Effects of Task Conditions on Use of Listening Strategies by Chinese

This paper reports an exploratory study on the effects of two task conditions (Zero-task and Task) on listening strategies used by Chinese EFL learners. The purpose of the study is to find out whether different task conditions affect strategy application, whether learners at different proficiency level use strategies differently on different task conditions, and whether strategy instructions have significant effects on learners’ strategy use. Fifty-two first-year university students participated in this study, with ages ranging from 18 to 20. A listening proficiency test (cronbach α = .832) differentiated them into two groups: Ghigh((N=29) and Glow(N=23). The instrument of the study is a questionnaire with 24 items of specific cognitive listening strategies. The participants were required to answer two questionnaires: one on Zero-task condition, the other on Task condition, with two months intervals in between. During the two months, the participants received some classroom strategy instructions. After the instructions, they replied to the questionnaire again. Data of both questionnaires were collected and analyzed. Results show that (1) from Zero-task to Task condition, the overall mean score of the reported frequency of strategy use increased slightly in Ghigh while decreased slightly in Glow; (2) Ghigh used strategies more frequently on Task condition than on Zero-task condition, while Glow didn’t show significant difference on both task conditions; (3) on the same task conditions, the two groups showed significant differences in their strategy application. Ghigh used strategies more frequently than Glow, especially on Task condition; (4) Ghigh responded more actively and effectively to strategy instructions than Glow.
1. Introduction
Learning strategies can be defined as conscious “steps taken by students to enhance their own learning”(Oxford 1990:1); more particularly, “specific actions taken by the learner to make learning easier, faster, more enjoyable, more self-directed, more effective, and more transferable to new situations”(Oxford 1990:8).
Since the 1990s, many studies with diverse focuses have been conducted to investigate effects of various factors on strategy use in language learning, such as language proficiency (Bedell & Oxford 1996; Dreyer & Oxford 1996; Green & Oxford 2003), gender（Ehrman & Oxford 1980; Green & Oxford 1995; Gu 2002）, etc. Task-based strategy studies aiming at assessing specific strategies applied on different task conditions during language learning have attracted more and more attentions.（Cohen,1998;Hsiao and Oxford 2002; Oxford et al 2004).
The present paper partially replicates Oxford et al.’s (2004) study about how task difficulty and proficiency affect the reported frequency of reading strategy used by language learners. The results indicated task-based strategy assessment provide a more detailed, more contextualized picture of strategy use (Hsiao & Oxford 2002) However, does this task-based strategy assessment apply to listening strategy used by Chinese EFL learners? Our study aims to investigate the effects of two task conditions (Zero-task and Task) on listening strategies use by Chinese EFL learners at different proficiency levels.
Firstly we generally review relevant theory and research. Then we explain the methodology of the present study, followed by presentation and discussion of the results. The paper concludes with the findings of our study and recommendation for further studies.
2. Review of research
In this chapter we generally review theory and research on the following two topics: listening learning strategies in L2 field, and task-based strategy assessment.
Listening comprehension and listening strategies
Listening comprehension
Researchers have pointed out that listening plays a key role in language learning because of its contribution to the development of the overall language proficiency (Wolvin & Coakley, 1996; Rost 2002). Listening comprehension is an active and complex process in which learners construct meanings from inputting information through a complex interaction between the characteristics of the input, the types of the declarative knowledge that are received, and the use of strategies to enhance understanding (Abd El Al, 2002: 42, Anderson: 1995). Cognitive theory holds that effective processing of information requires both top-down and bottom-up processing. The top-down processing is “a form of language processing that bases inferences on expectations and predictable generalisations cued by the incoming language’ (Rost, 2002: 96), while the bottom-up processing begins with analysing receiving messages from different levels of organisation of the utterance – from sound, to words, then to clauses, and then to sentences and finally get the intended meaning (Anderson and Lynch, 1988; Brown, 1990; Buck, 2001; Abd El Al, 2002). During this procedure, the learners (processors) use “various techniques to select information, organize it, relate it to what they have already known” (Abd El Al, 2002: 39).

