Jacqueline D. Beebe

When I meet Japanese individuals who are uncommonly good speakers of English I wonder how they developed their skill; if their schools or outside lessons were providing some special learning opportunities or if they were doing something special on their own. I also wonder why they got so good; if they had some special motivation or attitudes that spurred them on and supported them. My personal curiosity as a teacher led me to look in detail at the English learning careers of individuals.

This study is in the tradition of good language learner (GLL) studies (e.g. Rubin, 1975) that seek to eventually improve the learning of other less successful learners by finding out what the successful learners are doing right. I was influenced by the kind of questions asked in interviews of GLLs (especially in Naiman, Frölich, Stern, and Todesco, 1978) and by knowledge of factors often identified in GLL studies. For example, Ellis (1994, p. 546) surveys eight studies done between 1975 and 1989 and finds “(1) a concern for language form, (2) a concern for communication (functional practice), (3) an active task approach, (4) an awareness of the learning process, and (5) a capacity to use strategies flexibly in accordance with task requirements.”

This study also examines several factors which fall into the broad area of individual learner differences (IDs), where one finds studies of aptitude and other cognitive factors, age, learning style, personality, strategies, beliefs, motivation, and affective states such as anxiety. Skehan (1989) comprehensively examined the ID field, and recently looked again at learning styles (1998), while a recent general review of IDs is to be found in Segalowitz (1997). Researchers in Japan have surveyed learners’ beliefs and attitudes (e.g. Redfield,1995; Gobel, 1996; Gaies and Sakui, 1998), strategies (e.g., Nunnelley, 1993; Watanabe, 1992; Willis, 1996), motivation or learning orientation (e.g., Berwick and Ross, 1989; Koizumi and Matsuo, 1993; Johnson and Takeo, 1996; Yamashiro, 1998), and preferred learning styles (e.g., Hyland, 1994; Ishikawa, 1996; Yamashita, 1996).

Taken together, these and other second language acquisition (SLA) surveys of Japanese learners make plain the dissonance between what learners believe, what and how they want to learn, and the learning opportunities they find and perhaps exploit at their schools. However, when each of these types of ID factors are isolated and are examined in separate studies with a different population of learners, it is not easy to know how factors affect each other. One can only guess, for example, that teachers could tap into motivation A to encourage learners to engage in strategy B. Furthermore, the individual is lost sight of in questionnaire-based surveys of individual differences whose overly general questions do not provide a rounded picture of a specific learner acting for a specific purpose in a specific context.

This study therefore looks at IDs in context through a large case study. While I did not observe classrooms, I asked informants about each of their teachers’ classes in all their years of mainstream and supplementary English lessons. I looked at the particularities of how, for each informant, individual learner factors changed over time and in different circumstances inside and outside of school. In Beebe (1994) I found that seven female Japanese high school students who spoke English exceptionally well had much to tell about how and why they independently learned to speak English. In this current study I included both females and males, those with high and low oral proficiency, and students from a variety of schools. The heart of this study is systematic but flexible interviews concerning a wide range of learning factors.

The research questions addressed in this article are (a) What learning opportunities and individual learner differences (IDs) are associated with the successful development of oral proficiency in English as a foreign language by Japanese high school students? and (b) What can be learned by selecting and quantifying qualitatively-derived factors which vary between students with high and low oral proficiency?


Participants and Research Sites

 I focused on third year high school students, aged seventeen and eighteen. I met with forty-three students (“participants”) and eventually chose eighteen focal students  (“informants”) for my in-depth study. Informants came from eleven different schools, both in the Tokyo metropolitan area and some mid-sized cities several hours beyond commuting distance to Tokyo. The schools were academically-oriented schools from which most graduates continued on to two-year colleges or universities, as indeed all the informants planned to do. All of the informants were average to above-average students in virtually all their school subjects. While I designate five informants as Lows based on their poor speaking abilities, both High and Low informants had at least average grades in their English classes, where speaking is not tested. One High male is actually a Taiwanese Chinese who has lived in Japan since he was five and is better at Japanese than Chinese.

 I told the teachers who recruited participants for me at their schools that I wanted to interview mainly students who were exceptionally good English speakers but also some more typical students. In fact, I had a hard time finding participants who were clearly much worse than the best speakers, as apparently teachers wanted to show off their best students and these good speakers were more apt to agree to meet me. It was especially hard to recruit females who could not speak well, and the informant group is composed of eight High females and five High males and one Low female and four Low males. The longest total time that any informant had been abroad was six weeks.

Testing Oral Proficiency

I used a Story-Retelling task to select informants with High and Low oral proficiency and exclude the midrange participants by measuring the participants’ ability to (a) speak fluently, with speed and continuity; and (b) to use their knowledge of English, their speaking skills, and their memories to verbally convey the plot of a story in a comprehensible way. I tested students with a wide range of ability, so the task gave a great deal of support and made it possible to measure even very short and disjointed speech production with precision. The text came from an illustrated children’s story by Wells (1992). I shortened and simplified her story and added Japanese glossing of some words. Participants read the story and looked at the illustrations while also listening to a tape of the story being read aloud. They then had three minutes to study the story and ask questions. Then they read and heard the story once again before having to reproduce the story in either their own or the original words while turning pages of an illustration-only, no-text version of the story.

