ 國. 學. Taxonomy of Listening Comprehension Strategies. ‧. sit. y. Nat. Studies related to strategy use were first focused on those for language learning. io. er. purposes. According to Oxford (1992), learning strategies are consciously activated by FL/L2 learners to improve their language skills. O’Malley and Chamot (1990). al. n. v i n C strategies categorized language learning threeU types: cognitive, metacognitive, h e n ginto i h c. and social/affective strategies. Language learners use cognitive strategies to cope with the incoming information. Listening strategies under the scope of cognitive strategies are rehearsal, organization, inferencing, summarizing, deduction, imagery, transfer, and elaboration. As for metacognitive strategies, they help language learners to monitor, plan and evaluate the outcome of activities, and they need higher mental process. Metacognitive strategies in listening process include selective attention, planning, monitoring, reviewing, and checking comprehension. Social/affective strategies are interaction among people and the affective control, which facilitate learning outcome. In the listening process, social/affective strategies include 19.

(32) cooperation, questioning for clarification, and self-talk. In O’Malley and Chomat’s (1990) viewpoint, there are three main benefits of applying language learning strategies. First, learners don’t have to rely on others; they can be responsible for their own learning. Second, learners can build up schemata, which also help other kinds of learning. Third, strategies are easy to change or teach compared with other personal characteristics. It is important to instruct strategy in language classroom because it is teachable and flexible compared to other elements that influence performance (Rost, 2002). Generally speaking, listening strategies are. 政治大 listening strategies. He modified some of O’Malley and Chamot’s (1990) 立. like learning strategies. Vandergrift (1997) specifically focused on the types of. categorization of learning strategies and elaborated detailed subcategories under three. ‧ 國. 學. categories, which are also metacognitive, cognitive, and social-affective. ‧. strategies.The detailed classification of listening strategies by Vandergrift (1997) are. y. sit. io. er. Table 2.1. Nat. presented below in Table 2.1.. Vandergrift’s(1997) Classification of Listening Comprehension Strategies. al. n. 1. Metacognitive Strategies. Ch. i n U. (1) Planning a. Advance organization b. Directed attention c. Selective attention d. Self-management. engchi. v. (2) Monitoring a. Comprehension monitoring b. Auditory monitoring c. Double-check monitoring (3) Evaluation a. Performance evaluation b. Strategy evaluation (4) Problem identification. 20.

(33) 2. Cognitive Strategies. (1) Inferencing a. Linguistic inferencing b. Voice and paralinguistic inferencing c. Kinesic inferencing d. Extralinguistic inferencing e. Between parts inferencing (2) Elaboration a. Personal elaboration b. World elaboration c. Academic elaboration d. Questioning elaboration e. Creative elaboration f. Imagery. 治政 (3) Summarization大. 立. (5) Transfer (7) Resourcing (9) Note-taking (11) Substitution. 3. Socio/affective Strategies. ‧. ‧ 國. 學. (4) Translation (6) Repetition (8) Grouping (10) Deduction/induction. n. al. er. io. sit. y. Nat. (1) Questioning for clarification (2) Cooperation (3) Lowering anxiety (4) Self-encouragement (5) Taking emotional temperature. Ch. engchi. i n U. v. Related Studies of Listening Comprehension Test and Listening Strategies The following research also aimed to explore the relationship between GEPT test performance and strategy use. All of them chose subjects in Taiwan, and the subjects might share similar background knowledge and learning experience with the subjects in the present study. Therefore, the research is further discussed in the following passages.. 21.

