Language Management in MICE Industries

Listening cannot be overemphasized since it is a vital skill in terms of the four language arts, language learning, and language teaching. The strategies that MICE-major un iversity students might like are exp lored for their listening comprehension. Moreover, the research revealed the type of listening comprehension test item that could be difficult for students and indicated the difficulty that students might have in comprehending the test item so as to find some good strategies to overcome the d ifficu lties. Furthermore, the classificat ion of listening comprehension test items will conform to that of educational objectives of cognitive domain based on Bloom's (1974) taxonomies of educational learning [5]. Based on the findings, MICE-major university students will review their learning process and ask for more listening practice and strategic guidance in the hope that such activities will improve their listening comprehension and solve their listening difficult ies, so that language teachers will create a win-win opportunity to double-check instructional deficiencies and offer supplementary aids. The qualitative research was conducted through a semi-structure interview. Based on academic performance, subjects recruited in this study were included 24 MICE-major students, 12 of whom were chosen as HPL students while the other 12 as LPL ones at National Kaohsiung University of Hospitality and Tourism in Kaohsiung City, Taiwan. The instruments used in this study included the simulated listening comprehension test of the Intermediate Level of the GEPT, and listening comprehension strategy questionnaire designed by the researchers. The pedagogical implication of research results will help ESP teachers to enhance their MICE-major university students’ listening comprehension in their oral communication as required at work.


Introduction
Listening is an important skill among the four language arts because it plays an essential role not only in one's first language development but also in people's daily oral communicat ion. According to Davis (1980), listening ability is much more useful and important than speaking ability [26]. Likewise, second language (L2) learners build up their L2 co mpetence by understanding the comprehensible input, and thus speaking emerges (Krashen, 1985) [46]. To put it d ifferently, listening is the basis of language acquisition (Wing, 1986) [76]. Generally speaking, EFL learners do not have enough input of the target language in their daily life, so it is tougher for them to acquire listening competence. As a result, communicat ing with English speakers becomes difficult or even impossible for them.
Over the last few years, the MICE industry has become a strong economic force worth appro ximately £270 b illion in the global market (Kavanagh, 2010), which provides mass emp loyment [45]. Recently many businesses in the MICE industry have emp loyed people fro m d ifferent cultural and linguistic backgrounds. The ability to co mmunicate in more than one language with others is a necessity for this industry. Being competent in a second language, if not more, could help foster good work p ractices by reducing language barriers which could 'influence across cultural knowledge sharing, dialogue, relat ionship building and networking' (Bonache & Brewster, 2001) [6]. Apart fro m the establishment of good workp lace co mmunicat ion, MICE workers also need to be able to serve guests. This is why being competent in language communication for MICE workers become vital to the industry. In other words, competent listening and speaking skills are very important prerequisites for MICE employees. According to Jayawardena (2001:313), 'to sustain a thriving tourism industry, the host population must also be linguistically educated in the requirements of their target market.' [44] As a matter of fact, in all sorts of MICE roles, it is fundamentally important that staff would be able to communicate effectively and accurately with colleagues, guests and patrons. In particular, MICE staff needs to know how to communicate with others in a wide variety of situations, for example: with an unsatisfied guest, a stressed business man checking into a hotel or tourists who are lost or confused. It is also important that MICE staff is sufficiently confident with their use of verbal English, so that they could quickly and effectively put others at ease and provide the informat ion that they need. In MICE industries, assisting guests and patrons to feel relaxed and comfortable is always a very critical part of the job.
Undoubtedly, people working in the field of MICE should achieve a better English level of English in order to take up their job well. Thus, in this kind of situation, many educational institutes in Taiwan and others around the world have offered English-for-Specific-Pu rposes (ESP) courses for MICE-major students. The majority of these ESP courses are heavily orientated towards the way to provide their students with necessary and practical linguistic training and business skills required to ensure a successful career in the industry. Actually, of all the trainings, language learning is a challenge, but it could be overcame by showing students that languages are 'an integrated component of effective international businesses' (Embleton 1992:4) [29]. As a result, in order to pro mote active language learning in the aspects of listening and speaking abilit ies, institutes need to adopt some strategies as 'language training which requires a strategy within an integrated language strategysimp ly because of the vast range of choices and their associated pitfalls' (Hagen 1992:49) [80]. However, with respect to language acquisition process, listening skill is acquired primarily before other language skills. To enhance language learners' oral co mmunication proficiency, a language teacher should put emphasis on improving students' listening comprehension first while teaching English conversation. Therefore, the present study would explo re strategies that MICE-major university students might employ for their listening comprehension frequently the most, discuss those students' learning motivation, and address the correlation among the students' English academic performance, listening strategies, and learning motivation.