I roughly and holistically divided the participants into those with high, medium, and low speaking ability by listening to random samples of their interview and retelling tapes. The retelling tapes of potential informants—those who seemed to have the strongest and weakest speaking ability—were transcribed and scored to arrive at a final selection. Following Lennon (1990), Fluency scores were calculated by counting syllables per second, excluding disfluencies such as unfinished or repeated words and fillers such as uh. Information Conveyance scores rated both the degree to which speakers successfully communicated any of the information contained in the original text and the degree to which they communicated the most important points of the story. The story retellings of five native speakers of English (NS) were analyzed to divide the text into information units worth from one to three points depending on how many of the NSs mentioned each unit. Three NSs scored the potential informants’ transcripts, using the criteria of whether someone who was following the illustrations but did not know the story would be able to understand what the speaker meant to convey. By dividing the time taken to retell the story by the Information Conveyance score, I arrived at an Efficiency score. As an example of how the groups differed, on Efficiency, the NSs averaged .99, the Highs, .37, and the Lows, .16. T-tests of the scores of the Highs and Lows on Fluency, Information Conveyance, and Efficiency demonstrated the distinctiveness of the High and Low groups in oral ability. Details of both descriptive and inferential statistics, as well as inter-rater reliability figures are available in Beebe (1998).

The Interview Sessions

I conducted all the interviews myself, referring to a bilingual interview protocol and asking additional questions as appropriate, using English, Japanese or both in succession. All thirteen of the Highs spoke English during their interviews, with varying degrees of use of Japanese, while all but one of the five Lows did their interviews in Japanese. I am a female European-American who has taught English in Japan at both the secondary and post-secondary levels, but not at the schools the participants attended. I rate my oral-aural proficiency in Japanese at about the Advanced (the seventh highest of nine levels) on the ACTFL scale (see Clark and Clifford, 1988). The interviews were transcribed and translated into English by bilingual NSs of Japanese. Interviews ranged in time between forty minutes and over two hours.

I began audio-taping each session as I asked the participant questions about a bilingual Questionnaire which was filled out previously. It was composed of a grid with sections for the participant to fill in for elementary school and for each year of junior and senior high school (the six years when all participants studied English at their mainstream schools). Participants wrote down in either Japanese or English whether they had experienced (a) extra lessons of any kind outside of school (b) foreign teachers, English conversation classes, or an English club at school (c) contact with foreigners or returnees in Japan or a foreign penpal (d) travel abroad by the participant or family members and (e) use of any English language materials (including entertainment media) that were not assigned by a school.

Interview and Questionnaire Findings

Denzin (1978, p. 103) writes that “starting with loose sensitizing definitions of their concepts, [naturalistic researchers] empirically operationalize the concepts only after having entered the worlds*that they wish to understand.” I sifted through the interview data and arrived at thirty-two factors divided into the four categories of Schooling, Solitary Strategies, Interpersonal Strategies, and Attitudes. I considered these factors to have influenced the informants’ English learning on the basis of the informants’ comments on their practical or affective importance, my judgment of their potential for contributing to L2 oral proficiency, frequency of mention, and the extent to which the factor was stronger for either the High or the Low informants. I roughly and holistically quantified the presence and strength of each factor for each individual, judging whether the factor manifested in a strong (3), medium (2), weak (1), non-existent (0) or negative (-1) way. (See Beebe, 1998, for details.) What follows is a brief explanation and sometimes examples of each factor, with the means for Highs and Lows noted.


The English education that both the Highs and Lows received at school and in the test-preparation after-school cram schools that they had all attended was largely similar, but Highs more often encountered a teacher who supported the development of oral-aural skills. I asked the informants for each year of their schooling if their teachers ever used English to say things such as “Open your book to page...” or “Why are you late?”, and most of their teachers had not. However, somewhere along the way, the Highs were more apt to have encountered a teacher who spoke spontaneous English in the classroom, beyond reading sentences directly out of the textbook. The thirteen Highs averaged 2.2; medium strength on this factor, while the five Lows had a mean score of 1; weak (see Table 1). During my interview with one Low informant, his teacher entered the room to discuss a logistical matter with me, and after the teacher left, the informant expressed his amazement to find that his teacher could speak conversational English. He said that he had never once heard any teacher speak English.

Highs also more often had chances to speak English during their six years of mainstream school lessons; most Highs had had at least one teacher who had students do more than read an answer directly off of the page, while more than half of the Lows never once had such a chance. The Highs had a mean of 2 vs. the Lows’ mean of .6. Often such chances came from a teacher calling a student to answer a question without a textbook or homework assignment to read off of. Pair or small group work was very rare.

The Highs had more classroom time with NS English teachers than the Lows did (1.9 vs. 1). Although NS teachers working in the schools either permanently or as visitors did not always speak English or have students speak English, in general, more spontaneous English occurred in their classrooms and the informants’ interest in English was likely to be boosted.

Highs were more often chosen to represent their class or school in recitation or speech contests (1.1 vs. .6). This, of course, is at least as much a result as a cause of their English speaking ability. However, the contests occasioned intense oral production practice, and in some cases, one-to-one coaching with a teacher. Furthermore, the contestants were awarded public status as English speakers, not merely as English students, who are usually expected to remain silent, and it appears that the contests positively affected the self-identity and confidence of some informants.

