Chapter 1 Introduction
In this introductory chapter, the rationale is first presented, followed by the background and motivation. Definition of terms is particularly specified after background. Then research questions and hypotheses are subsequently posed to clarify the purpose of the study. After research questions and hypotheses, the significance of the study is revealed. Finally, organization of the study and the summary of the first chapter are expounded.
Rationale
With the growing interest in and adoption of communicative language teaching, most English foreign language teachers have realized that students should be taught to speak English with some fluency and with some communicative competence.
However, communicative ability is often equated with speaking proficiency and it is often neglected that communication is a two-way process involving at least two people: a speaker and a listener. Speaking does not constitute communication itself unless what is said is comprehended by another person. In English language teaching, listening comprehension has been a somewhat superficially treated skill, which is affirmed and reiterated by many scholars and educators. Chastain (1976), Paulston and Bruder (1976), and Glisan (1985) all consider that listening comprehension is probably the most neglected and the least stressed skill in second language teaching.
The problem is even more serious in situations where English is taught as a foreign language as in Taiwan (Yen, 1987). Less training in listening directly results in students’ less specialization in listening proficiency. According to Lai’s (1984) study,
she reports that approximately 40 percent of her subjects admits listening comprehension is their worst among the four language skills. This phenomenon can be approached from three perspectives.
First of all, the fact that Taiwanese students are deficient in their listening ability might be related to their way to learn English. Despite the popularity of computer equipment and video materials, written texts are, comparatively speaking, highly accessible and cheap sources of foreign language materials (Zhang, 2003). So to speak, written language in hard copy is still one of the major sources of contacting with foreign language. Another factor contributing to Taiwanese students’ listening deficiency is that most high school teachers in Taiwan take the pragmatic line of the college entrance examination, which only tests reading and writing skills. Blindly following the trend of the entrance examination, teachers give more weight to the development of students’ reading and writing skills while relatively leave their speaking and listening stagnant. Unless a listening section is included in public examinations, listening training will continue to receive little attention (Yen, 1987).
The third corresponding factor might be that listening itself is a demanding task, which involves a very complex processing. Richards (1983) proposes that listening process requires listeners to determine the type of speech event, recall scripts relevant to the situation, make inferences about the goals of the speaker, determine the propositional meaning of the utterance, assign an illocutionary meaning to the messages, and remember and act upon the information. It seems that listeners have to handle the above steps simultaneously even with the fleetingness of the speech. There is no wonder why Taiwanese students are confronted with great difficulty in listening.
Although Taiwanese students are deficient in listening ability, the importance of listening should not be underestimated. Based on Rankin’s (1978) study, adults spend roughly 45% of their time on listening, as compared with 30% to speaking, 16% to
reading, and a mere 9% to writing. That’s to say that listening is most frequently used skill. In addition, listening skills also serve as the basis for the development of speaking skills (K. Chastain, 1976). Oral communication is impossible without a listening skill. Moreover, listening comprehension is the keystone of many theories of second language acquisition (Asher, 1969; James, 1984; Krashen, Terrell, Ehrman, and Herzog, 1984; Winitz, 1978; Wipf, 1984). The underlying assumption is that language acquisition is an implicit process in which linguistic rules are internalized by extensive exposure to authentic texts (Postobsky, 1974; Winitz, 1978) and by comprehensible input (Krashen, 1980). What these linguists assert is that listening will enhance students’ comprehension and enhanced comprehension will in turn promote acquisition (Long, 1985). With the above reason, listening comprehension ought to be regarded at least as important as any of the other language skills.
As scholars grow to understand the significant role played by listening in language learning and communication, they recognize more and more importance of teaching listening comprehension in the foreign language classroom. This recognition has resulted in an increase in the approaches to enhance students’ listening comprehension. These approaches encompass listening strategy training (Bacon, 1992;
O’Malley, Chamot, and Kupper, 1989; Rost and Ross, 1991; Rubin, 1990; Vogely, 1995), the use of visual aids (Mueller, 1980; Chaudron, 1985; Blau, 1990; Chiang and Dunkel, 1992), and incorporation of previewing questions (Herron, 1994; Berne, 1995). While all of strategy training, visual aids and pre-listening activities prove their effectiveness in fostering students’ comprehension, the field of previewing questions, relatively speaking, is less highlighted. However, in the instructional settings, the use of previewing questions which takes little preparation time and class time is most accessible in comparison with the strategy training and visual aids. Given the fact that previewing question activity is effective in promoting listening comprehension and is
easiest to implement in the classroom learning, the present study focuses on the effect of previewing questions.