Purpose of the Study
The purpose of the present study is to investigate MICE-major university (MICEMU) students' performance on different types of listening comprehension test items. The correlation among MICE-major university students' English proficiency, learning motivation, students' English proficiency, and their use of listening comprehension strategies is probed in this study. In addition, the correlation between the high proficiency level students' learning performance and the low proficiency level students' learning performance on each type of listening comprehension test items of the Intermediate Level of the GEPT is exp lored. Moreover, another purpose of the study is to find out whether there is any relation between the performance on listening comp rehension test and the most frequently-used listening strategies. Furthermore, whether the listening test is valid and reliab le in evaluating MICE-major university students' listening comprehension would be attested in this study.

Research Questions
The present study is aimed at analyzing the correlat ion between MICE-major un iversity students' academic performance and their performance on the listening comprehension test items so as to explore their listening strategies and difficulties. Specifically, the following research questions are addressed in this study. 1. What difficult ies of listening comprehension would the MICE-major university students with high level and low level of proficiency encounter while they are taking the listening comprehension test of the Intermediate Level of the GEPT? 2. In listening comp rehension, are there any d ifferences in difficulties between the h igh-proficiency-level and the low-proficiency-level for these MICE-major university students? 3. What strategies in listening comprehension would MICE-major university students with high-level and low-level proficiency might employ while they are taking the listening comprehension test of the Intermediate Level of the GEPT? 4. Are there any differences in listening strategies for these MICE-major university students from high-level and low-level proficiency?

Nature of Listening
Listening is not the same as hearing. According to the definit ions of "to hear and to listen" from the Unabridged Random House Dict ionary, "hear" is ascribed to the capability of perceiving sounds with the ears whereas "listen" refers to the purpose of hearing. Hearing is regarded as the passive activity of what ears can do. Listening is considered the active activity of how the ears and the mind interact with each other. In other words, hearing refers to the physical activity of sound perception while listening signifies the complex mental activity that involves not only aural perception, but also interpretation of the meaning of utterance (Rixon, 1981) [58].
In fact, researchers (Dunkel, 1986;Glisan, 1985; have proved that listening comprehension should be an interactive and interpretative process in wh ich listeners are engaged in a dynamic construction of mean ing [28,33,59]. Each element in the successful interpretation of the utterance plays a critical ro le in the process of assigning mean ings to the utterance. As a result, any stimu li wh ich are clearly perceived are only the beginning when an interpretation is assigned to the perceived sounds.
Surely, "Listening" is a dynamic mental process in which listeners are actively involved in the construction of mean ing, in an interpretation of the speaker's intention, and in responding to spoken stimuli by using their linguistic and communicative co mpetence as well as world ly knowledge. Besides, listening is a mult ifaceted and dynamic mental activity. Effective listeners are capable of processing aural stimuli and applying linguistic and communicat ive knowledge successfully to decipher the speaker's intended meanings.

Models of Listening Comprehension Processes
In the past decades, research on listening has explored the mult idimensional features of listening. Chastain (1976) proposed five components of listening comprehension [22]. While listening, the listeners first distinguish and discriminate the sounds, intonation patterns, and the perceived voice qualities, so that they would perceive an entire message. After that, they would hold the message in their auditory memo ry and process it. Next, the listeners would decode the message as to what the speakers have uttered. At last, they would utilize the message and/or store it in their mind. Chastain's model revealed that it was the listener who constructed the meaning on the basis of the perceived sounds and then made use of them.
In addition, Wolff and Marsnik (1992) developed the receptive-transactional process of listening. They suggested a three-stage-process: hearing and attending, interpreting and receiving, and retaining and responding [78]. While the first two stages belonged to the construction process, the last one was the utilization process. Bro wn (1995) also emphasized the listener's active involvement in the co mprehension process and adopted a computing analogy for mo re exp lanation [7]. His model was described below: (a) identify ing the perceived message (the input or aural stimuli), (b) searching preexisting files to match the new message with what s/he listened to, (c) filing or saving the new message, and (d) making use of the new message (the output or the utilizat ion process). The listening processes of these two models are consistent with the two major processes proposed by Clark and Clark (1977) [24].

Cognitive Processes
Cognitive psychologists interpret language comprehension as informat ion processing, in wh ich aural input is first decoded (bottom-up processing) and then is processed on a general cognitive level (top-down processing) (van Dijk & Kintsch, 1983) [73]. Only part of the overall listening process is exp lained in models of listening processes. By contrast, information processing is generally accepted because this interactive listening processing is identified as bi-direct ional (Wolff, 1987) [77]. Further, it points out that listening comprehension would be affected not only by incoming aural stimuli (data-driven processing) but also by the listener's mental knowledge structure (concept-driven processing).

Factors Affecting Listening Comprehension
Factors affecting listening comprehension include text factors, speaker factors, and listener factors wh ich include language proficiency level, memory as hindrance, attention & affect, age, and background knowledge.

