Investigation into Learning Strategies Used by Effective and Less Effective EFL Learners in Korea

This study investigates whether or not effective learners used learning strategies differently from less effective learners in quantity as well as in quality. A total of 164 tertiary-level students learning English in Korea participated in the quantitative study, and eight students out of 164, four effective learners and four less effective earners, were chosen in the follow-up qualitative study. The quantitative study showed that effective learners used more learning strategies than less effective learners with the statistically significant but low correlation of .335 between learning strategy use and English proficiency. The qualitative study demonstrated that both effective and less effective learners, in general, actively used listening strategies to aid in listening comprehension, showing similarities as well as differences in strategy use. However, effective learners used their listening strategies more appropriately than less effective learners due to linguistic, cognitive, and affective variables which facilitated listening comprehension by intricately interacting with listening strategies. This study concludes with teaching implications and future study areas to verify these findings by different learners in different learning contexts.


Introduction
Traditionally, English teaching in Asia focused on teaching language skills, especially reading skills, grammar, and translation due to the dominant grammar-translation method.However, with the result of the paradigm shift in general education from teaching subject matter knowledge to teaching "learning how to learn," language learning strategies have been a crucial issue in the domain of L2 acquisition and teaching in Asia in general and in Korea in particular.
Since the trailblazing research on the "good language learner" studies (Rubin, 1975;Naiman et al., 1978), research on language learning strategies has burgeoned in several specific areas.Among the areas is whether effective language learners use more and different learning strategies than less effective language learners.In theory, it has been assumed that effective language learners use more learning strategies than less effective language learners in quantity and that the former use different learning strategies than the latter in quality.
However, these theoretical assumptions are still waiting to be tested empirically after more than 30 years of the good language learner studies.In quantitative studies, the studies to date have shown generally statistically significant but low correlations between strategy use and L2 proficiency (Griffiths, 2003;Nisbet et al., 2005;Phakiti, 2003).In qualitative studies, effective learners have not always used different learning strategies than less effective learners, depending on different learners in different learning contexts (Gan et al., 2004;O'Malley et al., 1989;Vandergrift, 1997).
The purposes of this study were to investigate whether effective learners use quantitatively more learning strategies than less effective learners in learning English in general, to investigate whether effective learners use qualitatively different learning strategies than less effective learners in listening to English in particular, and to attempt to explain the quantitative results by incorporating a quantitative study into a qualitative one.For these purposes, the following research questions were addressed to guide this study: (1) Do effective learners use quantitatively more learning strategies than less effective learners in English proficiency in general?
(2) Do effective learners use qualitatively different learning strategies than less effective learners in English listening comprehension in particular?
However, in spite of this assumption that effective language learners use quantitatively more learning strategies than less effective language learners, the relation of strategy use to L2 proficiency has been in general low with the correlation range from .01 to .47 (Griffiths, 2003;Mullins, 1992;Nisbet et al., 2005;Phakiti, 2003;Park, 1997;Politzer & McGroarty, 1985).For instance, Politzer and McGroarty (1985) investigated the correlations of different behaviors to proficiency gains and found that the correlations between behaviors and proficiency gains were ranged from .10 to .37.These low correlations of strategy use to L2 proficiency were also found in Asian contexts in which Mullins (1992) found that all the correlations of learning strategy use to various English scores failed to reach a significance level (r=.08-.18).