Background and Motivation
No one can deny that listening is not an easy matter, in which many Taiwanese students are confronted with great difficulty. The difficulty can possibly be attributed to the features of oral utterances. Oral utterances are usually regarded as context dependent, informal, unplanned and unedited (Chafe, 1982, 1985; Ur, 1984). The comprehension of the oral text may be impeded by elements inherent in orally produced messages, such as speaker’s pronunciation, stress, intonation and rate of delivery (Dunkel, 1991; Richards, 1983; Ur, 1984). Besides, listening also requires simultaneous interpretation since the listener is not allowed to review the presented information in most situations. Hence, the listener must rely on immediate comprehension and on the ability to retain information in memory for further analysis, which shows listening is an intricate and thorny task (Shohamy & Inbar, 1991). With more and more research directed at the listening process, researchers are equipped with more and more understanding of unique characteristics of the listening skills.
Nevertheless, the relevant research concerning the issue such as the effect of previewing activities on listening comprehension is still not widely researched.
Although several studies pioneered to tackle the issue of the previewing activity around two decades ago, a majority of them was aimed at testing the effectiveness of pre-reading activities on the reading comprehension (Hudson, 1982; Stahl, 1983;
Taglieber, Johnson, and Yarbrough, 1988). Based on their findings, pre-reading activities based on providing pictorial context and on pre-questioning facilitate reading comprehension performance. However, due to the differences between reading and listening comprehension that have been emerging in the research
(Bernhardt and James, 1987; Lund, 1991; Mecartty, 1993), it is difficult to make empirically-based recommendations regarding the use of pre-listening activities.
People scarcely pay attention to the pre-listening activities, particularly the prequestioning one. Prequestioning is seldom noticed until Underwood (1989) reports that multiple-choice questions might interfere with the test-takers “while” listening.
Underwood’s (1989) comments induce an issue of the placing of questions in listening comprehension tests. The issue whether placing questions before, while or after listening spurs researchers’ interests. But among these choices, placing questions prior to listening earns best acknowledgement. Weir (1993) suggests that question preview may both alter the nature of the listening process and affect the results of tests. This view is also partly agreed by Buck (1991b) who first hypothesizes that the questions to be previewed rather than be reserved after listening provides a purpose for the listeners. As Brown and Yule (1983) argue, listeners generally listen for a purpose, and this purpose affects both the level of their motivation and the listening strategies they choose. Since both of these directly influence the degree of comprehension, Brown and Yule (1983) suggest that tests of communicative listening ability ought to provide a listening purpose. If listeners are granted the questions prior to listening, the questions would help settle their listening purpose. Purpose would further influence the motivation. Their motivation, in this way, will be adapted to acquiring the information necessary to the test questions. As long as purpose and motivation are all set, listeners’ strategy use would be determined as well (Buck, 1991b). This further implies that the cases where listeners are given the questions before listening and where those are granted the questions after listening would yield diverse outcome.
Contrary to the view that previewing questions are helpful, some researchers reveal that previewing questions do not necessarily result in the positive influence as has been claimed. In Buck’s (1991b) study, he surprisingly finds out that reading questions before listening does not always influence listening strategies. He further explains that perhaps in some cases listeners merely scan the questions without pondering what they are asking for, or maybe listeners seek the answers in response to questions on some occasions but not on others. Similar to Buck’s (1991b) finding, Sherman (1997) demonstrates previewing questions do not yield the satisfactory results as expected, which can be attributed to the two factors. First of all, previewing questions may not assist top-down processing. It has to do with the fact that parameters of listening comprehension in effect shifts from top-down to bottom-up processing in accordance with lexical familiarity, distribution of attention, and known and new information, which are bound to vary individually and affect the nature of the processing (Boyle, 1984; Garrod, 1986; Hayashi, 1991; Kelly, 1991; Lund, 1991).