(34) The Relationship Between Strategy Use and GEPT Test Performance Wu's (2008) study investigated the relationships between strategy use and GEPT test performance. Subjects are 567 test takers of Taiwanese EFL learners. The age of the test takers ranged from 14 to 19, and the average age was 17.09. These test takers took a GEPT-Intermediate practice test two weeks before they participated in a formal GEPT-Intermediate test. The study only examined the part of reading and listening and excluded the part of speaking and writing. Before the subjects took the practice. 政治大 based on Purpura’s (1997) work. It contained 34 items of cognitive strategy use and 立 test, they were asked to fill out the strategy use questionnaire which was developed. 30 items of metacognitive strategy use. The questionnaire used a 5-point Likert scale. ‧ 國. 學. and was translated into Chinese. A pilot study was executed among a small sample. sit. y. Nat. clearly.. ‧. before the practice test to check whether test takers could understand the statements. io. er. The researcher concluded two main points based on the result of the study. First, learners used more metacognitive strategies than cognitive strategies. It is believe that. al. n. v i n the phenomenon might be deduced C from the second language h e n g c h i U testing situation. While taking language tests, learners are usually inclined to language performance rather than on language learning. Since performance is product-oriented and learning is process-oriented, it seems logical that learners employ more metacognitive strategies because they are relative to post-assessment process such as self-evaluation and self-testing. Second, when considering strategy use as predictors for GEPT performance, assessing the situation and monitoring were two strategies which were identified repeatedly in three different test results: sum score, reading score, and listening score. Monitoring had a positive effect while assessing the situation had negative consequence on test performance. It is noteworthy that the findings are 22.

(35) consistent with Paris’s (2002) assertion that strategies can have positive, negative, and neutral consequences. One of the possible explanations of negative influence on the study might be the statements under assessing the situation in the questionnaire were often used by low achievers. The statements often describe thinking process before or during a language performance. For example, Item 36 says “Before I begin an English test, I try to see which parts will be easy and which parts will be difficult”. Item 45 tells that “Before I begin and English test, I think about how the test will be scored”. For low achievers, they may have little confidence in their language competence.. 政治大 performance. For better learners, on the other hand, they might feel confident for the 立 Therefore, they need to estimate the test situation and try to optimize their. test and thus could focus on the test items single-mindedly.. ‧ 國. 學. Effects of Test Format and Text Type on Listening Comprehension and Strategy Use. ‧. sit. y. Nat. The study of Luo (2004) aimed at discovering the effects of test format (whether. io. er. questions are presented with pictures) and text type (the oral extent of the listening passage) on listening comprehension and strategy use. There were 106 third-year. al. n. v i n C hwere divided intoUhigh, medium, and low proficiency senior high school students who engchi groups by their performance on a listening comprehension test. Then, 32 students. from high proficiency group and 33 from low proficiency group participated in the study. After taking a listening test from GEPT, students were asked to fill in the questionnaire of strategy use in their real-time situation. The questionnaire contained 25 items and was based on Oxford’s (1990) classification of strategy. The findings of the study are presented as below. First, all subjects performed better on test item with pictures than without pictures. The researchers concluded two possible reasons of the phenomena. One is that test with pictures avoid testing other language skills, such as reading. In test items without pictures, test-takers need to read 23.

(36) questions and rely on cues from the passage and thus might underestimate test-takers’ listening ability. Another reason is that pictures provide contextual cues, which can make up for incomplete understanding of the input. Second, the study found that students have preference in using strategies. For high proficiency learners, they used compensation, metacognitve, and social strategies more often than low proficiency learners. The researcher suggested that compensation and metacognitive strategies require higher level of mental processing, and thus can be served to distinguish high and low proficiency learners. Third, subject tended to use metacognitive and affective. 政治大 subjects’ test performance in the test item without pictures. In conclusion, test format 立. strategies in the test item with pictures. Metacognitvie strategies were also related to. and text type had significant effect on listening performance and strategy use. And. ‧ 國. 學. metacognitve strategies were found to enhance subjects’ overall performance. The. sit. y. Nat. instructing listening comprehension in teaching.. ‧. researcher suggested language teachers to be aware of these elements when. er. io. Effect of Text Types on College Students' Listening Strategy Choice. al. n. v i n C h the effect of text Su’s (2007) study intended to examine type on listening strategy engchi U. choice while taking listening comprehension test. The subjects were 360 college. students form the universities in central Taiwan. They took a listening comprehension test of GEPT intermediate level. There are three types of text in the test: picture, dialogue and conversation. After taking each part of question, they were asked to fill in a questionnaire. The questionnaire was designed to understand strategy use in different kind of text types. In other words, they needed to fill the same questionnaire three times. The questionnaire was based on Vandergrift’s (1997) classification of listening strategy and contained 34 items. 33 were 5-point Likert scale and one was an open question. After finishing the test and questionnaire, 30 subjects from each high, 24.