The low correlations between learning strategy use and L2 proficiency in the aforementioned studies might be because the advanced learners automatized their learning strategies and could not report them (Faerch & Kasper, 1983), as evidenced by Phillips (1991) and Hong-Nam and Leavell (2006).They found that the mid-proficiency group of ESL adults in the USA used more learning strategies than the higher and lower proficiency groups, showing a curvilinear relationship between learning strategy use and L2 proficiency.However, Park (1997) found that the relation of learning strategy use to the Test of English as a Foreign Language (TOEFL) by EFL university students was linear, indicating that the high proficiency group used more learning strategies than the mid-proficiency and the low proficiency groups.
Investigation into the relation of strategy use to L2 proficiency has been on the track until recently among various learners in various learning contexts.Griffiths (2003) investigated this relation by ESL learners of various nationalities in New Zealand and found that the correlation between learning strategies and course level was significant (r=.29).Phakiti (2003) also found the significant correlation between cognitive and metacognitive strategy use and EFL reading achievement test (r=.391 and .469, respectively) by EFL university students in Thailand.More recently, Nisbet et al. (2005) found that the correlation of Chinese university students' use of learning strategies to the TOEFL scores failed to reach a significance level (r=.109).
The reason for the lower-than-expected correlation of learning strategy use to L2 proficiency can be explained by closely looking into data in qualitative studies.The earlier good language learner studies utilized qualitative data through observations, diaries, or interviews (Rubin, 1975;Naiman et al., 1978).These studies imply that compared with effective learners, less effective learners use learning strategies inactively or inappropriately.In order to test this implication, several studies have been undertaken to systematically investigate learning strategies used by both effective and less effective learners (Gan et al., 2004;O'Malley et al., 1989;Porte, 1988;Vandergrift, 1997;Vann & Abraham, 1990).
In the vanguard of interest in less effective learners' strategy use, Porte (1988) found that contrary to common beliefs and hunches, less effective learners shared similar learning strategies found in the good language learner studies.However, the difference between the two groups of learners was that less effective learners used less sophisticated and less suitable learning strategies than effective learners, a finding consistent with those of subsequent studies (O'Malley at al., 1989;Vandergrift, 1997;Vann & Abraham, 1990).Vann and Abraham (1990), for instance, found that less effective ESL learners in the USA used learning strategies actively and that their strategy repertoire was similar to those of effective ESL learners.However, the problem of less effective learners was that they often used learning strategies inappropriately to carry out their learning tasks at hand.The inappropriate use of learning strategies by less effective learners was also found in other studies (O'Malley et al., 1989;Vandergrift, 1997).In a seminal research on what learners do in listening tasks, O'Malley et al. (1989) investigated listening comprehension processes and listening comprehension strategy differences between effective and less effective listeners.They found that both effective and less effective listeners used listening strategies actively and similarly in general, although effective listeners used self-monitoring, elaboration, and inferencing more appropriately than their counterparts.
One of the most interesting findings in this regard was made by Vandergrift (1997) who called attention to the importance of investigation into learning strategies by combining both quantitative and qualitative data.Using the data reported by learners of French in Canada, he found that, in general, intermediate listeners used more metacognitive strategies than novice listeners.However, the reverse held true in cognitive strategies.That is, novice listeners generally used more cognitive strategies than intermediate listeners, a crucial finding that provides reason for the low correlation between strategy use and L2 proficiency in quantitative studies.
Most studies described so far were conducted by either a quantitative approach or a qualitative one with a few exceptions of utilizing both approaches (Phakiti, 2003;Vandergrift, 1997).Since each approach has its advantage, combining the two approaches in one study will render insightful explanation about quantitative results from qualitative perspectives (Johnson & Onwuegbuzie, 2004).