Listening strategies
Learning strategies are conscious “steps taken by students to enhance their own learning”(Oxford 1990:1) and are “a sequence of steps taken by a learner, deliberately, in a specific order for a specific purpose that is to learn, recall or comprehend the target language”(Abd El Al, 2002: 50). Hedge (2000) proposed a four-broadly- categorised taxonomy about learning strategies: cognitive, Matacognitive, communicative and socio-affective strategies. In regard of the nature of listening comprehension, we focus on cognitive strategies which involve both top-down and bottom-up processing approaches.
Some descriptive studies use various methods to describe and identify listening strategies used by successful and less successful listeners. For example, O’Malley, Chamot & Küpper (1989) use think-aloud protocols with high school SEL students to examine what strategies are used in different phases and whether there are any differences of using strategies between effective and ineffective listeners. Their major strategy headings are self-monitoring, elaboration and inferencing. Results show significant differences of strategies used between effective and ineffective listeners. Vandergrift (1992) uses semi-structured interviews, stimulated recall and think-aloud protocols to investigate the strategies used by Core French high school students in transactional and interactional tasks. Findings indicate that there is an overall increase in the total amount of used strategies by the increase of proficiency level and the less successful listeners have to understand the texts by focusing on semantic cues, contextual clues etc. Goh (1998) uses retrospective verbal reports to investigate the cognitive and metacognitive strategies and tactics used by Chinese ESL learners in a university in Singapore to compare the way high- and low- ability listeners apply them. The study shows that the high-ability listeners use more strategies and tactics than the low ability ones.
Yu (2004) uses a written questionnaire with thirty-six specific listening strategies to investigate the effects of strategy use on learners’ autonomy of 160 Chinese EFL learners at two proficiency levels. These strategies fell into three major categories metacognitive, cognitive ones, socio-affective strategies. Results showed high-ability students used more metacognitive and cognitive strategies than low-ability students, while there is little difference in socio-affective strategy use. Another finding in Yu (2004)’s study indicated that students perceived high value of strategies and require strategy instructions to improve their listening abilities.
These studies on listening strategies mainly focus on probing, describing and identifying listening strategies used by successful and less successful listeners. Although they confirm the same finding that both good and poor listeners use strategies, the differences between the two groups of listeners have not been significantly revealed. Moreover, the total number of strategies used and the frequency of use of each strategy “are not necessarily indicators of how successful [students] will be on a [specific] language task” (Cohen, 1998:8-9). On a particular L2 task, the appropriateness of the strategies to the task is likely to be more important than the numbers and frequencies of strategies employed (Oxford et al., 2004).