Backwash Effect
The effect of testing on teaching and learning is known as backwash or washback. It could be beneficial or harmful. Test designers should make sure that the test's backwash effects are beneficial.
Some ELT specialists (Hughes, 1989;Weir, 1990) have given cogent advice on how to produce beneficial backwash [40,75]. It is widely thought that authenticity and directness of assessment facilitate positive consequences for teaching and learning. Hughes (1989: 44-47) outlined seven ways to pro mote positive backwash [40]: 1. Test the ability of students' development you want to enlarge. 2. Sample widely and unpredictably. 3. Use direct testing. 4. Make testing as criterion-referenced as possible. 5. Base achievement tests on objectives. 6. Ensure that test is known and understood by students and teachers. 7. If there's any necessity, provide assistance to teachers.

Listening Comprehension Test and Difficulties in Listening Comprehension
Listening is a complex and active mental process that involves perception, attention, cognition, and memory. During the process of listening comprehension, various factors may affect learner listening ability. A nu mber of experimental findings have reported on the listening difficult ies experienced by language learners in different contexts.

Differences in Listening Difficulty between High and Low Proficiency Listeners
With the instrument of the GEPT listening comprehension test, it was found in Chao and Cheng's study (2004) that most of the difficult ies (43 items out of 49) between senior high school listeners with high level and low level proficiency achieved significant differences. In terms of the relationship between listeners' proficiency levels and their listening difficulties, low achievers encountered more difficu lties in listening activities than high achievers did; especially in listeners' factors and their listening processes.
The subjects in the questionnaire revealed that the most prominent difficu lty for high ach ievers was in text factor category-unknown words in texts, while the most common difficulty for low achievers was attributed to listeners' factor category-little practice in English listening. In addition, low achievers tended to focus on word level, which resulted in their inability to grasp the overall mean ing of the listening comprehension. This was consistent with the findings by Hwang (2004) that less proficient EFL listeners met greater comprehension problems at perceptual stage than more proficient listeners did [43].

Listening Comprehension Strategies
Listening comprehension strategies refer to skills or methods for listeners to directly or indirectly achieve the purpose of listening comprehension of the spoken input. Nu merous studies related to listening comprehension strategies of language learners have been published in the past two decades.
Murphy (1986) conducted a study to investigate differences between more and less proficient learners in listening strategy use [52]. The results indicated that more proficient learners used a wide variety of strategies than less proficient learners. In addition, mo re proficient learners focused more on rhetorical organization and main ideas, while less proficient learners focused more on mean ing and pronunciation of unknown words. O'Malley, Chamot, and Küpper (1989) used think-aloud procedures to investigate the listening processes of Hispanic students learning English as a second language (ESL) [54]. The findings pointed out that the mental processes students used in listening co mprehension paralleled three phases of comprehension process: perceptual processing, parsing, and utilizat ion. In the initial (perceptual p rocessing) stage, more effective students monitored their attention better than less effective students did. During the second (parsing) stage, effective learners emp loyed more top-down processing strategies than bottom-up strategies. In the final (utilization) stage, effective learners used elaboration and inferencing strategies more frequently than the less effective learners did. Vandergrift (1996) conducted a study to investigate what strategies high school students used when learning French, and how course level and gender affected their strategy choices [72]. The results showed that compared with metacognitive and socio-cognitive strategies, high proficiency students most frequently adopted cognitive strategies. It was reported that the mean nu mber of metacognitive strategies increased with level. Furthermore, female students used a greater variety of metacognitive strategies than male counterparts, which was also consistent with Bacon's (1992) finding [4]. Goh (2000) explored the use of cognitive and metacognitive strategies among listeners with high and low proficiency [34]. She used retrospective verbal reports to gather data from the participants. The findings indicated that high proficiency listeners used more strategies and tactics than low proficiency ones did.

English for Specific Purposes (ESP)
ESP could be defined to meet specific needs of the learners. Strevens (1988) argued ESP may be restricted if the language skills were to be learned instead of being taught [65]. Hutchinson and Waters (1987) indicated that "ESP is an approach to language teaching once (the) content and method are based on the learner's reason for learning (p.19) [41]. Loren zo (2005) stated that "ESP students are usually adults who already have some acquaintance with English and those whose purpose is to communicate with others through a set of professional skills in order to perform particular job-related functions" (para. 1) [49]. Hutchinson and Waters (1987) provided three reasons for the emergence of ESP: demand in the new world, revolution in linguistics, and new focus on learners [49]. Dudley-Evans (1998) defined three characteristics of ESP (p.4): (a) ESP meets the specific needs of learners; (b) ESP uses the underlying methodology and activities of the discipline it serves; and (c) ESP focuses on the appropriate activities for grammar, lexis, register, study skills, discourse, and genre [27]. Anthony (1997) indicated that ESP is generally designed for intermed iate or advanced students since ESP courses requires some basic knowledge of the language system, but it could be used with beginners [2]. Carter (1983) classified ESP into three types: English as a restricted language, English for Academic and Occupational Purposes (EAOP), and English with specific topics [16]. Moreover, three features are common to ESP: (a) authentic materials; (b) purpose-related orientation; and (c) self-direction (Dudley-Evans, 1998) [27]. Sysoyev (2000) introduced a framework for the development of the ESP course, including student analysis, goals, objectives, teaching materials, content design, course planning, and evaluation [68]. Fu rthermore, there are five addit ional co mponents: authenticity, research-base, language/text, needs, and learning/ methodology that originate fro m the real world and ESP pedagogy (Swale, 1990) [67]. Browne (2011) indicated that junior and senior undergraduate students require higher-level skills [12]; thus, the scope of English education needs to be narrowed fro m general English materials to ESP materials. English for specific purposes (ESP) for Taiwan MICE industries have been transformed recently fro m a labor-intensive economy to a knowledge-based economy. The Taiwan Ministry of Education priorit ized touris m and hospitality relevant to MICE for national develop ment under the service industry (Chang & Hsu, 2010) [19].