(37) intermediate and low proficiency level received an interview for exploring their in-depth use of strategy during the test. The research concluded three points by the study. First, the result showed that EFL learners in Taiwan were accustomed to use certain strategies in different kind of text types though these strategies might not be facilitative or appropriate. Second, the EFL listeners tended to use certain listening strategies with different frequency in different text types. One possible explanation is that listeners perceived that they need to adjust their strategy use according to text types, so they changed their frequency of. 政治大 the strategy use. However, there was no interaction effect between proficiency level 立. using strategy. Third, both proficiency level and text type had significant influence on. and text type.. ‧ 國. 學. The limitation of the study is that it didn’t take subjects’ major into. ‧. consideration. Whether subjects were from English department or not might change. sit. y. Nat. the result of the study to some extent. The background of subjects could be more. io. er. similar to one another. The other part that needed to be improved is the way of applying interview. It was carried out after the listening test. Therefore, some thinking. al. n. v i n process of the listeners couldCnot be reflected rightU h e n g c h i away. To have more holistic picture of how listeners use strategies, think-aloud method might be helpful. Concluding Remarks of Chapter Two. The researcher has reviewed the issue on listening text, listening strategies and relative studies. It was found that several factors and some strategies were contributed to listening performance. However, most of the studies focused on listeners of higher level who are from senior high school or college. Few or even no studies have been conducted to examine listening strategy use of junior high school students. Secondly, the instrument of listening comprehension text varied, some were from authentic 25.

(38) audio tape, and some were from standardized test. Even studies which used GEPT listening test often analyzed listening text in dichotomy, either text with picture and non-picture or text of monologue and dialogue. However, the GEPT listening test has been modified from three question types to four since 2010. Therefore, more work on the issue is needed.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 26. i n U. v.

(39) Chapter Three Methodology. The primary aim of the present study is to explore listening performance and strategy use in different question types. The study used GEPT elementary level. 政治大 strategy use questionnaire. 立The questionnaires for test-takers were designed to collect listening comprehension test. After taking test on each question type, test-takers took. ‧ 國. 學. necessary data for further analysis. This chapter comprises four sections. The first section describes the background information of the participants. The second section. ‧. discusses the instruments used in the study. Procedure of the study is briefly. sit. n. al. Ch. Participants. engchi. er. io. the last section.. y. Nat. introduced in the third section. Techniques used for analyzing data are presented in. i n U. v. In order to investigate the effects of question types on listening strategy in junior high school, participants were selected from one junior high school in Yilan. The school consists of eighteen classes in total; each class includes approximately 30 students.177 students from six classes of eighth grade were invited to participate in the study. The participants have received formal English class for at least five years. When entering junior high school, they took entrance examination implemented by Yilan County Government Education Department. The entrance examination aimed to allocate students equally to each class by their achievement. Therefore, students in each class were equally distributed in terms of their academic ability. 27.

(40) Instruments There were two instruments used in the study, a listening comprehension test from GEPT elementary level and an adapted questionnaire based on Vandergrift’s (1997) classification of listening strategy. GEPT The listening comprehension test was chosen from GEPT (General English Proficiency Test) elementary level. The test was published in 2010 by the Language Training & Testing Center. GEPT is a standardized test and the elementary level aims. 政治大 native speakers of America and the speed of talking is slower than normal speech for 立 to test the language proficiency level of junior high school students. It is recorded by. the purpose of testing. There is not any distinctive accent or fast speech rate. The. ‧ 國. 學. GEPT elementary level consists four parts, picture description, answering questions,. y. sit. io. The Structure of the GEPT Listening Comprehension Test. al. n. Section. Ch. er. Table 3.1. Nat. comprehension test.. ‧. conversations and short talks. Table 3.1 shows the structure of the listening. i n U. v. Presentation of questions. Number of speakers. pictorial. one. 5. 2. Answering Questions. Non-picture. one. 10. 3. Conversations. Non-picture. two. 10. pictorial. one. 5. 1. Picture Description. 4. Short Talks. engchi. Number of questions. Taking the test of section one (picture description) needs about three minutes for all five questions, each question takes about 30 seconds. There is a five-second pause between questions in picture description. In section two (answering questions), every question lasts about 20 seconds, and is followed by a 15-second pause for test-takers 28.