Participants
The participants were 164 university students in Korea who took the Test of English for International Communication (TOEIC) and agreed to participate in this study.The students consisted of 64 juniors and 100 seniors, with 87 male students and 77 female students.Their ages were ranged between 21 and 36 with the average age of 24.They were majoring in various academic domains, with 47 humanities, 41 social sciences, 16 natural sciences, and 60 engineering.The participants had learned English for six years in secondary schools as a required course and continued to learn English in college.

Instrument
3.2.1The Strategy Inventory for Language Learning (SILL) Oxford (1990) developed a spearheading questionnaire of the Strategy Inventory for Language Learning (SILL) to measure the participants' quantitative use of language learning strategies.It consists of six strategy categories: memory strategies for remembering and retrieving new information; cognitive strategies for comprehending and producing a language; compensation strategies for overcoming a knowledge gap, metacognitive strategies for coordinating learning, affective strategies for regulating emotions, and social strategies for cooperative learning with others.The SILL is a five-point Likert-scale questionnaire from (1) never or almost never true of me to (5) always or almost always true of me.The SILL has been widely used around the world to measure learning strategy use, and the reliability and validity of the SILL were reported high (Oxford & Burry-Stock, 1995).The internal consistency reliability of the SILL used in this study determined by the Cronbach's alpha was .923.The SILL was translated into the native language of the participants in order to minimize any possible bias resulting from their comprehension of each item.

Test of English for International Communication (TOEIC)
The participants' English proficiency was determined by the Test of English for International Communication (TOEIC) developed by the Chauncey Group International (2002).The TOEIC consists of listening and reading comprehension sections with score range from 10 to 990 (5 to 495 each section), in increments of 5 points.The internal consistency reliability of TOEIC was reported 0.96, and the standard error of measurement was ±25 scale points in each section.The criterion validity of TOEIC measured by the relationship between the TOEIC scores and the Language Proficiency Interview (LPI) was reported .74(The Chauncey Group International, 2002).

Data Collection and Analysis
One data, the TOEIC scores, was obtained by utilizing a special university program designed to help students improve their English proficiency by funding for those who took a certified TOEIC in the fall semester of 2006.After taking the TOEIC, the students were required to report their scores to the university.The current author asked the university to release the TOEIC scores of those who agreed to participate in this study, and used the TOEIC scores as data to determine their English proficiency.
Another data, the SILL scores, was administered by a research assistant, a graduate student majoring in English education, either in groups or individually for one week when the participants came to an office to register for the TOEIC.Before administering the SILL, the research assistant briefly explained what the SILL was about and how to respond each item of the SILL.Then, he asked them to respond the SILL sincerely and honestly and encouraged them to question any items beyond their comprehension.It took about 25 minutes to administer the SILL.The data was analyzed using SPSS 14 for windows.

Findings
The participants' English proficiency measured by the TOEIC scores and their strategy use determined by the SILL scores averaged 502 out of 990 and 3.08 out of 5.00, respectively.The TOEIC score, 502, approximately matches with the "1+" level described by The American Foreign Service Institute (FSI) (The Chauncey Group International, 2002), and the SILL score, 3.08, can be interpreted as "medium" strategy users (Oxford, 1990).
In order to answer Research Question 1, the participants were divided into four subgroups on the basis of their TOEIC scores: the least effective group comparable to FSI level 1 (Group 1), the less effective group comparable to FSI level 1+ (Group 2), the effective group comparable to FSI level 2 (Group 3), and the most effective group comparable to FSI level 2+ (Group 4).The TOEIC scores of each group are shown in Table 1, followed by the schematic relation of the SILL scores to the TOEIC scores in Figure 1.ANOVA result showed that relatively more effective learners (Group 3 and Group 4) used significantly more learning strategies than relatively less effective learners (Group 1 and Group 2) [F(3, 156) = 6.94; p < .001],with the significant correlation of .335 between the SILL scores and the TOEIC scores (p < .01).In the follow-up post-hoc Tukey test, however, only the strategy use between the less effective group (Group 2) and the effective group (Group 3) reached a significance level (p < .05).It should also be noted that the relationship between the SILL scores and the TOEIC scores was linear rather than curvilinear, as shown in Figure 1.

Participants
The participants in the qualitative study were selected from the participants in the quantitative study who agreed to participate in the qualitative study as well.They were in the humanities consisting of four effective learners and four less effective learners, as determined by the TOEIC scores.The TOEIC scores of effective learners were ranged between 710 and 820 (the FSI level 2 or 2+) in total scores and between 365-440 in listening comprehension scores, whereas the TOEIC of less effective learners were ranged between 285 and 375 (the FSI level 1) in total scores and 155-195 in listening comprehension scores (The Chauncey Group International, 2002).A Mann-Whitney test showed that both the TOEIC total scores and the TOEIC listening comprehension scores of effective learners were significantly higher than those of less effective learners (p < .05).