Listening Strategies training
Strategy training aids learners to become more consciously aware of what strategies might be useful in a given learning situation. Some studies have suggested the implications for L2 listening strategy training. For example, Vandergrift (1999) has explained how L2 listeners can use strategies to enhance their learning processes and how teachers can nurture the development of listening strategies. In her empirical study, Teng (2003) instructs her fifty college students in National Taiwan University of Science & Technology with thirty EFL listening strategies for fifteen weeks. Results show that systematic strategy instruction enhances the improvement of strategy use for EFL listener. Ma (2005) investigates the effects of listening strategy training on Chinese EFL learners. After fifteen-week’s instructions, the sixty-two EFL learners at three proficiency levels (high, intermediate and low) show significant difference in strategy use, especially to the intermediate students. .
Strategy training helps learners become more self-directed, autonomous via the improved use of strategies (Wenden, 1991, 1998; Granham, 1997; Cohen, 1998, 1999); become more aware of their learning processes and in turn increase the willingness and ability to manage their learning (Niyokos, 1996; Mendelsohn, 1995, 1998); enhance motivation (Wenden, 1991, 1998); remove anxiety, reduce uncertainty and foster self-confidence (Niyokos, 1996).
Task-based strategy assessment
As Cohen (1998) argued that when learners respond to a strategy questionnaire without actually doing a L2 task, they may become less accurate about their actual behavior. Learners’ different interpretations of their task experience may influence their reporting of the strategy use in terms of the comparability of responses to the same items.
Ikeda and Takeuchi (2000) empirically examined whether reported strategy use is affected by the presence of absence of an actual task and how types and frequencies of strategies varied according to task difficulty. Results show that having students complete a task significantly affected the reporting of reading strategies. The reported frequencies of strategy use in both the high and low proficiency learners declined in three task situations: no task, easy task and difficult task. Oxford et al. (2004) conducted a similar study to investigate effects of including or not including a language task as part of strategy assessment and how students’ reported strategies differ when the language task is easy versus difficult. Except for the similar findings of significant effects of task situations on reported strategy use, they found that low-proficiency students reported to employ more strategies than high-proficiency students in specific task situations, because they tend to use strategies to compensate the knowledge inadequacy. On the other hand, high-proficiency students, reported less strategy use though, more often tended to use “top-down” strategies like predicting, guessing, finding the main ideas, trying to understand without translating into the native language.
Similar task-based strategy assessment can be applied to listening strategy use by L2 learners. In regard of the nature of listening comprehension and listening strategies, different task conditions may influence learners of different listening ability in their listening strategy use.
3. Research questions in the present study
The present study aimed to examine the effects of the present (namely task hereinafter) or absence (namely Zero-task hereinafter) of a listening task on students’ reporting of their listening strategy use. Since strategy use is closely related to L2 proficiency, we took students’ listening comprehension ability into consideration. Furthermore, we investigate the impact of listening strategy instructions on the reported frequency of students’ strategy use under the two different task conditions.
Specifically, the research questions of the present study are:
1. considering the cognitive listening strategy as a whole, are there any main effects or interaction effects on listening strategy frequencies based on the two independent variables, namely task condition (Task or Zero-task) and listening proficiency level (high or low)?
2. considering each of the strategy item on cognitive listening strategy questionnaire, do high- and low-proficiency listeners report different frequencies of strategy use within each of the task conditions？
3. does classroom listening strategy instruction have any effects on students’ strategy use?
4. Methodology
4.1 Research Subjects
Altogether 83 first-year college students participated in the present study, of whom 47 were male students and 36 were female students, with ages ranging from 18 to 20. They majored in laws and computer science in a large university in southern China. They received College English instructions six hours a week as a compulsory course, which include Integrated Skills of English, Learning Strategy and Listening. They were required to take a nationwide English proficiency test (College English Test Band 4, CET-4) after two years’ study.
All the 83 students were supposed to reply the listening strategy questionnaire twice, respectively before and after the eight-week listening strategy training. However, some of the students failed to reply both questionnaires completely. Finally we got 52 questionnaires that were valid to our research.
After the eight-week listening strategy training, we conducted a listening comprehension test, which consisted of 50 multiple choice test items in three parts: short dialogues, long conversations and short passages. The total score was 50 points. A statistical validity analysis showed that the listening proficiency test was reasonably valid with α (Cronbach’s Alpha) value of .832. With the result of this test, we tangibly classified the 52 participants into two groups. Those whose scores were above the average 35 points were in the high-level group ( Ghigh, N=29), while those scored equal or below the average were in low-level group (Glow, N=23). A t-test in SPSS revealed that the two groups were significantly different in their listening comprehension ability (See Table 1).