Buck (1991a; 1991b) also affirms that “getting the right answer” may be a different process for every combination of listener, text and questions---in fact that every person answers a different question. Since so, the previewing questions are unable to guarantee to aid listeners’ top-down processing. In addition, previewing questions from Sherman’s (1997) perspective are likely to divert listeners’ attention. Bartlett (1932) makes the general point that “the character of what is perceived is lost the moment we seek to analyze it into partial percepts.” This corollary in listening might be that previewing questions affiliating single propositions are likely to distract attention from global meaning.
In consequence, Sherman (1997) in her study recommends another procedure, the Sandwich version, which distributes participants the questions between two listenings. Sandwich yields the best outcome compared with other groups such as
Questions Before, Questions After, and Free Recall. Free Recall, which turns out to be a quite problematic test method in her study, is left without discussion here. For Sandwich version, it not only shows the most satisfactory result, but also wins the supportive evaluation in the follow-up questionnaire. However, Shih (2004) points out that both the wording description and the scoring criteria in the questionnaire are problematic, which in turn are likely to affect the real effect of testees’ perceptions of different procedures. Besides the errors in the questionnaire, Sherman (1997) does not probe into the relationship of the testing procedures and test-takers’ proficiency level to see which procedure is particularly suitable for which level of participants. Also, Sherman does not explore the issue of question types. She does not group her questions into different types such as global, local and inferential ones to see whether previewing is more meaningful for certain types of questions. Since Sherman (1997) mistakenly involves an unnecessary Free Recall group, employs the ambiguous wording description, incorporates the false scoring criteria, and fails to highlight the relationship of testing procedures and proficiency levels and relationship of testing procedures and question types as issues, the present study is thus motivated.
In order to fill this gap, the present study aims to investigate the three test procedures on listening comprehension, namely Questions Before, Questions After, and Sandwich in respect. Besides this, the present study is interested in inspecting the relationship between the test procedures and proficiency levels and that between test procedures and question types. It is expected that the result of the study can clarify the issue and provide some useful insights for the classroom teachers.
Definition of Terms
Test Procedures
Questions Before: It is the procedure that allows students to preview listening
comprehension questions prior to the two hearings of the text. The procedure that grants questions in advance is Questions Before.
Questions After: It is the procedure that gives students the listening comprehension
questions after two hearings of the text. The procedure which does not give questions prior to but after listening is Questions After.
Sandwich: It is the procedure that allows students to read the listening comprehension
questions after one hearing of the text but before the other. The procedure which metes out questions between two hearings is Sandwich Method.
Question Types
Global Questions: These questions require the test taker to obtain the gist of the text,
and synthesize information. The answers to global questions can be determined directly from the listening text.
Local Questions: These questions require the test taker to locate details, understand
single words, such as name, time, and location. The answers to local questions can be heard from the listening text.
Inferential Questions: These questions demand the test taker to make inference of
the stated facts. The answers to inferential questions cannot be elicited directly from the listening text without thinking. Listeners have to make extra associations with the stated facts.
Research Questions
On the basis of the observations above, four main research questions are posed to investigate the effects of listening test procedures on the listening comprehension of
Taiwanese college students, and the respective relationship of testing procedures with proficiency levels and that with question types.
1. How do Questions Before, Questions After, and Sandwich differ in affecting participants’ listening comprehension?
2. How do participants of high, intermediate, and low proficiency level diversify in the performance of the listening test with different procedures?
3. How do the effects of global questions, local questions, and inferential questions differ with different procedures?
4. What do participants feel and think about Questions Before, Questions After, and Sandwich?
Hypothesis of Research Questions
Based on four research questions, four hypotheses are formed in response to each question above.
Hypothesis 1:
Sohamy and Inbar (1991) suggest that previewing questions provide a purpose for the listeners and assist top-down processing. It is thus predicted that participants treated with Questions Before and Sandwich (previewing before the second hearing) would exceed participants who have Questions After in their listening comprehension.
Nevertheless, Sherman (1997) argues that Sandwich version, unlike Questions Before which might divert participants’ attention right in the beginning, is the best-performing one in her result. She further explains that Sandwich which determines a purpose but does not distract test-takers’ attention compensates the shortcomings of Questions After and Questions Before. Since so, it is hypothesized that Sandwich version in the present study would be the best-performing group.