(41) to respond. As for section three (conversations), time for each question is around 40 seconds, the waiting time for answering question is 15 seconds. The question in section four (short talks) needs about 40 seconds, and before the statement, questions that needed to be answer is spoken by a narrator. The pause for test-takers to answer question takes 5 seconds. The subjects took all of the four sections and the test took about twenty minutes to answer all the listening questions. Rubin (1994) reviewed Weinstein-shir and Griffiths’ (1992) study about the effect of speech rate in listening comprehension. According to the results from the. 政治大 presented at the speed of 127 w.p.m (word per minute). Table 3.2 shows the duration 立 study, low-intermediate level listeners performed the best when the listening text is. and speed of four sections in GEPT listening test. The text is example questions from. ‧ 國. 學. a GEPT elementary listening test. The result shows that the speech rate of GEPT. ‧. listening test isn’t too difficult for junior high school students. Though the speed in. sit. y. Nat. the section of answering questions seems faster than others, it did not trouble students. io. er. while taking test. The reason is due to its simple question stem. There is only one speaker asking one question at one time. And the pause between questions is long. n. al. Ch. enough for students to respond. Table 3-2. engchi. i n U. v. Speech Rate of Four Sections in GEPT Elementary Level Listening Test Section. Number of words. Duration. Speed (w.p.m). in each question One. 16. 0’33. 123.75. Two. 12. 0’27. 135. Three. 39. 0’19. 123.15. Four. 42. 0’20. 126. 29.

(42) Questionnaire Researches on listening strategy choice used various methods. Some use think-aloud procedure (Bacon, 1992), others use listening journals (Katchen, 1996) and still others use interview (Bacon 1992 ; Su, 2007). Considering the level of junior high school, it might be difficult for them to articulate strategy use precisely. Therefore, the study explored junior high school students’ listening strategy use by questionnaire. The questionnaire is adapted from Su’s (2007) study, which was based. 政治大 subcategory of cognitive strategies, association, was not included in the questionnaire. 立 on the classification of Vandergrift’s (1997) classification of listening strategy. One. The original description of association in Su's (2007) study was “I associate the word. ‧ 國. 學. I hear with the word I have learned." Because students who joined the pilot study. ‧. found the description confusing, they did not know what kind of association might. sit. y. Nat. occur in the listening process. Students could only judge whether the word is familiar. io. er. for them in limited time for response, they were unable to associate words that they have learned before. Therefore, the subcategory of association in cognitive strategies. al. n. v i n C hwere two parts in the was deleted in the present study. There questionnaire; one is engchi U. about the subjects’ background, such as their gender, years of learning English. The other part is regarding the listening strategy they use. There were 24 items in the five-point Likert scale, ranging from “always” to “never”. One open-ended question will be used to examine the subjects’ other strategies (See Appendix A). The subjects filled the questionnaire after they listened to each question type. Table 3.3 shows the categories of listening strategy covered in the questionnaire and their correspondent items. The description for each item is presented in Table 3.4.. 30.

(43) Table 3.3 The Classification of the Strategy Items in the Listening Comprehension Strategy Questionnaire Strategy. Item. 1. Metacognitive strategies a. Planning Advance organizers Directed attention Selective attention Self-management b. Monitoring c. Self-evaluation. 政治大. ‧ 國. ‧. n. engchi. sit er. io. Ch. y. Nat. al. 9. 8. 19. 12. 22. 13. 15. 2. 1. 3.. 學. 2. Cognitive strategies a. Repetition b. Grouping c. Imagery d. Auditory Representation e. Elaboration f. Inferencing g. Note-taking h. Summarizing i. Translation j. Association k. Bottom-up l. Top-down m. Listening-specific. 立. 14. 18. 6. 10. 4. 21. 5. 7. 17. 24.. i n U. v. 16. 20. 23. 11.. Table 3.4 Description of Items Item 1 2 3 4 5. Descriptions I make a mental or written summary when I listen to the text. I take notes when I listen. (Ex. Jotting down key words). I translate key words or sentences into Chinese. I adjust myself to catch up the speed. I check if I really understand the meaning and never complete the test by guessing. 31.