Instrument
Two passages with one relatively shorter and the other relatively longer were chosen to investigate whether effective learners use qualitatively different learning strategies than less effective learners in English listening comprehension.The shorter passage was excerpted from the practice version of the TOEIC (Si-sa, 1994).It was about "Lead Poisoning" and consisted of a total of 8 sentences and 112 words with 14 words per each sentence.The longer passage was excerpted from the practice version of the TOEFL test (Educational Testing Service, 1998).It was about "Prehistoric People of the Desert" and consisted of 14 sentences and 194 words with about 14 words per each sentence.Both passages were tape recorded by male native speakers of English with the speech rate of about 165 words per minute.

Data Collection and Analysis
The author contacted the participants of the qualitative study and asked for their cooperation with this study.After assuring their cooperation, the author collected the data on an individual basis for a two week period.The data consisted of two different sessions: the training session and the main session.In the first week of the think-aloud training session designed to help the participants understand the think-aloud procedure, the author asked each participant to think aloud in answering three simple math questions, describing two pictures, and listening to an English passage excerpted from the practice version of TOEFL (Educational Testing Service, 1998).After this training procedure which continued for about 60 minutes, all the participants were told to visit the author's office for the main study at the time set for each participant in one week.
In the second week of the main session, the data were collected through the following three stages as per the recommendation of O' Malley et al. (1989) and Vandergrift (1997): the pre-collection stage to relieve the students' anxiety by having a light conversation, the in-collection stage to collect the data, and the post-collection stage to report any missing data.During the in-collection stage, the students were told that the goal of listening comprehension was to summarize and answer two multiple-choice questions after listening to each passage.To reach the goal of listening, they were advised to think aloud in Korean to elicit as many and accurate data as possible with the focus on comprehension processes, that is, how they comprehended words, phrases, and sentences using their linguistic and background knowledge and listening strategies.Apart from this goal, the students were encouraged to report whatever came to their mind whether or not it was crucial to or interfering with their comprehension of the passages.There were pauses in every one to two sentences in each passage in order to help the students to think aloud normally, to elicit as many listening strategies as possible, and to ask questions at points where the students did not think aloud clearly or where the author wanted to elaborate what the students verbalized.The relatively short passage was used first, followed by the longer one in order for the students to build confidence in this listening comprehension task and in the think-aloud procedures.All the think-aloud processes were tape-recorded, transcribed, and analyzed by the researcher with the cooperation of the research assistant who collected the quantitative data.It took about 25 to 35 minutes to think aloud each passage.

Findings
The findings of the qualitative study were described with the focus on the three categories of learning strategies: Metacognitive strategies such as attention and monitoring; cognitive strategies including prediction, parsing/translation, inferencing, and elaboration; and affective strategies such as anxiety/confidence.It should be noted that both effective learners (E1-E4) and less effective learners (I1-I4) often used these strategies concurrently, say monitoring and elaboration, parsing and translation, and prediction and inferencing, in order to accomplish listening comprehension goals.

Attention
Attention is a fundamental step to perceiving and retaining information on a text.Effective learners paid close attention to what they listened to without being interfered with unexpected noise around the interview site.For instance, E3 did not notice the intermittent noise coming from the construction site nearby while he was listening to the English passages.In addition, effective learners were able to direct their attention to important listening points.
(Interviewer: What do you do when you could not listen to a sentence?)E1: I direct my attention to the sentence I am about to hear rather than thinking of what I could not listen to previously.
The reverse was true for less effective learners.That is, less effective learners did not pay attention to their listening stimuli as well as effective learners did, often wandering aimlessly from their listening comprehension and giving up their listening task easily.This was true especially in a difficult sentence, as in these examples: I1: Hmm, area and I? I could not comprehend what I listened to because it spoke too fast and because I thought of eating rice cakes with my friends after this interview.I2: The sentence was too long and hard to understand, and I gave up further listening.