Fifteen out of the eighteen informants went to outside English lessons that were not mainly aimed at raising test grades, either before junior high or during the secondary years. Only three out of thirty-two factors were stronger for the Lows than for the Highs. One such factor was exposure to English in formal lessons prior to starting English lessons at school in junior high school. The Lows scored 1.8 and the Highs 1.5. The Highs however, had more of a conversation orientation to their lessons and had more fun lessons, with games, songs, etc., which introduced them to techniques for using and studying English that they seldom encountered in their subsequent mainstream school lessons. Outside conversation lessons during the secondary years were much stronger for Highs than for Lows (2 vs. .4), both in the amount of time devoted to such non-test related lessons and in the amount of English conversation that actually occurred at so-called conversation classes.

Table 1. Learning Factors Related to Oral Proficiency

Factor / Group Mean Highs n= 13 Lows n=5
English teacher/s spoke spontaneous English
Students had spontaneous speaking chances 
Native speaker English teacher 
Recitation/speech contest participation 
English lessons prior to junior high 
Outside conversation lessons
Solitary Strategies 
Self-chosen entertainment listening materials 
Pedagogical listening 
Intense repetitive listening 
Read and/or repeated aloud 
Sang English songs 
Form-focused textbooks and reference books 
Content-focused reading 
Advice: practice with media—music, radio, movies, etc. 
Prefers to focus on meaning rather than form 
Thought in English as a practice technique
Interpersonal Strategies 
Advice: speaking practice with a partner 
Sought non-Japanese, especially NS interlocutors 
Wants to talk to non-Japanese, especially NSs 
Use of Japanese interlocutors for English practice 
Volunteered to speak English at secondary school 
Traveled abroad (used English) 
Wrote letters in English to non-Japanese 
Believes s/he will become an excellent English speaker 
Effect of failure on motivation to study (+ or -) 
Desire to master communicative, interactive English 
Integrative motivation—interest in people and culture 
Wants to live abroad 
Distancing motivation 
Future career plans requiring oral proficiency 
Specific experience sparked interest in English 
Academic study plans calling for English

Scoring Key
 3 = strong (behavior often engaged in, opinion strongly expressed, etc.)
 2 = medium or strength unknown
 1 = weak
 0 = informant did not report this factor or said that it does not apply
 -1 = the opposite is true for this informant

The pre-junior high lessons were usually instigated by the informants’ mothers. Conversation lessons in the secondary years only occurred when the informants themselves talked their parents into letting them attend these lessons that even the informants themselves felt did not contribute much to their test scores. The Lows voiced the opinion along with the Highs that being able to speak English is a good thing, but the Lows do not consider it a priority at this point in their English careers. The Lows say that they may go to a conversation school in their college years, but they believe that cram school is much more important now. More of the Highs on the other hand, refuse to give up enjoying learning to speak and listen, while they also do well in reading and writing tests.

Solitary Learning Strategies

 In this study, for want of a better term, strategy is used more broadly than in most SLA research. I classify as strategies those ways in which the informants chose to use English beyond study and homework required for school and cram school. Strategy refers to a language use activity, whether engaged in for communication, for enjoyment, or for language learning, and I include behaviors as broad as traveling abroad and as narrow as reading aloud. In a few cases I also include behaviors which the informants said they would or would not engage in if given a chance or would advise others to follow.

 One of the most striking findings of this study is the extent to which solitary exposure to and use of English seems to have contributed to the ability of the Highs to interact face-to-face in English, as demonstrated by both their scores on the Story Retelling task and their willingness and ability to conduct their interviews in English, which for some of them was their first sustained interaction in English or first time to talk to a non-Japanese. Highs made more use of self-chosen entertainment listening materials; music, movies, TV shows, and radio stations with English-speaking DJs (1.5 vs. .2). They often used such materials for fun more than as a study technique, and often used them very intensively. For example, one High watched the same Disney Beauty and the Beast  video without subtitles fifteen to twenty times, understanding fifty percent the first time and seventy or eighty percent now. Another stopped and rewound portions of an entertainment movie video in order to take dictation of the dialog or consult a dictionary. Both Highs and Lows made good use of pedagogical listening materials  (2 vs. 1.8), and some Highs listened to radio English lessons almost daily for five or six years. One High would memorize the essays on the pedagogical tape she used and recite them aloud and write them out. On especially intensive or repetitive use of either type of listening material, Highs scored 1.8 and Low .4.

 Highs spent more time repeating aloud after tapes or broadcast texts and also reading aloud (1.9 vs. 1.4). One High female likes to read so loudly that her family has to tell her to shut up. One of the biggest differences between the two groups is in singing English songs  (2.2 vs. .2). Several Highs can sing fifteen to twenty English songs by memory. Again, the Highs mainly sing for fun, although some informants checked the meaning of songs with the dictionary or wrote out a Japanese translation. Songs thus provide the informants with both linguistic input and mechanical practice that leads to oral fluency.

 In general, when studying autonomously, Lows often do more of the same sort of study they must do for assignments, while Highs adventure into techniques which they devise on their own or have learned at their extra non-academic English lessons. The Lows were more apt to use self-chosen reading materials than the Highs were. Lows scored 2.6 and Highs scored 1.8 on use of form-focused textbooks and reference books. The Lows scored 1.4 and the Highs .9 on use of content-focused materials such as magazines for English learners or the sixty Sherlock Holmes stories that a High informant read in the original unsimplified English after having read the stories in Japanese.