Table 1 Descriptive data of participants’ proficiency difference
Group Number Mean SD SE t-test(sig.)
Ghigh N=29 39.48 2.29 .43

.016

Glow N=23 30.26 3.86 .81


4.2 Instrumentation
4.2.1 Listening comprehension assessment. As explained earlier, the listening comprehension test consisted 50 test items in three different parts: 15 short conversations (one-round dialogues between a man and a woman followed by comprehension questions), three long conversations (four to six-round dialogues between a man and a woman) followed by 5 comprehension questions for each one, and four short passages (about 150 words for each passage) followed by 5 questions for each one. All test items were multiple choices with one point for one item. The participants were required to finish the test in about thirty minutes. As stated previously, the listening comprehension proficiency test was reasonably reliable (α =.832)

4.2.1 Listening strategy questionnaire. We specified the cognitive listening strategy questionnaire (LSQ) developed by Yu (2004) into 24 strategies under the three major categories of before, while and after listening. The questionnaire consisted of 24 statements of listening strategies (See appendix A), and the students’ responses were recorded using Likert Scale of 0 (almost never) to 5 (almost always) to measure each strategy. The reliability of the questionnaire employed by the present study was .869 using Cronbach’s Alpha coefficient of internal consistency.
4.3 Data collection procedures
It was made clear to the students that this study would not affect their course grades. They responded to the questionnaires anonymously, but the listening comprehension test would be part of their course grades in order to ensure the validity of the group pattern. Data collection was conducted in the following steps:
Firstly, the students were asked to complete the listening strategy questionnaire (LSQ) without doing any listening task, called the “Zero-task condition” in the present study.
Then eight weeks later, during which listening strategy instructions were implemented once a week into their English course, the students were asked to take the listening comprehension test which involved conversations and passages. The amount of time allotted to the test was thirty minutes. The papers were collected for grouping.
Right after the thirty-minute listening test, the students were required to respond to LSQ again. We consider this as Task condition to differentiate the first stage “Zero-task condition” in the present study.
All the data were collected for analysis to investigate the research questions.
4.2 Data analysis procedures
To answer Research Question 1 in the present study, i.e. whether there are any main effects or interaction effects on listening strategy frequencies under the two task conditions (task or Zero-task) and listening proficiency level (high or low), we used repeated measures analysis of ANOVA to identify the significance. The results of ANOVA would tell us:
-- whether the reported mean frequency of strategy use was significantly different on the two task conditions (i.e. the main effect for task condition)
--- whether the reported mean frequency of strategy use was significantly different across the two proficiency levels(i.e. the main effect for proficiency levels)
-- and whether Ghigh and Glow reported different mean frequencies of strategy use across the two task conditions,
To answer Research Question 2 and 3, i.e. whether high- or low-level listeners report different frequencies of strategy use within each of the task condition in the present study, we employed multiple measures of Paired T-test and Independent T-test. We set the significance level at p<.05 for both T-tests, which would help us uncover the patterns of difference in listening strategy use and be useful for further study. The results of Paired T-test would tell us whether there would be significant difference in specific strategy use by the same level group on the two task conditions. Independent T-test would help us find out whether there would be any significant differences between the two proficiency groups in specific strategy use on either of the two task conditions. Results followed by discussions were presented in the next part of the paper.
5. Results and discussion
5.1 Answer to Research Question 1: Considering the cognitive listening strategy as a whole, are there any main effects or interaction effects on listening strategy frequencies based on the two independent variables, namely task condition (Task or Zero-task) and listening proficiency level (high or low)?
5.1.1 An overall pattern of reported listening strategy use by two proficiency groups across Zero-task and Task conditions
We compared the mean score of LSQ reported by the two proficiency groups on the two different task conditions. Table 2 presents the means and standard deviations of the two groups across Zero-task and Task conditions.