Research Design
The present study adopted a quantitative approach using a semi-structure interview to collect and analy ze the data for findings regarding MICE-major university students' language learning motivation and their use of listening comprehension strategies. The target subjects consisted of MICE-major university students who had received instruction in English for at least one year, fro m a university in southern Taiwan. Students of two groups were selected fro m three d ifferent English proficiency levels (HP, M P, and LP levels), inclusive of 150 participants in this study. Instruments employed in the study were: 1) Intermed iate listening proficiency test of GEPT, 2) the Indiv idual Background Questionnaire (IBQ), and 3) The Interview Question Guide (IQG).

Research Structure
Based on the research questions, the research structure could be addressed as follows. between the high-proficiency-level and the low-p roficiency-level M ICE-major university students have encountered in their listening comprehension test. 2. C2a-C2b indicates the comparison of listening comprehension strategies the high-proficiency-level and the low-proficiency-level MICE-major university students have used in their listening comprehension test.

Subjects
Based on their academic performance, subjects recruited in this study included 24 students from Depart ment of Hotel Management, Depart ment o f Food and Beverage Management, Depart ment of Hospitality, Exhib ition & Marketing Management at Nat ional Kaohsiung University of Hospitality and Tourism in Kaohsiung City. According to their academic performance, 12 subjects were chosen fro m the total students at high proficiency level, and another 12 subjects were selected at low proficiency level. Therefore, 24 students were interviewed respectively.

Instruments
Two sorts of instruments were used in the present study. The first instrument used in this study was the simulated listening comprehension test of the Intermediate Level of the GEPT, which was included in the test-book published by the Language Training & Testing Center (LTTC) in 2015. The other one included three parts: Background Information, Learning Motivation, and Listening Comprehension Strategies.
With regard to the scheme of the listening comprehension test from the Intermediate Level o f the GEPT, it comprised three sections, as shown in Table 2.  In Section One, eight pictures were used as questions.
Subjects were asked to look at the picture and then choose the optimal one fro m the four aural descriptions, wh ich best matched the picture. The first section was aimed to test the subject' s comprehension. Since the subjects were ready to comprehend the meaning of the pictures in Section One, they might choose the matching aural description.
As for Section Two, there were ten mult iple-choice questions. In the second section, subjects were required to complete a dialogue by selecting the correct response or choosing a proper answer to a question from the four choices after they had listened to the question. The main focus in the section was to test subjects' ability to comp lete the dialogue. Likewise, there were ten mult iple -choice questions in Section Three. Ho wever, in the third section, subjects were not asked to complete a dialogue. Instead, they were asked to choose an appropriate response from four options after a question was introduced at the end of the dialogue. Actually, it demanded much higher levels of cognitive processing than the previous two sections.
With respect to the Interview Question Guide (IQG), 6 questions comprised the Questions for Listening Difficulties, while the Questions for Using Listening Comprehension Strategies consisted of 21 sub-questions, as categorized into 10 main questions, as shown in Table 3.

Procedures
At the start, the researcher told the subjects about the purposes and procedures of the present study and asked them to cooperate together during conducting the process of the test and the Interview Question Guide. The subjects were told that their test results and answers to interview questions would be counted and analyzed. Before the test started, the researcher required the subjects to follow the directions in the test book and listen to the tape carefully. They were also told to choose only one correct response to each question. In addition, the subjects were asked to pay attention to the procedures and take the test seriously. It took the subjects 35 minutes to finish answering the listening comprehension test.
Moreover, qualitative data was obtained by conducting a semi-structure interview. In order to explore MICE-major university students' difficult ies and strategies while they were taking the GEPT listening comprehension test from a more profound perspective, 24 students of different proficiency levels were asked to part icipate in an interv iew. Moreover, to make them well -prepared in the interview process, an interview guide was designed before the implementation of the interview. The interview guide included 6 open-ended questions related to the subjects' most difficult part in the GEPT listening comprehension test, as well as difficult ies resulting fro m the listening processes, listener factors, speaker factors, and text factors. In addition, 21 questions relevant to the listening strategies the subjects used were also investigated. It took each participant about 40 minutes to answer all the questions in the interview.