 Highs have more fun studying English and also advise a hypothetical teacher or student of English to use audio or visual media and games to learn English more often than Lows do (1.9 vs. .4). Highs more often than Lows said that they prefer to focus on meaning rather than on form (1.7 vs. .8).

These differences in the types of solitary techniques employed probably both contribute to and are a result of the higher oral proficiency of the Highs. For example, it was only some of the Highs who report that they sometimes think in English as a study technique (.8 vs. 0), and the Highs are more apt to use challenging authentic materials. This solitary study prepares Highs for the difficult task of conversing face-to-face in real time, while the Lows tend to freeze at that prospect. The Highs have discovered that they can tackle difficult materials because in solitude they can listen repeatedly, consult reference books, and be unobserved and therefore unjudged in their struggles. One Low informant who tried listening to the English of a video once and decided that it was too hard and that she could not help reading the subtitles. Compare her with the High informant who tapes paper over the bottom of his TV screen so he cannot read the subtitles. He says that he can understand ten percent of the English dialog the first time. He then watches the movie again, reading the Japanese subtitles, and the third time, when he watches the movie with the subtitles again covered, he can understand fifty percent.

Interpersonal Strategies

 Highs and Lows in equal numbers advise a hypothetical student to practice speaking English with someone (both 1.6), but it is the Highs who have actually taken risks to do this. As mentioned under Schooling, Highs more often went to outside conversation lessons, and they often did this as a means of speaking English with foreigners. The Highs more often sought out NS or foreign interlocutors (the informants rarely distinguish between the two) than the Lows did (2.4 vs. .6), for example, taking the initiative to strike up conversations with foreign teachers or students at their high schools or going to international exchange events. The Lows have rarely sought such chances, and most say they would probably avoid them as they “can’t speak English.” When it comes to English speaking partners, Highs prefer foreigners and Lows prefer Japanese interlocutors. The Highs scored 1.9 on wanting to speak with foreigners and the Lows scored -.6. The Highs scored 1.4 on wanting to use or having found Japanese interlocutors, while the Lows scored .6. Some informants have found one or several schoolmates to practice speaking English with informally, just for the fun of using English or talking about popular culture. This is not easy to do, as they virtually never see anyone, teachers or students, speaking English outside the classroom. The informants fear picking up incorrect English from a Japanese peer or lapsing back into Japanese, and some who have tried speaking buddies still do not recommend the practice. The Highs, however, can at least conceive of themselves as English speakers. More of them have volunteered to speak English at school at some point (1.8 vs. .2), for example, raising their hands to ask a question of a visiting foreign teacher.

 An obvious occasion for speaking English is during travel abroad, and the Highs have been abroad more and have also used English more while there (these combine as a score of 1.8 for the Highs and .8 for the Lows). The one Low who spoke English abroad, on a homestay trip, was the only Low who did his interview in English. The Highs often persuaded their parents to let them take a trip abroad and while there, would go up to strangers and speak English. One High spoke English with her Japanese roommate on a school trip to the U.S., and she says that none of the other girls from her school did so. She also made a point of talking to the Mexican students at the U.S. English language school rather than to the Japanese students.

Trips abroad, as well as penpal clubs, gave many Highs the chance to correspond in English (1.5 vs. .4). Letters offer a rare chance to compose English, as school lessons usually only call for translation and sentence transformation writing. One High wrote love letters to a woman he met on a trip in his high school years until she got married. Another High currently writes in English to three people about five times a year each.


 I use the term attitudes to cover a variety of factors having to do with affect, goals, preferences, motivation, and self-identity. They largely overlap with the ID factors that Gardner and MacIntyre (1993) group as affective variables, “those emotionally relevant characteristics of the individual that influence how she/he will respond to any situation” (p. 1). The factors I highlight as attitudes are more changeable and more easily influenced by the individual student or teacher than factors such as aptitude and personality, and they in turn can powerfully influence learning for good or ill.

In general, the Highs feel much more positive about English. More of the Highs believe that they will some day become excellent speakers of English (1.8 vs. .4), some believing it is possible to do so even without leaving Japan. For those informants who brought up a specific experience of failure with English, the effect was usually positive for Highs (.2) and usually negative for Lows (-.2). For example, some Highs spoke of finding out while abroad that their academic studies had not prepared them to communicate face-to-face and therefore being spurred on to study English conversation on returning to Japan.

 Highs more often and more strongly expressed a desire to master communicative English (2.3 vs. 1), complaining more, for example, about the emphasis on grammar at school. They also showed more of an integrative motivation (2 vs. 1.6). Expressions of such desires as to go on a homestay, to travel in order to make foreign friends, to marry a foreigner, or to learn about other cultures were rated as indications of integrative motivation. Gardner (1985), and more recently, Gardner and MacIntyre (1991) contrast an integrative motivation, directed effort toward language learning fueled by personal interest in the people and culture of the other language, with instrumental motivation, stemming from the practical advantages of learning the language, such as job opportunities. As an example of integrative motivation, since age six, one High female has been attracted to the glamour of blue-eyed blond movie stars and singers, and has wanted to learn English and French. Many informants want to live abroad for instrumental or integrative reasons, or both. Whether for work, study, or for the life style or relationships that other countries afford, more Highs than Lows want to live abroad some day (1.5 vs. .2).