Table 2 Means and standard deviations of proficiency groups across Zero-task and Task conditions
Proficiency Zero-task Task
Ghigh Mean 2.97 3.09
N 29 29
SD 1.4 1.49
Glow Mean 2.82 2.76
N 23 23
SD 1.5 1.4

The mean LSQ scores in the two proficiency groups did not show the same directions on the two tasks. In Ghigh, the reported frequency of strategy use on the Zero-task condition (X = 2.97, SD=1.4) was lower than on the Task condition (X = 3.09, SD=1.49). Thus, high-level students’ overall strategy use tended to increase slightly from Zero-task condition to Task condition.
In contrast, learners in Glow did not show an increase but a slight decrease from Zero-task to Task condition. The reported frequency of strategy use on the Zero-task condition (X = 2.82, SD=1.5) was slightly higher than on the Task condition (X = 2.76, SD=1.4).
This overall pattern did not agree with the pattern of reading strategy use revealed by Ikeda and Takeuchi (2000) and Oxford et al. (2004), whose results showed slight increase from no task to difficult task for low proficiency group while slight decrease for the high-proficiency group. The difference may partly lie in the different nature of listening and reading and their different interaction between task presence and learners’ strategy use.
On the other hand, the mean LSQ scores in Ghigh, both on Zero-task and Task conditions, were higher than that in Glow. This result further supports the previous findings (e.g. Vandergrift,1992; Goh, 1998; Yu, 2004, etc.) that the high-ability listeners use more strategies and tactics than the low ability ones.
Studies (e.g. Vandergrift, 1999; Teng, 2003; Ma, 2005, etc.) showed that systematic strategy training enhances the improvement of strategy use for EFL learners. In the present study, the LSQ in task condition was conducted after an eight-week strategy instruction. From the mean LSQ scores in table 2, we found that Ghigh showed a positive effect of strategy instructions while Glow showed a negative one. In other words, strategy training is more effective for high level students, but less effective for low level students.
To find out the statistical significance of these differences, we used repeated ANOVA with SPSS 13.0

5.1.2 Main effects and interaction effects of task conditions and proficiency levels
Table 3 presents the results of the repeated ANOVA on mean frequency of strategy use on Zero-task and Task conditions. The results showed that there were no significant main effects for tasks and listening levels. In other words, the overall reported mean frequency of strategy use did not differ significantly on Zero-task conditions for the two proficiency group. Neither of the two groups differed significantly on the reported mean frequency of strategy use across the two different task conditions. However, the results revealed a significant finding: on the Task condition, the reported overall mean scores of LSQ were significantly different between the students of the two groups (F(1, 50)=7.485, p<.05). This finding supported the previous discussion that high level students used more listening strategy and were more sensitive to strategy training.

Table 3 Interaction effects of task conditions and proficiency levels
Sum of Squares df Mean Square F Sig.
Z-task
Ghigh *Glow Between Groups .371 1 .371 1.703 .198
Within Groups 10.891 50 .218
Total 11.262 51
Task
Ghigh *Glow Between Groups 1.318 1 1.318 7.485 .009
Within Groups 8.801 50 .176
Total 10.118 51
Ghigh
Z-task* task Between Groups .136 1 .136 .828 .367
Within Groups 9.221 56 .165
Total 9.357 57
Glow
Z-task* task Between Groups .033 1 .033 .138 .712
Within Groups 10.472 44 .238
Total 10.505 45


5.2 Answer to Research Question 2: Considering each of the strategy item on cognitive listening strategy questionnaire, do high- and low-proficiency listeners report different frequencies of strategy use within each of the task conditions？
5.2.1 Main effects of the two task conditions on strategy-by-strategy use by either of the two proficiency levels.
To find out the different effects of task conditions on listening proficiency in terms of each specific listening strategy use, we used paired T-test. Table 4 presents the results.