Figure 2. Procedures for Conducting this Study
After the test and the interview questions were completed, the researcher began to calculate the correct responses to each of the questions on the listening comprehension test. Also, all the data collected from interview were co mputed and analyzed qualitatively. The whole set of procedures of the study are illustrated in Figure 2.

Data Analyses
The section could be divided into two parts. The first part was the counting of the scores of the three item types fro m the listening comprehension tests taken by those subjects. The collected data and accounts for the students' answers to the interview questions were analyzed in the second part.

The Relationship between Proficiency Differences and Performance on Three Types of Test Item
Based on Table 4, both HPL and LPL students did very well in the first type of test item. In addit ion, LPL students performed better in the second type of test item than in the third one. On the contrary, HPL students scored higher in the third type of test item than in the second one. However, consistent with Bloo m (1974) [5], if there are t wo questions about different abilities of cognitive processing, it would be easier for students to answer a question which requires the ability of applicat ion than a question that requires the ability of analysis. To put it d ifferently, the third type of test item, which was aimed to test students' ability of analysis, must be mo re d ifficult than the second one that was aimed to test students' ability of application. But in the present study, the degree of difficulty was only reflected in LPL students' performance on the three types of test item, not in HPL students'. HPL students performed better on the third type of test item that was more difficult than the second one. Consequently, we could argue that HPL students' ability of analysis is better than their ability of application. However, to justify this argument, we might have to increase the number of our subjects in the further study.
To sum up, according to Buck (2001), listening comprehension involved various kinds of knowledge--linguistic knowledge, details of the contexts, extra-linguistic knowledge, and general world knowledge [14]. In other words, any kind of knowledge would influence listeners' understanding of the listening text and then would compensate for each other. For high proficiency listeners, their better linguistic knowledge, world knowledge, and better strategies would make listening tasks easier. On the contrary, low proficiency listeners who were not equipped with various or adequ ate knowledge wou ld be unable to apply the needed knowledge in the listening process. As a result, during the listening process, low p roficiency level students would encounter difficu lties in the listening material aspect more frequently than high proficiency level ones would.

Interview Question Guide
There were 24 MICE-major university students recruited as the interviewees in the present research. Of them, 12 students were selected as high-proficiency level students categorized HPLS1-S12, wh ile the other 12 ones were chosen as low-proficiency level students classified as LPLS1-S12. The interviewees were asked about their listening difficult ies while they were taking the GEPT listening comprehension test. When it comes to the most difficu lt part in the test, almost half of the interviews (11 participants) considered Part 3 (i.e., Short Conversation) to be the most difficult section. In addit ion, nine students had the most difficulty with Part 2 (i.e., Question or Statement Response), and two subjects encountered problems in Part 1 (i.e., Picture Description).
Finally, one interviewee thought both Question Response and Short Conversation were the most difficu lt ones, and one MICE-major university student indicated that she had difficu lty in all the three parts. Therefore, in terms of the overall interview results, Part 3 was the most difficult section, followed by Part 2 and then Part 1.
As for the difficu lties in Part 2, Question or Statement Response, almost all of the subjects responded that the text was too short for them to speculate the answers. In addition, the difficu lties in Part 2 also consisted of too many questions, unknown vocabulary and phrases, unfamiliar test type, unknown connotation, and lack of attention at the beginning of listening. The excerpts below are quoted from the interviewees of LPLS5 and HPLS2. LPLS5 was a female freshman who got a GEPT listening score of 21, and HPLS2 was a male sophomore who got a GEPT listening score of 37 With regard to the difficulties in Part 3, short conversation, one subject said that it was harder to judge the meaning of the sentences in Chinese in this part than in Part 1 and 2, and another interviewee indicated that it was not easy to guess the answers in Part 3. Furthermore, the subjects also confessed their difficulty with unfocused attention in the conversation section. The following statements are quoted fro m HPLS8 and LPLS1. HPLS8 was a female sophomore who got a GEPT listening score of 39, and LPLS1 was a male freshman who got a GEPT listening score of 18.
[HPLS8]「第一大題只是選圖，問題也不會難，而 第三大題就沒有那麼好理解，因為它比較長吧！而 且有些單字比較專業所以聽不太懂。而第二大題一 句 話通常不會太難啊！」 "The Part 1 was to choose a picture, and the question was not difficult. But it was not so easy to comprehend the conversation in the third part because the texts were longer and there were some specialized wo rds.' Besides, the sentences of the second part were not very difficult." [LPLS1] 「 Conversation，因為它給你一段話講完馬上判斷它 的語意然後在選出正確答 案。就是判斷它文中的意思然後在選出正確答案， 其它兩部份還不會這樣 子。...第三部分對話也很容易讓人分心。」 "Conversation was the most difficult part, because it gave you a stream of speech to judge right after listening. That is, you had to judge its Chinese meaning in order to choose the right answer. This would not happen in the other two parts…The third part was the most difficult one because I was easily distracted by the long conversation." There were two interviewees whose difficult parts were more than one. One subject, HPLS4, expressed that his difficulty lay in Part 2 (i.e. question response) and Part 3 (i.e., short conversation) The texts were longer, so he had to pay attention to dialects, slang or unknown phrases. If he did not understand what he heard, he would get lost and miss the subsequent messages. HPLS4 was a male sophomore who got a GEPT listening score of 36. The other participant, LPLS6, said all three parts were d ifficu lt for her. She attributed her d ifficulty to little practice in English listening, the fast speed of delivery, and the linking sounds between words. In addition, even if she heard familiar words, she still had to think about them for a wh ile to recall the mean ings. LPLS6 was a female fresh man who got a GEPT listening score of 16. The following are their explanations.
[HPLS4] 「 基本上我覺得大概是question response 跟長對話的部分 (註：第二、三部份 )要稍微注意一 下，因為就是裡面可能會有美國的slang，或者是dia lect，這些方面可能就是對非英語系國家的學生來講 會蠻吃力的。然後再來就是不懂的片語，就是我們 不熟悉的片語方面我們沒有辦法去確實的掌握它的 整 句的語意部分。」 "Basically, I thin k long conversation and lecture (i.e., part B and C) should be noticed because there were American d ialects or slang, wh ich were difficult for students from non-English speaking countries. Then, the phrases that we didn't know caused difficu lties. When hearing unfamiliar phrases, we were unable to capture the meanings of the whole sentences exactly." [LPLS6] 「我平常都沒有在聽英文聽力之類的東西，所以聽 力 不是很好。」 "Because I do not usually practice Eng lish-listening, my listening is not good enough." Although the interviewees had different parts of difficult ies in Intermed iate GEPT listening comprehension test, the hierarchy of difficult ies were represented by their scores of test items. They got fewer scores in Part 3 Short Conversation, while they gained more scores in Part 1 Picture Description. Therefore , it could be concluded that the interview results were echoed from the results found by their performance on the GEPT listening comprehension test items.