 Another related motivation the Highs, but not the Lows, displayed in the interviews was a distancing motivation (1.8 vs. 0). Their unusual English ability gives these informants the means to get out of Japan on their own and also contributes to their self-image as something other than a stereotypical Japanese boy or girl with a stereotypical future. Perhaps having a self-identity as an English-speaker, with the mostly positive connotations that has in Japan, (educated, internationalized, cool, etc.) affords the informants some psychic breathing space in their current lives. The Highs do not usually mind being thought of as strange for speaking English in front of their friends. Several of the females said they want careers of more substance than being “office ladies” pouring tea for a few years before quitting, and they believe that their English skills will give them access to more interesting jobs. One says that she may never marry. Some of the males are proud of things like being Americanized, being able to talk to beautiful Australian women on the beach, or liking English pop music instead of the Japanese pop music his friends like. Gun, the Taiwanese informant, attributed his sudden increase in English study and the use of new independent study techniques during his third year of high school to his increased interest in learning English, which he said was motivated by an increased disinterest in the Japanese language and everything about Japan and his life in Japan.

 Many of the Highs have career plans or dreams which will require more conversational English ability than that obtained by most Japanese university graduates (2.4 vs. .6). One could say that these informants have an instrumental motivation, however, the age of the informants and the glamorous nature of some of their desired jobs suggests that their choice concerns more than earning a paycheck, and is related to a distancing and integrative motivation, too. Hisako (the one who likes glamorous blondes) wants to be a movie marketer, either in America, or to please her parents, in Japan. (All the girls have been given names ending in ko.) Masa wants to be a bilingual DJ; Osamu, an international journalist; Eriko, an international tour conductor. Rieko wanted to be a translator, then a secretary who would use English, and now she wants to be a flight attendant. Kazu and Noriko have more mundane goals—they want to be English teachers, and Gun (from Taiwan) wants to be a dental engineer in Japan until he is thirty-five and then go abroad and become a Japanese teacher.

 Often clearly remembered events or influences sparked the motivation that led to an informant’s special interest in English (1.4 vs. .8). These sometimes also led to a career interest; as in Masa being inspired by a bilingual DJ on the radio or Eriko reading a mystery with an international tour conductor as heroine. At around age ten, Wakako dubbed the movie ET off the TV bilingually and was surprised and fascinated that she could not understand the English soundtrack. This inspired her to take radio English lessons, using the accompanying textbook, every day for two years.

 Concrete needs for English for academic study plans also motivate speaking and listening practice. More of the Highs than the Lows have academic plans for sometime after high school that will call for conversational English skills (1.9 vs. .4). Several want to participate in exchange programs or obtain degrees abroad, and two have taken steps such as researching foreign universities and trying the TOEFL test.

The Highs have concrete goals demanding exceptional English proficiency, while the Lows simply think that English is a good thing to know. Both Highs and Lows think that English is best learned abroad, but the conclusion of the Lows is that it is therefore better to forget about speaking in the meantime and concentrate on passing tests, while the conclusion of the Highs is that they had better start getting good enough at speaking to be able to go abroad.


Research Implications

The solutions to both educational and research problems are to be found in the context of that problem. I believe that the field of ID research would benefit from less dependence on large-scale questionnaire surveys. Such surveys have enriched our knowledge of the variety of ways in which learners can differ and have helped researchers to more precisely distinguish between factors by disseminating a shared language. Computer-assisted analysis of surveys offers new views of the relationship between factors. Surveys also seem to offer a quick easy way to find out about a group of learners, but Griffee (1999) explains the time consuming process of questionnaire development, and Sakamoto (1996) questions the extent to which translations of questionnaires are measuring the same construct as the original. Careful questionnaire research may take as much time as in-depth interviews, observing in natural settings, collecting learners’ diaries, and other such research methodologies that might provide much richer data. My impression is that reports on these other non-experimental research methodologies are less common than surveys in the SLA ID literature, including JALT publications and presentations.

One would not want to depend only on satellite maps of the hemisphere to find one’s way across town. The SLA field will be moved forward and teachers will learn more about their own students if surveys are used as one source of triangulation. For example, Simmons (1996) presents data from strategy surveys filled out before and after six weeks of study along with field notes taken from interview/strategy training sessions and learners’ diaries. Ishikawa (1996) suggests that we will learn more about our students by simply having them write in an unstructured way about what they wish to do in the class than by having them fill out a generic questionnaire. Snow, Hyland, Kamhi-Stein, and Yu (1996) interviewed junior high students with a card sort activity that allowed each student to explain and rank each of their choices for ingredients of an ideal class as well as to add two ingredients of their own. These choices could then be quantified and statistically analyzed with some assurance that the researchers and the students shared an understanding of the factor’s meaning. LoCastro (1994) employed group discussions to allow learners to explain situational contingencies and strategies they had not been able to detail with Oxford’s (1990) Strategy Inventory for Language Learning (SILL). Keim, Furuya, Doye, and Carlson (1996) examined the mismatch between the learning strategies and attitudes that learners profess to value when surveyed and the way they actually behave in the classroom. The researchers arrived at a much richer understanding of the realities of learning or teaching a language than if one year had done a study of strategies and the next year a study of attitudes.