Hypothesis 2:
Although Sherman (1997) does not touch upon the issue of the proficiency levels, Mueller (1980) once divided his participants into high-level and low-level while treating them with Visual-Before, Visual-After, and No-Visual. Based on his result, the low-level group with Visual-Before has grades significantly higher than Visual-After and No-Visual. But for the high-level group, it does not show any significant difference from one treatment to another. In sum, the lower level the learners are, the greater indispensability they need the previewing of visual organizers.
Similarly, in the current study, it is hypothesized that Questions Before and Sandwich would be relatively more meaningful for the low-proficiency group. But for high-proficient participants, their performance would not be significantly affected by the test procedures.
Hypothesis 3:
Sohamy and Inbar (1991) examine the relationship between the question types and text genres. They deliberately design their questions as global, local and trivial questions. According to their observation, although global questions are able to enhance participants’ top-down processing, they are relatively more difficult to handle compared with the local and trivial questions. Participants who answer global questions right seldom do it wrong in the local questions. But the reverse situation does not exist. Thereby, it is predicted that previewing global questions would be more meaningful than local questions. However, since the essence of local and trivial questions is similar, trivial questions are left without discussion in the present study.
But the present study incorporates inferential questions which are highly recommended by Nunan (1999; 2002). Due to the fact that inferential questions force the learners to make extra associations and process the incoming text more deeply,
they are quite tough to handle. Therefore, it is hypothesized that inferential questions would be as meaningful to preview as global questions.
Hypothesis 4:
In Sherman’s (1997) findings, Questions Before is the most widely favored procedure although her participants’ preference does not parallel their grades. Shih (2004) who replicates Sherman’s (1997) study receives the same response as well. It is generally found that participants tend to prefer Questions Before because they would obtain more sense of security and know which part in the listening text they should particularly attend to. Based on these pioneering studies, it is expected that a majority of participants would favor Questions Before most.
Significance of the Study
By comparing the effect of the three listening test procedures---Questions Before,
Questions After and Sandwich Version---on listening comprehension, this study will offer English teachers in Taiwan deeper insights into English listening process and comprehension. Teachers will understand better how different listening procedures may affect local college English majors’ listening comprehension. Thereby, they may be better informed of which procedure assists students to comprehend a listening text best. Hence, they might try to vary the locus of the comprehension questions rather than always propose the questions after the text. In addition, with the outcome of which proficiency level of participants particularly good at which version, teachers can adjust the placing of comprehension questions according to students’ listening proficiency so that they may learn to listen most effectively. Moreover, the present study deliberately divides the question types into global questions, local questions, and the inferential ones, which Sherman (1997) fails to discuss. With the result of the
study, teachers may know which type of questions is particularly effective for previewing and which type is not.
By surveying students regarding their preference of the three procedures, this study will provide teachers a better understanding of local college students’
perception about the three loci of the comprehension questions. Brown (2000) suggests that “understanding how human beings feel and respond and believe and value is an exceedingly important aspect of a theory of second language acquisition.”
Grabbling up teaching techniques does not give a teacher all-important ability. The rule of thumb is that teachers always bear whom they teach in mind (Brown, 2001).
Teachers’ acquaintance with students’ feelings and thoughts through the questionnaire offers an excellent opportunity for them to reflect upon their ways of training and testing. By urging teachers to check whether their training and testing meet students’
expectation, it allows them to further ponder the necessity of adjustment. Owing to this opportunity of self-inspection, teachers are able to create a more supportive learning environment for students and help them to comprehend the listening materials with greatest efficacy.
Summary and Organization of the Study
This study mainly explores the effect of test procedures on listening
comprehension of EFL college students in Taiwan. Besides the test procedures, two more variables are added: proficiency levels and question types. Thereby, the study also investigates how the relationship of test procedures and proficiency levels, and that of procedures and question types influence their listening performance. Overall, this thesis consists of five chapters. The introduction, proposing research purpose background, definition of terms, research questions, hypotheses, the significance of the study, and its organization is presented in Chapter 1. After this introductory
chapter, a review of previous research related to previewing questions and test procedures is presented in Chapter 2. Chapter 3 describes the research design, including a description of the participants, the instruments, and the experimental procedures. Chapter 4 presents the results and discussion of findings. Chapter 5 contains the summary of major findings of this study, as well as its limitations, pedagogical implications, and suggestions for future research.