Monitoring
Monitoring is a strategy pertaining to checking one's understanding and misunderstanding while listening and tracking the sources of it.Both effective learners and less effective learners checked their understanding on the basis of variables including linguistic knowledge and background knowledge about the passages.In addition, both groups of learners were able to articulate their comprehension problems while dealing with the passages.
Both E4 and I4, as in the examples below, knew the outcome when they utilized inaccurate background knowledge about a text.E3 focused on words and he was also aware of the negative side of focusing too much on words rather than on meaning, whereas I1 reported the importance of grammatical knowledge to analyze the sentence and pronunciation in listening comprehension.
Effective learners: E3: I tend to focus too much on words in listening comprehension, which, I think, interferes with the understanding of the whole passage.

Prediction
A psychological theory of listening contends that listeners are assumed to be active in predicting and revising what they hear based on various types of cues.This assumption was confirmed by both effective and less effective learners.However, what differentiated the two groups of learners was the accurateness of predictions.That is, effective learners predicted the text more accurately than less effective learners did probably because the former had more cues such as words and pronunciation available than the latter.E1, E2, and I4 below predicted the coming sentences accurately based on key words such as 'sorry' and 'where.'However, the prediction of I3 was wrong because of misunderstanding of previous sentences.

Effective learners:
E1: She predicted what she would hear based on a key word she heard, as in "When I heard 'sorry,' I focused on the reason for sorry [in the coming sentences.]"E2: (Listening Text: They actually followed a series of carefully planned moves.Where they moved depended on where food was available-places where plants were ripening or fish were spawning.)As I expected, they did not settle in a place.Then, the following sentences will explain why and where they were from.

Less effective learners:
I3: (Listening Text: Even though the cave's very large, it was certainly too dark and dusty for the travelers to live in.But it was a great place to hide things, and tremendous amounts of food supplies and artifacts have been found there.)Since I listened to 'but' and 'high,' (He misunderstood 'hide' for 'high') and since the sentence described 'darkness,' the next phrase after 'but' will be something about 'low' and 'brightness.'I4: Since I listened to 'where,' it might have explained where they moved to.Since it was about movement, the next may explain why and when.

Parsing/translation
Parsing is concerned with breaking down a sentence into smaller units such as words and phrases when necessary, and translation occurs when thinking of the meaning of L2 words in L1.Listeners segment a sentence into smaller units on the basis of their grammatical knowledge and meaningful interpretation.To better understand the listening process with regard to parsing, listeners were asked to answer what they focused on when they heard a sentence.In general, effective learners comprehended a whole sentence at a time and broke down a sentence into smaller units when the tasks were challenging: E1: Since I've heard this type of sentence several times, I comprehended the whole sentence at a time.(Interviewer: What did you do to comprehend the sentence?)I just comprehended the sentence without doing anything.
E2: When a sentence is easy to understand, I listen to the sentence as a whole.However, when a sentence is hard to understand, I focus on words and call upon the background knowledge about the sentence.E4: I focused on words because this sentence was getting longer.(Interviewer: What about a short sentence?)When a sentence is short, I absorb the whole sentence as it is.
On the contrary, less effective learners listened to words, and often translated the words into Korean while listening.Even though some less effective learners wanted to listen to beyond a word level, they couldn't because they failed to process the meaning of words or phrases they heard automatically.When they think of the meaning of the words and phrases in Korean while they are listening, they can't attend to the next words and phrases, ending up with the misunderstanding of what they listen to, as reported in I1 and I3.I1: I try to listen to a whole sentence.However, when I focus on the first part of a sentence and think of the meaning of it in Korean, I miss the next part.I3: When I hear a word, I try to think of the meaning of it in Korean.The problem is that while I am thinking of the meaning of the word in Korean, speech which I do not attend to keeps flowing.Then, I hear another word and think of the meaning of it in Korean again, [while speech keeps flowing.When this process is repeated,] I end up with storing up scattered words failing to connect them meaningfully.
Interestingly, three out of the four less effective listeners complained of their shortage of memory ability because they did not remember what they understood while they were listening.This might be because they focused on words rather than meaning in listening comprehension or because they had limited cognitive ability left to remember what they heard.I1: When I take the TOEIC, I forget what I heard about a text when I answer the items.I3: I am frustrated because even though I think I understood the sentence at the time I was listening, I forget it while reporting.I4: When I listen to the first part of a sentence, I retain words.The problem is that when I continue to listen to the next part, I forget the first part of a sentence I heard before.