The extent to which individual learners have relatively fixed characteristic learning styles is open to question in the SLA field. Yamashita (1996) found that reported learning styles varied according to the current year in high school and whether time had been spent abroad. Reid’s position seems to have evolved from a firm belief in fixedness (1995, p. viii) to an acknowledgment that styles can evolve (1998, p. xiii). My study’s retrospective approach revealed instances of informants’ autonomous strategies becoming much more aurally-oriented after various sorts of stimulation, such as exposure to new study techniques in conversation classes, being attracted to foreign culture, realizing how much conversational skills were lacking after being abroad, wanting foreign friends, deciding on long-range plans to live abroad, feeling distanced from mainstream Japanese culture, or reacting against excessive test-oriented study.

LoCastro (1994) writes that the learning context influences purpose, which influences strategy choice. Qualitative research can best highlight such relationships. While an optional background questionnaire on Oxford’s (1990) SILL asks respondents why they want to learn another language, it still does not connect a particular strategy with a particular purpose. Some of my informants could almost be said to be learning two separate languages, as one might study both Latin and a modern language—written test-oriented English at school and conversational English on their own. Depending on which sort of English they were thinking of, those informants might have very different answers to questions about strategies employed, motivation, preferences, beliefs, etc.

The informants balance what they want to do for their own short-term purposes (having fun, connecting with people, etc.) and long-term purposes (work, life abroad, etc.) with what they need to do for school and entrance exams. This study thus demonstrates the need for “a new conception of strategy research, one that focuses not on learner strategies but rather on learning strategies and the intricate interplay of learner and teacher in their determination,” (Woods, 1997, p. 115). Factors may indeed consistently cluster into apparent learning styles when surveys are analyzed, but how much will having a taxonomy of types of styles or strategies tell us about the problems and potential of a particular individual’s complex learning history? We must know what world the learner lives in and what identity has developed there, what experiences have been encountered and what they meant to that learner, where the learner wants to head and how far she or he expects to get. (The dissertation this paper is based on includes several case studies of individual informants.)
Reid’s (1998) widely-used Perceptual Learning Style Preference survey identifies second language (L2) learner types such as “auditory” learners, whom she says benefit from hearing lectures or audiotapes or conversing with their teacher. This category would not fit several of my Lows who enjoy taped listening exercises but are loath to speak English to anyone. Learners do not manifest a particular type of learning style in a social vacuum. Many of my informants have never encountered a social context where those with less than native-like skills have the right to speak English freely, so they say that they can’t speak English and refuse to try.

Because this paper briefly summarizes an in-depth study, my results may sound similar to those found with large-scale questionnaire-based surveys. Surveys of large numbers of learners with precise yes/no or scaled answer choices give an appearance of exact knowledge. But many surveys do not specify, for example, whether reference is being made to the learners’ current language class/es or their whole learning career, use of the L1 or L2, interaction with or one-way input from a NS or NNS, or speaking to a partner or to the whole class.

When an informant in this current study reported studying with videos I soon knew if she was the one informant who once had one teacher who used a video in class one day or the informant who studied at home with ten to twenty different movie videos. Of course, a survey could ask learners whether they “seldom,” “often,” etc. use videos, but I also know that the classroom teacher did nothing with the video but play it once, while at home Eriko watches each movie twice, for Japanese, then English, and also rewinds the videotape to catch bits she misses. A survey will tell me what percentage of a group wants to speak English or has conversed in English, but in my study I know that the informant who says that she wants to learn how to speak English is the same one seldom says a word when she has a chance to talk to a foreigner.

In theory, questionnaires could specify as much detail in questions and answers and be customized to a specific context almost as much as face-to-face interviews, but the reality is usually different. Developing questions meaningful to respondents, readers, and researchers takes time, and replication is an important step in the positivistic research process, so it is tempting to transfer questionnaires to contexts to which they are not appropriate. Reid’s (1998) survey for ESL/EFL learners on perceptual learning styles, for example, stems from general classroom education research, and despite her modifications, still seems to be a more appropriate way to investigate how one learns history in one’s L1 than how one learns to converse in an L2. Her survey makes no reference to audiovisual materials or learning by speaking. Reid identifies auditory learners with questions such as “I learn better in class when I listen to someone,” (p. 163). Some of my informants complained about how badly certain teachers spoke English. Is the student answering Reid’s survey question envisioning a NS or a NNS, speaking the L1 or the L2 well or badly, on tape or live? The survey seems to float in a cloud of learning style theory far above the terrain of real-life language learners. For example, it asks respondents if they learn better by making drawings as they study but does not ask about learning by writing words or sentences. (The method for learning new vocabulary that was mentioned the most often by my informants was that of writing words over and over again.) Reid herself (personal communication, 1998) has said that “surveys are not great research instruments,” and that her survey is best used for learners to explore their own styles. I believe that individual interviews such as I employed qualitatively illuminate much that remains unknown when we read and attempt to interpret survey results.