The Differences of Listening Comprehension Difficulties between HPL and LPL MICE-Major University Students
Regarding interviewees' listening difficu lties, the interviews showed that there were many differences between high-and low-proficiency listeners. In terms of the process differences, four low-proficiency listeners were found having difficu lty in translating the messages into understandable Chinese, whereas no high-proficiency listeners didn't have any troubles in this part. . One high-proficiency listener even said that he (or she?) did thinking without translation. In addition, two low-p roficiency listeners said that they did not get the main idea after listening, but no high-proficiency listeners responded the same difficulty. Moreover, there were t wo low-p roficiency listeners who had a hard time in recognizing indiv idual words; however, high-proficiency listeners did not sense this difficulty. Finally, one high-proficiency listener said that there was no difficu lty in the process of listening for h im. The fo llo wing is an excerpt fro m LPLS2. LPLS2 was a female freshman who got a GEPT listening score of 20.
Q2:「 當你這次在做 GEPT聽力測驗時，你在理解聽 力 測驗內容的過程所遭遇到的聽力困難有哪些？」 "In terms of the listening process, what difficulties did you encounter while taking the listening section of the GEPT?" [LPLS2] 「要聽的時候還要把它轉換成中文，聽起來可能會 比 較慢。」 "While listening, I needed to translate the message into Chinese. In this way, I might listen slowly." With regard to difficult ies in the listener's factors, there were many differences between listeners' two levels of proficiency, which indicated that lower-proficiency listeners had a higher frequency of meeting difficult ies than higher-level ones did. First of all, high-proficiency listeners were more confident than low-proficiency ones. There were two h igh-proficiency listeners showing no difficulty, while two low-p roficiency listeners indicated their lack of confidence in English listening. One low-p roficiency listener even said that she was poor in overall English competence. Second, two low-proficiency listeners felt nervous, and one of the subjects also suffered fro m an xiety while taking the test; however, no high-proficiency listener felt these difficult ies. Third, low-p roficiency listeners had a greater proportion of subjects than high proficiency ones in lack of vocabulary. The interviews indicated that two high-proficiency listeners and five lo w-proficiency listeners said that their limited vocabulary knowledge caused difficulties in listening. At last, low-proficiency listeners showed a larger proportion of participants than high-proficiency listeners in lack of listening practice. It was found that one high-and five lo w-proficiency listeners seldom listened to Eng lish. Take LPLS1 and LPLS2 for instance. LPLS1 was a male freshman who got a GEPT listening score of 18, and LPLS2 was a female fresh man who got a GEPT listening score of 20.
Q3: 「你覺得自己的哪些因素造成你在聽 GEPT時的困 難 ？」 "What factors caused you to encounter difficu lties while you were taking the listening section of the GEPT?" [LPLS1] 「 就覺得自已聽力很爛，沒有多聽多練習。」 "I felt my listening ability was poor, and I did not practice English listening a lot." [LPLS2] 「 練習不夠。」 "I didn't practice enough." As far as the difficult ies of the speaker are concerned, the number of lower-proficiency listeners was larger than that of higher ones in each difficulty. Two low-proficiency students experienced difficult ies in male speakers' speech, while no high-proficiency listener had the same problem. However, two high-proficiency listeners, and three low ones had confronted difficulties in speakers' accents.
In addition, there were two listeners with high-level proficiency and four listeners with low-level proficiency who felt that the speakers' speed was too fast for them.
Although the three different kinds of proficiency learners almost experienced the same difficulties, there were slight differences. Co mpared with h igher ones, lower-proficiency listeners tended to experience the difficulties more frequently. It was even found that five students with high-level proficiency and three with low-level proficiency felt no difficulty fro m the speakers. The interview findings indicated that learners with lo w-level pro ficiency faced greater difficult ies fro m male voices, speakers' accents, and fast speed rate. The excerpt fro m HPLS9 is shown below. HPLS9 was a male sophomore who got a GEPT listening score of 40.
Q4: 「當你這次在做 GEPT聽力測驗時，有哪些講話者 的 因素會造成你的英語聽力困難？」 What factors caused you to encounter difficulties while you were taking the listening section of the GEPT?" [HPLS9] 「應 該沒有吧！我覺得都還可以接受。」 "None! I think it was all acceptable." With regard to d ifferences of d ifficu lties in the text factor, the most frequently mentioned difficu lty was texts fro m specialized field. There were three listeners with high-level proficiency and seven listeners with low-level proficiency who encountered difficu lty in terminology. Clearly, listeners with low-level proficiency had trouble understanding specialized texts more frequently than higher ones did. Therefore, the interviews have shown that although all listeners suffered fro m texts in specialized fields, listeners with low-level proficiency had more problems than higher-proficiency listeners did in comprehending the spoken texts. To take LPLS8 for example. LPLS8 was a female freshman who got a GEPT listening score of 19.
Q5: 「當你在做GEPT聽 力測驗時，有哪些聽的內容會 造 成你的英語聽力困難？」 "What factors caused you to encounter difficu lties while you were taking the listening section of the GEPT?" [LPLS8] 「比 較專業的內容，例如說專有名詞。」 "Texts that are selected from specialized fields. For example, the terminology was difficult." To sum up, the data from the qualitative interview both indicated that listeners with high-level proficiency experienced difficulties less frequently than listeners with low-level proficiency. Besides, the findings fro m the interview proved differences of difficulties between HPL and LPL MICE-major university students in the listening process, listener factor, speaker factor, and text factor. Therefore, it could be concluded that the listeners' degree of difficulty was highly related with their listening proficiency level. To sum up, the data fro m the qualitative interview both indicated that listeners with high-level proficiency experienced difficu lties less frequently than listeners with low-level proficiency.