Table 4 Significance of strategy use on two task conditions by Ghigh or Glow
Proficiency Strategy Zero-task
Mean Task
Mean Sig.
p<.05 Difference
Z-t>T T>Z-t
Ghigh No.3 2.62 2.03 .041 ×
No.4 3.31 4.03 .006 ×
No.7 2.62 3.24 .036 ×
No.9 1.62 1.03 .035 ×
No.10 1.69 2.31 .039 ×
No.18 2.76 3.55 .015 ×
No.24 1.52 2.45 .001 ×

Glow —— —— —— —— —— ——

Within the same proficiency level, the students in Glow did not report to use more strategies either on Zero-task or on Task condition, but the students in Ghigh employed two more strategies on Zero-task condition while they employed five more strategies on Task condition. The two strategies the students in Ghigh reported use more on Zero-task conditions were No.3 (I try to understand the meaning of every word) and No. 9 (I try to give a word-for-word translation from English to Chinese). This indicates that high levels students had better awareness of listening strategies and would try such strategies as detail associations and translation to help them better understand listening materials. This can be proved on the Task condition where high level students employed five more specific strategies including No.4 (I try to think ahead and anticipate what comes next), No.7 (I try to connect what I hear with information I have gained form experience in the world), No.10 (I group information into different meaningful groups), No.18 (I try to take notes about important information) and No.24 ( I make a summary of the information I heard).
The results showed that high level students were good at listening strategy use. They were more sensitive to listening strategy training. After the eight-week training, they had better understanding and awareness of listening strategies. For specific Task condition, they tended to employ more such cognitive strategies as inferring, associating, grouping, note-taking and summarizing skills to help them gain more information from listening tasks. They tended to use strategies to anticipate before listening and effectively process information by the means of note-taking and grouping, and after the listening they would summarize the information to have a good understanding. On the contrary, low level students didn’t show any significant effects on strategy use on Task condition. This may suggest that low level students were not familiar about specific listening strategies, especially on Task condition, where their reported frequency declined slightly.