Inferencing
Inferencing is a strategy used by listeners to guess difficult words or fill in missing information.In general, this strategy was used more by less effective listeners than effective listeners because the former encountered more unfamiliar words, phrases, and sentences than the latter.However, effective listeners used inferencing to their advantage, whereas less effective listeners did not take advantage of it, guessing wildly based on a few words they could listen.For instance, E3 inferred what he missed out of context correctly.However, I3 wildly guessed the meaning of a sentence by the words "late" and "sorry."I4 inferred wrong by misunderstanding wandering for wonder, as in the examples: Effective listener: E3: (Listening Text: You can see their size in relation to the hands holding them.)I could not comprehend all the words.However, since the previous sentences were about "finding foods, instruments, and other things," I guessed that this sentence was about finding artifacts and showing them to the listeners.

Less effective listeners:
I3: (Listening Text: I am really sorry that this was brought to my attention so late.)Since I listened to 'late' and 'sorry,' I guess it is about apologizing for being late.I4: (Listening Text: You might think that they were wandering aimlessly-far from it!)I guess this sentence describes something which is full of wonder.

Elaboration
Elaboration in listening comprehension pertains to relating listening stimuli to existing knowledge in long-term memory.Both effective and less effective listeners used elaboration, as reported by E1, E4, and I4.However, compared with E1 and E4 who made use of elaboration to comprehend a text, I4 misused it and composed a different story, as in the following examples: Effective listeners: E1: I think that this passage is similar to the passage in the TOEFL writing test I took before, which was about residents' opinions on building a factory nearby the village.E4: Well, they move with their belongings and equipment for foods.(Interviewer: Did you think of anything else while you are listening?)I thought of nomads moving from places to places.

Less effective listener:
I4: (Listening text: But it was a great place to hide things, and tremendous amounts of food supplies and artifacts have been found there.The food includes dried fish, seeds, and nuts.The artifacts include stone spear points and knives; the spear points are actually rather small.Here's a picture of some that were found.You can see their size in relation to the hands holding them.)'Movement' and 'foot' in this sentence hint that this sentence is concerned with movement, foot, and medical perspectives…… It is about the size of a foot…… In this sentence, a person shows a picture of prehistoric people's feet…… It also describes the size of feet followed by hands.It seems that the text will describe their hands from now on.

Anxiety/Confidence
Anxiety plays an important role in L2 listening because when listeners are apprehensive or worried too much, their cognitive system and affective system do not function normally and interfere with listening comprehension.
When asked to answer how they feel when they are listening, all the four effective listeners answered that they were a little bit worried, but not too much.They also reported that when they are worried in a test situation, they use listening strategies such as trying to relax and improving attention.For instance, E4 reported that 'This sentence is getting longer.Thus, I am a little bit worried for a moment but decided to pay more attention to the next sentence.'Compared with effective listeners, two less effective listeners answered 'a little bit worried' and the other two answered 'worried much.'The less effective listeners used various listening strategies to deal with anxiety such as trying to improve attention, stretch and take a deep breath, and close eyes and relax.For instance, I4 reported that he closes his eyes and tries to relax when he is worried too much.
It should be noted that anxiety was intricately intertwined with confidence because the anxiety of those two less effective listeners who answered 'worried much' was lack of confidence in listening comprehension.One the two worried listeners, I2 was not confident throughout the listening task.For instance, she reported that 'I can't comprehend unfamiliar words.I comprehend only the words that I know such as "ground."This is a short sentence.Nevertheless, my comprehension is wrong, isn't it?'In contrast, E3 showed confidence in this task.He reported that 'even though I didn't comprehend this sentence, I think I can comprehend it if I could listen to it one more time.'