Pedagogical Applications

This study suggests that teachers who even occasionally used spontaneous English or let students do so help students to attain higher oral proficiency. This suggests that teachers do not need to be helplessly stuck in the impasse of schools that teach for the tests and tests that only measure what students have learned in their grammar-translation dominated mainstream classes and testing-skills cram schools. There are other students in Japan who like most of the Highs are eager to learn English conversation. If teachers take just a little time to model helpful behaviors—speaking English, trying authentic materials, etc.-- and allow students to network, it may catalyze changes in attitudes and inspire independent study or practice with peers by interested students. Murphey (1995) speaks of the importance of both non-native speaker (NNS) teachers and students serving as role models. By themselves speaking some spontaneous English in class, NNS teachers can assist their students by demonstrating that one does not have to have native-like proficiency in order to have the ability or the right to speak English. Holliday (1996, p. 235) writes that “all English language educators need to be constantly critical and aware of the social influences and implications of what they do,” and by allowing themselves to unapologetically speak imperfect English, teachers demonstrate how students too can empower themselves. The power of this example was evident to me when the Low informant expressed amazement to see his teacher speak English.

Teachers may feel that they cannot devote enough time away from presenting and explaining textbook lessons for speaking and listening activities to make any practical difference. However, if teachers devoted a little time to requiring students to speak some English not written on a page or immediately repeated after a tape, students would experience themselves as English speakers and not just English students. Students would be forced to at least revise their idea that they “cannot speak English” to they “cannot speak well,” and if a speaking activity was repeated even once a semester over the years, students would have the chance to see some improvement in their speaking skills.

One High told me how she spoke English for fun with a few girls at her junior high but in three years had never found an English speaking buddy at her high school. Teachers may not organize pair work or small group speaking activities because they assume or know from experience that students will simply speak Japanese, but even if most students speak Japanese or remain silent, the few in each class who choose to speak English will have a way of finding each other for possible speaking practice outside of class. Those less reticent students can both offer each other moral support and serve as a model for other students who might try speaking English the next time.

Teachers can encourage independent study by having students bring in and discuss any extra materials or resources they know of. Ryan (1997) and Davis (1995) offer lists of ideas, including print and broadcast resources in Japan. Students could loan each other the sort of magazines, tapes, and videos that my informants bought or recorded off of TV or radio. Some informants were bolstered in their independent study by finding peers they could talk to about English-language music and movies, so teachers should make it clear that non-pedagogical materials are welcome, and that if the students are exchanging information about English use, some Japanese language will be tolerated.

 Many of the informants in this study seem to have learned to speak English in large part through the highly motivated use of self-chosen listening materials. Teachers who must stick to textbooks for the main content of their teaching could still devote one lesson to working with a video or song in the sorts of ways my informants did, such as taking dictation, singing along, repeating, translating, consulting dictionaries, and listening to the English first before reading the Japanese translation (seldom done at school). In this way, new strategies could be modeled that some students might try at home when more time is available, and when they can choose their favorite materials, such as the Beauty and the Beast video that was bought because it made the informant cry. In fact, the autonomous study that many of the informants engaged in was an extension of the sort of intensively analytic or memory-based strategies they learn at school, but with the difference that informants used self-chosen, authentic, and audio materials--all rarities at school.

Another finding of this study is that situations and goals requiring conversation skills often motivate the Highs to study listening and speaking, as when one High escorted a group of foreign students around her high school. Schools can put up posters for homestay, penpal, host family, and foreign travel and study programs, or better yet, provide information or organize such programs. Japanese adults who use English in their careers could also be invited to visit English classes, and the informants often mentioned liking their teachers’ personal stories of foreign travel or foreign friends, even when these stories were told in Japanese. One goal of many informants is to make foreign friends, so schools can therefore invite foreign teachers and students for long-term or one-day visits. Such classroom visits by foreigners were when some informants first realized that English was a communicative tool and not just an intellectual exercise.

Finally, this study suggests to teachers the extent to which some students who are sitting looking bored in our classrooms may actually be highly motivated to study English; motivated in the sense of liking English, having learning goals and purposes, and spending a great deal of time on autonomous English study. Before we underestimate or give up on students, we should find out what they are doing on their own and why, and what lessons their experiences can offer to the teacher and other students.