The Differences of Listening Comprehension
Strategies MICE-Major University Students with High-Level and Low-Level Proficiency May Employ While They Are Taking the Listening Co mprehension Test of the Intermediate Level of the GEPT In the interview data, among the 24 interv iewees, several subjects said that they would read the four options before listening. It was said that three listeners with h igh-level proficiency and one with low-level proficiency read the items and then decide possible answers prior to listening for specific aspects of the texts. The result ind icated that listeners with h igh-level proficiency showed the best performance in selective attention. The fo llo wing example demonstrates HPLS6 interviewee's strategy use. HPLS6 was a male sophomore who got a GEPT listening score of 35 There were two d ifferent strategies used while the learners listened to the texts. To begin with, one interviewee with high-level proficiency said that he/she would think only in English without any translation; however, no listeners with low-level proficiency employed this strategy. As to another cognitive strategy use, four high-and two lo w-proficiency four listeners with high-level p roficiency and two with lo w-level paid attention to key words. To take HPLS 11 and LPLS9 for example , HPLS 11 and LPLS9 were female sophomores who got a GEPT listening score of 38 and 22 respectively. Q8: 「在聽 GEPT測驗的時候，你用什麼方法幫助您聽 懂 測驗內容? 」 "What strategy did you use to help understand the texts while listening?" [HPLS11]「我會記關鍵字 …我比較會用英文去思考 。 」 "I would memo rize key words…I tended to think in English." [LPLS9] 「就是看題目上的關鍵字然後去聽，就是如果它唸 出 來，那個關鍵字有出來就會比較注意聽吧！」 "I would read the options and pay attention to the key words. If the key words appeared, I would listen to the content with more attention." These findings suggested that there was a general tendency which reflected greater use and wider range of use for listeners with h igh-level proficiency, who only think in English and find key words.
After listening, recalling what was heard and crossing out the impossible answers were the most common strategies applied by all the subjects. As for the strategy differences, listeners with high-level proficiency chose to recall the main ideas, p ick up the most reasonable answer, and then select the answer by co mmon sense. Listeners with low-level proficiency chose to translate English into Chinese or select the answer with longest sentence. In addition, there was one listener with lo w-level proficiency who focused on details, and another listener with low-level proficiency said that he/she didn't use any strategies since the time for responding to each question was too short. Therefore, listeners with h igh-level proficiency utilized a variety of strategies, wh ile listeners with low-level proficiency relied more on translation strategy. The following excerpts are fro m LPLS3 and LPLS4. LPLS3 was a female freshman who got a GEPT listening score of 15, and LPLS4 was a female freshman who got a GEPT listening score of 17 It was found that higher-proficiency listeners tended to use strategies more frequently and flexib ly. On the other hand, lower-proficiency listeners favored bottom-up strategies such as translation. Generally speaking, the findings fro m the qualitative data indicated that higher-proficiency listeners with high-level proficiency used listening strategies better.