Discussion
The finding of the quantitative study was that the correlation of students' use of English learning strategies to English proficiency was statistically significant but low (r=.335).The significant correlation between learning strategy use and L2 proficiency provides new insight into the conflicting findings (Griffiths, 2003;Mullins, 1992;Nisbet et al., 2005;Park, 1997;Phakiti, 2003).For instance, Nisbet et al. (2005) reported the correlation between these two variables as .109which is very low, whereas Park (1997) reported as .34.
The low correlation of learning strategies to English proficiency can be attributed to several points.First, there are lots of variables, which account for L2 acquisition, other than learning strategies such as linguistic knowledge, background knowledge, motivation, and acculturation (Ellis, 1994;Horwitz, 2008;Park, 2004).So, there will be limitation for one variable, say learning strategies, to account for L2 proficiency.Second, since the classical test theory posits that observed scores consist of true scores and error scores, the observed scores measured by the SILL and TOEIC might not be consistent and trustworthy (Oxford & Burry-Stock, 1995;The Chauncey Group International, 2002).Third, learning strategies are supposed to be used to attain goals or solve problems.Even though effective learners, in general, used more learning strategies than less effective learners to attain goals, effective learners would not use learning strategies when problems did not occur in performing tasks which were not challenging enough to use specific strategies.Fourth, the low correlation might be due to effective learners' automaticity of learning strategies and failure to report the strategies they used subconsciously (Cohen & Macaro, 2007;Faerch & Kasper, 1983).The fifth reason might be that the use of English learning strategies in quantity alone could not account for English proficiency because effective English learners might use learning strategies differently from less effective learners in quality (Vandergrift, 1999;Vann & Abraham, 1990).
All the five speculations above might cause the low correlation of learning strategy use to L2 proficiency one way or another.Among the five, the fifth speculation was related to the findings of the follow-up qualitative study in which both effective learners and less effective learners actively used listening strategies to enhance listening comprehension such as metacognitive strategies, cognitive strategies, and affective strategies.However, effective learners used these strategies more appropriately than less effective learners due to other variables such as linguistic, cognitive, and affective variables which work interactively with listening strategies in performing tasks.In this regard, learning strategies might be multidimensional even though they were categorized into a unidimensional construct of strategic competence in the current literature of learning strategies (Cohen & Macaro, 2007;Oxford, 1990).
In metacognitive strategies, contrary to previous findings (Phakiti, 2003;Vandergrift, 1997), both effective and less effective learners actively used monitoring strategies in order to check their listening comprehension drawing on linguistic and background knowledge.However, a conspicuous difference between the groups of learners was found in attention to listening.That is, effective learners paid and directed attention to listening stimuli, whereas less effective learners were often interfered with unexpected noise and wandered from their listening comprehension (O'Malley et al., 1989).Less effective learners' attention problem might result from their helplessness in the comprehension of the passages after trying hard to listen to the passage, but in vain.
In cognitive strategies, both effective learners and less effective learners actively employed listening strategies in order to make sense of the oral passages, but effective learners used these strategies to their benefits.For instance, both effective learners and less effective learners used cognitive strategies such as prediction, inferencing, and elaboration.What differentiated the two groups of learners, however, was that effective learners used these strategies more successfully than less effective learners did.That is, effective learners predicted the next passage, guessed unfamiliar words out of context, and used their background knowledge accurately, whereas less effective learners used these listening strategies inaccurately and wildly (O'Malley et al., 1989;Vandergrift, 1997).This might be because less effective learners, compared with effective learners, were lacking linguistic knowledge and had problems recognizing words, phrases, and sentences which, in turn, made them draw too much on the background knowledge about the passages to compensate for the lack of linguistic knowledge.Caution should be exercised because even though less effective or novice learners embedded in extracting meaning of individual words, they used interactive strategies utilizing both word-based bottom-up strategies and schema-based top-down strategies.However, compared with less effective learners, effective learners took advantage of their listening strategies in performing their tasks due to linguistic, cognitive, and affective variables.
In addition, both effective and less effective learners tried to comprehend a whole sentence and parsed a sentence into smaller chunks when a sentence was long and hard to understand.It was the size of these units that differentiated effective learners from less effective learners.That is, effective learners segmented sentences into bigger units than less effective learners did, a finding also noted by previous studies (O 'Malley et al., 1989;Vandergrift, 1997).Compared with effective learners who were able to process textual units such as words, phrases, and sentences automatically, less effective learners had problems in processing these textual units, tended to fixate their attention to words, and translated the words into L1 in a controlled way.This caused their cognitive processing capacity and memory ability about what they had heard to be limited.
In affective strategies, both groups of learners felt anxiety in listening comprehension, especially in a test situation, and shared similar listening strategies such as 'trying to relax' in order to relieve their anxiety.In the extent of anxiety level, however, less effective learners were, in general, more worried about listening comprehension than effective learners.In addition, it was found that the anxiety level was reversely related to the confidence level because effective learners were more confident and less worried in their listening comprehension than less effective learners (Gan et al., 2004).It is noteworthy that when L2 learners are worried too much and have low confidence in performing listening tasks, their cognitive system and affective system do not function normally and block their comprehension tasks.Accordingly, the low anxiety level combined with the high confidence level held by effective learners might aid better in the listening comprehension tasks than their counterparts who were more worried and less confident in the tasks.