5.2.2 Main effects of either of the two task conditions on strategy-by-strategy use by the two proficiency levels.
To compare the difference of strategy use by the two groups in a given task condition, i.e. either on Zero-task or Task condition, we used repeated independent T-test to find out the significance.
Table 5 Significance of strategy use by the two groups on given task conditions
Task condition Strategy Ghigh
Mean Glow
Mean Sig.
p<.05 Difference
Ghigh> Glow Glow > Ghigh
Zero-task No.9 1.62 1.83 .038 ×
No.17 3.93 3.60 .049 ×
Task No.4 4.03 3.39 .029 ×
No.5 3.86 3.00 .022 ×
No.6 3.52 2.73 .036 ×
No.7 3.24 2.30 .025 ×
No.18 3.55 2.30 .010 ×
No.21 3.34 2.26 .007 ×
From table 5, we found that on Zero-task condition, students in Ghigh employed strategy No.17 (If I do not understand some information such as a word or a phrase, I try to guess its meaning basing on relevant information about the topic.) more frequently than students in Glow, while students in Glow used strategy No. 9 (I try to give a word-for-word translation from English to Chinese) more often than students in Ghigh. This indicated that high level students generally would employ their world knowledge to process the incoming information, i.e. the top-down processing, while low level students would employ bottom-up processing method by translating into their mother language in order to help them get the intended meaning of incoming information.
When the two groups were given a specific task, their reported strategy use frequencies varied significantly in six specific strategies. The students in Ghigh reported to employ all of these strategies more frequently than the students in Glow. The six strategies included referencing (No.6 I try to relate what I hear to information I have already studied), making personal association (No.5 I try to make personal association with what I hear and No.7 I try to connect what I hear with information I have gained from experience in the world ), resourcing (No.6 I try to relate what I hear to information I have already studied), and note-taking skills (No.18 I try to take notes about important information and No.21 I try to understand the meaning of the key words). This result supported our findings that high level students use more listening strategies than low level students. In order to achieve a good understanding of listening tasks, high level students tended to use referencing technique more often to anticipate what they would listen. They would also make personal associations and use resourcing skills much more often than low levels students to help them process information more effectively. Moreover, high level students would usually take notes about important information more often than low level students. In a word, when processing information in listening tasks, high level students would employ more cognitive listening strategies and more effectively combine these strategies to achieve good comprehension than low level students do.
5.3 Answer to Research Question 3: Does classroom listening strategy instruction have any effects on students’ strategy use?
Researchers hold that strategy training enhances learners’ motivation (Wenden, 1991, 1998), remove anxiety, reduce uncertainty and foster self-confidence (Niyokos, 1996), thus help improve their language proficiency. Some empirical studies (e.g. Vandergrift, 1999; Teng, 2003; Ma, 2005, etc) revealed that implementing strategy training into classroom instruction could enhance students learning process and improvement of strategy use.
In the present study, before the LSQ on Task condition, the students in both groups received about sixteen hours’ (two hours a week for eight weeks) classroom listening strategy training, including such strategies as how to anticipate, how to infer, how to associate listening information with world knowledge, how to make good use of various resources while listening and how to take notes, etc. One of the present researchers gave them the training. After the training procedures, the students took the listening comprehension test. As soon as they finished the thirty-minute test, the students were asked to do the same LSQ as the one they replied on Zero-task condition.
From the results presented in Table 4, we found out that students in Ghigh did demonstrate significant more strategy use (five specific strategies) on Task condition than they did on Zero-task condition. However, students in Glow did not show any significant difference in the strategy use on the two task conditions. On the other hand, results from Table 5 showed that on the Task condition (after the strategy instructions), students in Ghigh employed more listening strategies (six strategies with significant difference) than students in Glow.
Thus, we drew a tentative conclusion that listening strategy training had significant effects on high level students but had no effects on low level students in their strategy use.
6. Conclusion
The present study investigated the effects of two task conditions (Zero-task and Task) on cognitive listening strategies used by different levels of Chinese EFL learners. Results revealed the following findings:
Firstly, taking cognitive listening strategy as a whole, there were no significant interactive effects between the two variables of tasks and listening proficiency. Even though there was a slight increase for high-level learners and slight decline for low level learners in reported strategy use frequencies from Zero-task to Task condition, the differences were not statistically significant. However, the two groups demonstrated a significant difference in overall strategy use frequency when they had a task. High level learners reported to use strategies much more often then their low level peers.
Secondly, for the specific strategy-by-strategy use, high level learners proved to be better strategy users. On one hand, high level students tend to employ some top-down strategies as inferring, resoucing, making personal association to help them process information. They used these strategies more often when they did a Task than when they had no task. On the other hand, low level learners did not show any significant difference on the two task conditions for their specific strategy use.
Thirdly, strategy training was more effective to high level learners than to low level ones. It enhanced high level students’ improvement of strategy awareness and strategy utilization. However, strategy training was less likely to work well with low level students.

Nevertheless, the current study had its limitations, and further studies need to be conducted to prove the findings.
First, the sample was not large enough to generalize the pattern of learners’ reported strategy use. It would be plausible to use a larger population sample to investigate a more generalized pattern in listening strategy use.
Second, the subjects were grouped by the result of one listening test. It would be more convincing to group subjects with a valid and reliable proficiency test.
Third, the task conditions in the present study were not very clearly defined. We took the listening test including conversations and passages as Task condition. It would be advisable to clearly define tasks, for example, conversation or lectures. What’s more, strategy use frequencies may differ dramatically according to the difficulty of tasks.
Fourth, the listening training duration was not long enough to form a pattern for learners at different proficiency levels. Results may be different if systematic listening strategy training could last for longer time, say several months or a year.
Finally, the current study focused only on cognitive listening strategies. What would be the pattern if more categories of listening strategies were involved?
All these limitations need to be taken into consideration for any further studies about interactive effects of task and proficiency.