Conclusions
In terms of MICE-major university students' listening difficult ies wh ile they took the test of GEPT, it was found that text factor was the most difficult category, follo wed by listener factor, listening process, and then speaker factor. When it co mes to the top-ten difficult ies, four difficulties in listener factor, three in text factor, t wo in listening process, and one in speaker factor were included. As for the listener factor, the difficult ies consisted of MICE-major university students' English proficiency (i.e., limited vocabulary), unfamiliar contents, lack of practice, and insufficient background knowledge. Difficulties in the text factor were repetition (i.e., unrepeated materials), unknown phrases and unknown words. In the listening process, two perception problems were included: unable to pay attention to the subsequent informat ion while thin king about the previous parts and inattentive at the beginning of listening so that they miss the initial messages. Lastly, speed rate (i.e., fast speed) was the only obstacle in the speaker factor that ranked as the highest difficulty.
As for listening strategies frequently used by MICE-major university students, the listening strategy category most often used by them was metacognitive strategy, followed by cognitive strategy and then affective strategy. Concerning the top-ten most frequently applied strategies, five metacognitive strategies and five cognitive strategies were included. The most often used five metacognitive strategies were: preparing to concentrate, trying to keep up with the speed, being aware of the attention, reading the four options in advance, and evaluating the understanding. On the other hand, the five most frequently adopted cognitive strategies were: finding key words, predicting fro m the options, comprehending texts by knowledge about English, guessing unknown words by context, and putting the details together to comprehend texts. In addition, listening strategies used by most students in the present study also include constant listening practice, listening for words or phrases that occurred in the options, getting involved in contexts, closing eyes for concentration, paying attention to the second speaker, noticing the speakers' tone and intonation, deleting the impossible answers, and selecting the best answers by logic and synonyms.
Regarding strategy differences in proficiency levels, the interviewee results indicated that listeners with high-level proficiency used a wider range of strategies more frequently than lower-level ones did, especially thinking only in English and finding key words. Selective attention (i.e., reading the four options before listening) was applied more often by high-proficiency listeners than by low-p roficiency counterparts. However, translation strategy was more frequently used by listeners with low-level proficiency. After listening, the MICE-major university students more frequently employed two strategies, recalling what was heard and crossing out the impossible answers.

Pedagogical Implications
Based on the findings of the present study, some pedagogical imp licat ions are presented. First, in MICE courses, teachers could offer a series of exercises for practicing listening sub-skills, such as distinguishing word boundaries, identify ing stress patterns, recognizing reduced forms in speech (Brown, 2001). In addit ion, Goh (2000) claimed that regular perception practice might be good for learners to familiarize themselves with the features of the target language [34]. Moreover, in order to have students get used to real-life listening outside the classroom through training, the listening activities had better to be as realistic as possible. Authenticity, mean ingfulness, interaction, and interest are all features of good listening activities in the MICE field. Furthermore, adopting appropriate listening strategies (e.g., looking for keywords, predicting the speaker's intentions, and grasping the main ideas) would enhance listeners' comprehension in the listening process. As a result, it is essential for teachers to equip students with different sorts of listening strategies useful for them to perform better in the future MICE business.
Second, since the three types of test item in this study could be matched with Bloo m's taxonomy on educational objectives (i.e., co mprehension, applicat ion, and analysis), language teachers should use test items to measure what objectives students have achieved, and thus realize what their learning difficult ies might exist so as to provide them with extra aids. In fact, performance on the listening comprehension test is a more valid and stronger indicator of one's communicat ive competence which could even simu late real co mmunication in a MICE wo rkp lace and define one's language level .
Third, it is crucial for language teachers to notice the potential co mplexity and difficu lty of higher cognitive domains imposed on the students with low-level proficiency when they take the listening comprehension test. Language teachers should track their simple curriculu m goals to complicated ones as they could guide their students toward development of MICE communicative competence.
Finally, when EFL learners show deficient L2 linguistic knowledge and face a sequence of strange L2 sounds, they would need additional help to co mpensate for their inadequate L2 linguistic knowledge if teachers could provide them with well-developed background knowledge and well-designed visual aids. To sum up, EFL teachers could take advantage of some realia and contextual p ictures to help students upgrade their listening comprehension so as to enhance their aural and oral co mmun ication proficiency in MICE industries.