Conclusion
Paralleling with paradigm shift from teaching to leaning how to learn in general education, language learning strategies have been a crucial issue around the world.In order to meet this demand, the current study investigated whether or not effective learners used learning strategies differently from less effective learners in quantity as well as in quality and found several important findings.
The findings of this study provide implications for teaching a second language in two areas.First, since effective learners used more learning strategies than less effective learners, strategy training should be implemented in a language classroom to help learners become autonomous both in and outside of the classroom (Thompson & Rubin, 1996;Vandergrift, 1999).However, teachers still focus on language training ignoring strategy training in mainstream foreign language classes across academia because learning strategies are still beyond the concern of many teachers, textbook authors, and administrators.Thus, the findings of this study will sensitize educational administrators, textbook authors, English teachers and learners around the world to the importance of learning and teaching strategies in L2 acquisition.Second, since effective learners used various learning strategies to their benefits due to other variables, teachers should keep these variables in mind in strategy training in order to maximize training effects.These variables are countless but can be categorized into linguistic, cognitive, and affective variables, as have been centered in second/foreign language learning and teaching.Drawing on these tentative findings, teachers should fine tune their model of learning strategy training that is compatible with instructional goals, students' learning styles and proficiency levels, and learning settings.
Even though several important findings were made in this study, this study is not immune to criticism in generalization because it was undertaken with limited numbers of participants learning English in college level in Korea and because there are various data collection methods such as observation and interviews other than self-report data.Thus, more studies should be conducted among different learners in different learning contexts using different data collection methods to replicate the findings of this study and to shed further light on the strategy differences used by effective and less effective learners.This study as well as future studies on learning strategies will contribute to more effective second language learning and teaching, which in turn will lead to a E4: Did I understand the sentences accurately?It seems that I sometimes guess incorrectly what I could not understand out of my background knowledge, leading to a total misunderstanding of the text like writing a novel …… When I make a novel, I make two to three wrong answers out of the three questions [about a passage in the TOEIC].Less effective learners: I1: This sentence was hard to understand because the sentence was too long to analyze [grammatically], and I was not used to English pronunciation.I4: I thought of prehistoric people carrying around stones and something else [while listening].(Interviewer: Does the thought or image help your listening comprehension?)It facilitates or interferes with listening comprehension.(Interviewer: Can you tell me more specifically?)When a sentence is hard to understand, I focus on words and draw images about what I hear.The problem is that when I draw a different picture, it interferes with listening comprehension.

Figure 1 .
Figure 1.The Relationship between Learning Strategy Use and L2 Proficiency