  • Beebe, J. D. (1994). "To speak English with my voice": A study of seven successful English conversationalists. In D. H. Tatsuki et al. (Eds.), Temple University, Japan Campus, Research Studies in TESOL, II (1), 25-45.
  • Beebe, J. D. (1998). Individual differences in context: Japanese secondary students learning to speak English. Unpublished doctoral dissertation, Temple University.
  • Berwick, R., & Ross, S. (1989). Motivation after matriculation: Are Japanese learners of English still alive after examination hell? JALT Journal, 11, 193-210.
  • Clark, J. D. L. & Clifford, R. T. (1988). The FSI/ACTFL proficiency scales and testing techniques: Development, current status and needed research. Studies in Second Language Acquisition, 10, 129-147.
  • Davis, R. (1995). Using English beyond the classroom. The Language Teacher, 19 (5), 52-53, 55.
  • Denzin, N. K. (1978). The research act: A theoretical introduction to sociological methods (2nd ed.). New York: McGraw-Hill Book Co.
  • Ellis, R. (1994). The study of second language acquisition. Oxford: Oxford University Press.
  • Gaies, S., & Sakui, K. (1998, November). Learners’ beliefs about methodological alternatives. Paper presented at JALT 98, Omiya, Japan.
  • Gardner, R. C. (1985). Social psychology and second language learning: The role of attitudes and motivation. London: Edward Arnold.
  • Gardner, R. C., & MacIntyre, R. D. (1991). An instrumental motivation in language study: who says it isn’t effective? Studies in Second Language Acquisition, 13, 57-72.
  • Gardner, R. C. & MacIntyre, P. D. (1993). A student’s contributions to second-language learning. Part II: Affective variables. Language Teaching, 26, 1-11.
  • Gobel, P. (1996). Student beliefs and attitudes toward language learning in university education. Temple University, Japan Campus Working Papers in Applied Linguistics, 8, 41-47.
  • Griffee, D. T. (1999). Questionnaire construction and classroom research. The Language Teacher, 23 (1), 19-21.
  • Holliday, A. (1996). Developing a sociological imagination: Expanding ethnography in international English language education. Applied Linguistics, 17, 234-255.
  • Hyland, K. (1994). Culture and learning: A study of the learning style preferences of Japanese students. RELC Journal, 24 (2) 69-91.
  • Ishikawa, S. (1996). Determining learning styles of Japanese college students through questionnaires and essays. Temple University, Japan Campus Working Papers in Applied Linguistics, 9, 42-55.
  • Johnson, J., & Takeo, M. (1996). Learning orientations and motivational strength in Japanese university students. Temple University, Japan Campus, Research Studies in TESOL, 3, 1-10.
  • Keim, B., Furuya, R. Doye, C., & Carlson, A. (1996). A survey of the attitudes and beliefs about foreign language learning of Japanese university students taking communicative English courses. JACET Bulletin, 27, 87-106.
  • Koizumi, R., & Matsuo, K. (1993). A longitudinal study of attitudes and motivation in learning English among Japanese seventh-grade students. Japanese Psychological Research, 35, 1-11.
  • Lennon, P. (1990). Investigating fluency in EFL: A quantitative approach. Language Learning, 40, 387-417.
  • LoCastro, V. (1994). Learning strategies and learning environments. TESOL Quarterly, 28, 409-414.
  • Murphey, T. (1995). Identity and beliefs in language learning. The Language Teacher, 19(4), 34-36.
  • Naiman, N., Fröhlich, M., Stern, H. H., & Todesco, A. (1978). The good language learner. Research in Education Series, 7, 1-112.
  • Nunnelley, T. P. (1993). Cognitive style, learning mode and Japanese EFL learners. Temple University, Japan Campus Research Studies in TESOL, 1 (1), 80-96.
  • Oxford, R. L. (1990). Language learning strategies: What every teacher should know. New York: Newbury House.
  • Redfield, M. (1995). Student views on language learning. Speech Communication Education, 8, 107-119.
  • Reid, J. (Ed.). (1995). Learning styles in the ESL/EFL classroom. New York: Heinle and Heinle Publishers.
  • Reid, J. (1998). Appendix 3: Perceptual learning style preference survey. In J. Reid (Ed.), Understanding Learning Styles in the Second Language Classroom (pp. 162-166). Upper Saddle River, NJ: Prentice Hall Regents.
  • Rubin, J. (1975). What the “good language learner” can teach us. TESOL Quarterly, 9, 41-52.
  • Ryan, S. M. (1997). Preparing learners for independence: resources beyond the classroom. In P. Benson, & P. Voller (Eds.), Autonomy and Independence in Language Learning (pp. 215-224). London: Longman.
  • Sakamoto, M. (1996). The effect of translating survey questions. Temple University, Japan Campus Working Papers in Applied Linguistics, 9, 82-88.
  • Sawaki, Y. (1997). Japanese learners’ language learning motivation: A preliminary study. JACET Bulletin, 28, 83-96.
  • Segalowitz, N. (1997). Individual differences in second language acquisition. In A. M. B. d. Groot & J. F. Kroll (Eds.), Tutorials in bilingualism (pp. 85-112). Mahwah, NJ: Erlbaum.
  • Simmons, D. (1996). A study of strategy use in independent learners. In R. Pemberton, E. S. L. Li, W. W. F. Or, & H. D. Pierson (Eds.), Taking control: Autonomy in language learning (pp. 60-75). Hong Kong: Hong Kong University Press.
  • Skehan, P. (1989). Individual differences in second-language learning. London: Edward Arnold.
  • Skehan, P. (1998). A cognitive approach to language learning. Oxford: Oxford University Press.
  • Snow, M. A., Hyland, J., Kamhi-Stein, L., & Yu, J. H. (1996). U.S. language minority students: Voices from the junior high classroom. In R. M. Bailey & D. Nunan (Eds.), Voices From the Language Classroom: Qualitative research in second language education. Cambridge: Cambridge University Press.
  • Watanabe, Y. (1992). Washback effects of college entrance examination on language learning strategies. JACET Bulletin, 23,175-194.
  • Wells, R. (1992). Max's Dragon Shirt. Kiddy CAT, 3, 7-30.
  • Willis, M. (1996). The learning strategies of EFL Learners and their relationship with communicative and receptive contact with the target language. Temple University, Japan Campus Research Studies in TESOL, 6, 69-88.
  • Woods, D. (1997). Decision-making in language learning: A lens for examining learner strategies. The Language Teacher, 21, (10) 67-75, 115.
  • Yamashiro, A. D. (1998). Investigating modeling relationships among attitudinal factors and language proficiency in Japanese college students. Nihon University Journal of Humanities and Sciences, 4 (2) 31-49.
  • Yamashita, S. O. (1996). The Japanese returnee students’ learning styles. Japanese Society, 1, 124-146.