(44) 立. 政治大. 學. ‧. 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24. I won’t give up even if I don’t understand thoroughly. Instead, I focus on the following text. I try to concentrate when I find myself being distracted. I group words I heard according to their features. I repeat key words in my mind once or twice. I read possible distracters before listening. I listen to chunks of words once. I used pause and tone to speculate words that I don’t know. I relate what I heard with my experience and background knowledge. I decide what to focus on before listening. I predict the following text and outcome when seeing questions. I focus on the details of the content. I evaluate how much I have known about the topic. I utilize the description of questions to help comprehension. I have mental pictures when listening to the texts. I piece up the gist with detailed information. I ask myself to think in English when listening. I use intonation to segment words and phrases. I focus on the gist of the contents. I evaluate how much I have understood the contents.. ‧ 國. 6. sit. y. Nat. n. al. er. io. Almost each item corresponds with one subcategory of listening strategy. The. i n U. v. classification of Vandergrift’s (1997) listening strategies included three parts:. Ch. engchi. metacognitive, cognitive and social/affective strategies, social/affective strategies were crossed out in the questionnaire. Since the present study aims to investigate the strategy use under the listening comprehension test, it is not possible for students to ask help from the peer while testing. Therefore, the listening strategies under social/affective category were not used in the questionnaire.. Procedure The research procedure includes three parts: modifying questionnaire, implementing formal test, and grouping students. 32.

(45) The questionnaire is first adapted from Su’s (2007) study. She investigated the relationship between listening strategy choice and text type. Participants of the study were college students from central Taiwan; therefore the wording of the questionnaire might be too difficult for junior high school students. The researcher had five students answer the questionnaire to check whether it is clear and friendly for junior high school students. And the questionnaire was modified by students’ responses. For example, in the description of note-taking, Su (2007) wrote that "I take notes while listening." But students who joined the pilot study reflected that it is not possible for. 政治大 whole sentence for them. Therefore, the researcher gave an example of taking notes in 立 them to take notes in such a short time because taking notes means writing down a. the end of the description, i.e., jotting down key words.. ‧ 國. 學. The researcher had students fulfill the task in two different periods. In the first. ‧. period, students listened to the explanation of the listening test and listening strategy. sit. y. Nat. questionnaire. The researcher read through items on the questionnaire and checked. io. er. whether students understood the words. Then, students answered questions in the section of picture description and answering questions. At the end of each section, the. al. n. v i n C h and had studentsUfill in the questionnaire. Another researcher stopped the CD player engchi two sections, conversations and short talks were given to students two weeks away. from the first class. The reason to split the procedure into two periods is because that the subjects needed to answer the same questionnaire four times after taking GEPT listening test. The GEPT listening test took about twenty minutes, and answering 24 items on the questionnaire carefully for four times took about forty minutes. For fear that subjects might not have abundant time to fulfill listening test and questionnaire in one period, and having students fill in the same questionnaire for four times in two consecutive periods might cause fatigue or practice effect, the researcher decided to have an interval of two weeks to reduce possible influence of practice effect. 33.

(46) In the formal test, the researcher introduced the procedure and purpose of the study, and students were told that the result wouldn’t influence their academic scores. Students were asked to spend three to five minutes filling personal information in the first part of questionnaire. Then they started the listening test. The content of listening text was played only once. After answering questions about the listening text, the researcher stopped the CD for a moment. The 24 items on the questionnaire were read through by the researcher for making sure that every student understood the meaning of the items. During the time of filling questionnaire, students were allowed to ask. 政治大 subject had finished every question of the first part, and started playing the second 立. any questions about the questionnaire. Then, the researcher made sure whether every. part of listing test. In the second part, the subjects also answered the listening. ‧ 國. 學. comprehension question first and fill the questionnaire later. It took about forty-five. ‧. minutes for the whole procedure. The whole procedure was presented in figure 3.1.. y. Nat. n. er. io. al. sit. Figure 3.1 A Flowchart of the Research Procedures. Adapting Questionnaire. Ch. engchi. i n U. v. Used Su’s (2007) questionnaire and had five students answer.. Checked whether students were clear about the description.. Decreased number of items and elaborated descriptions.. 34.