References:
Abd El Al, A. 2002. ‘The effect of a strategy-based instruction programme on developing EFL listening comprehension skills’. Unpublished PhD dissertation, University of Warwick, UK
Anderson, A. and T. Lynch. 1988. Listening. Oxford: Oxford University Press.
Anderson, J. 1995. Cognitive psychology and its implications. 5th Edition. New York: W.H. Freeman.
Bedell, David A. and Oxford, R. 1996. Cross-cultural comparisons of language learning strategies in the People’s Republic of China and other countries. In Language Learning Strategies around the World: Cross-cultural Perspectives Oxford (ed.) Manoa: University of Hawaii Press
Brown, G. 1990. Teaching the spoken English. 2nd Edition. London: Longman.
Buck, G. 2001. Assessing listening. Cambridge: Cambridge University Press.
Cohen, A. 1998. Strategies in learning and using a second language. London: Longman
Dreyer, Carisma and Oxford, R. 1996. Learning strategies and other preicators of ESL proficiency among Afrikaans speakers in South Africa. In Language Learning Strategies around the World: Cross-cultural Perspectives Oxford (ed.) Manoa: University of Hawaii Press
Green, J. and Oxford, R. 2003. At the theoretical crossroads between ESL and EFL: Predicting English proficiency and exploring learning strategies in Puerto Rico, Manuscript in progress
Goh, M. 1998. ‘How ESL learners with different listening abilities use comprehension strategies and tactics’. Language Teaching Research 2/2: 124 – 47.
Gu, Yongqi 2002. Gender, academic major, and vocabulary learning strategies of Chinese EFL learners. RELC Journal 33:35-55
Hedge, T. 2000. Teaching and learning in the language classroom. Oxford: Oxford University Press
Hsiao, T.-y. and R. L. Oxford 2002. Comparing theories of language learning strategies: A confirmatory factor analysis. The Modern Language Journal 86(3): 368-383.
Ikeda, M. and Takeuchi, O. 2000. Tasks and strategy use: Empirical implications for questionnaire studies, JACET Bulletin 31:21-32
Laviosa, F. 1999. A cognitive apprenticeship model to L2 listening: An experimental study with Italian. Paper, AILA, Tokyo.
Ma, Yunxia. 2005. Incorporating strategy Training into EFL Listening Class: from theory to Practice. Unpublished MA thesis in Huhan University of Technology, Huhan, China, P.R.
Nyikos, M. 1996. ‘The conceptual shift to learner-centred classrooms: increasing teachers and students strategic awareness’. in R. Oxford (ed). Language learning strategies around the world: cross-cultural perspectives. Manoa: Second Language Teaching & Curriculum Center, University of Hawaii
O’Malley, M., A. Chamot. and L. Küpper. 1989. ‘Listening comprehension strategies in second language acquisition’. Applied linguistics 10/4: 418 – 437.
Oxford,R. et al. 2004. Effect of the presence and difficulty of task on strategy use. RELC Journal 42:1-47
Oxford, R. 1990. Language learning strategies: what every teacher should know. Boston, Mass.: Heinle & Heinle
Rost, M. 2002. Teaching and researching listening. London: Longman.
Teng, Huei Chun. 2003. Teaching EFL listening strategies to college students in Taiwan: Paper presented at 12th English Language Teaching Symposium in Taiwan, China
Vandergrift, L. 1992. ‘The comprehension strategies of second language (French) listeners’. Unpublished PhD dissertation, University of Alberta, Canada.
Vandergrift, L. 1999. Facilitating second language listening comprehension: acquiring successful strategies. ELT Journal, 53, 168-176.
Wenden, A. 1991. Learner strategies for learner autonomy: planning and implementing learner training for language learners. Englewood Cliffs, N.J.: Prentice Hall.
Wenden, A. 1998. ‘Metacognitive knowledge: a language learning’. Applied linguistics 19/4: 515– 37.
Wolvin, A. and C. Coakley. 1996. Listening. 5th Edition. Dubuque, IA: Brown & Benchmark.
Yu Zhuoya. 2004. An Investigation into Chinese University Students’ Use of Listening Strategies in Their Self-studies on Listening Comprehension: MA thesis, University of Warwick