efl teachers ' perceptions of communicative , audiolingual and yakudoku activities

In recent years, the learning of English as a Foreign Language in Japanese high schools has become the focus of new educational policies applied at the national level. One of these is The Course of Study issue by the Ministry of Education, in which teachers are, for the first time in a long series of curriculum guidelines, adjured to develop students' "positive attitudes towards communicating in English." Another is the JET program, which has put thousands of native English speaking assistant language teachers (ALTs) into Japanese secondary classrooms for the purpose of team teaching with Japanese teachers. Data resulting from a survey project of 876 Japanese high school English teachers was used to provide empirical evidence of teachers' levels of approval of communicative, audiolingual and traditional (yakudoku) activities. Teachers were also asked to rate the strengths of a variety of influences on their instruction, including university entrance exams, and preand in-service teacher education programs. Teachers' perceptions of both activities and instructional influences were examined in light of teachers' length of career, type of school (private versus public, academic versus vocational), and level of contact with an ALT. The data revealed the complexities of imposing broad, national educational policies on a diverse group of teachers, and in an educational culture which likely precludes teachers' use of communicative activities.


Introduction
In recent years, the teaching of English as a Foreign Language in Japanese secondary schools has become the focus of a variety of new educational policies applied at the national level.In 1989, the Ministry of Education issued a new set of curriculum guidelines and course descriptions for the instruction of English in high schools, called The Course of Study (Ministry of Education, Science, and Culture, 1992).For the first time, descriptions for the mainstream, four skills English I and II courses in the new Course of Study included the startling injunction that high school teachers were to instill a "positive attitude towards communicating in English" in their students (McConnell, 1995).
Another major change in foreign language education policy in secondary schools applied at the national level was the 1987 advent of the JET program, which brought native English speaking "assistant language teachers" (ALTs) into Japanese junior and senior high school English classes (McConnell, 1995;Wada & Cominos, 1994).The purpose of the JET program was to "provide increased opportunities for interaction in the schools between [ALTs] and Japanese teachers of foreign languages," and by extension, promote the teaching of communicative English (Wada & Cominos, 1994: 1).The JET program is well endowed, with an annual operating budget of US$222,000,000 (McConnell, 1995).The JET program is currently in its twelfth year, and employs 5,361 ALTs from numerous countries ("JET program," 1998).Given the conservative leanings of the Japanese education sector (Lincicome, 1993), these two policies are radical.
However, there are several obvious aspects of the Japanese high school educational culture that work against teachers' acceptance of activities designed to promote students' communicative abilities (McConnell, 1995), implying a mismatch between this politically inspired plan and the realities of Japanese high school EFL education.Further, it is not even clear what Japanese high school English teachers believe about communicative activities.No empirical research on teachers' perceptions based on a generalizable sample has been done, even though The Course of Study has been in force in the majority of Japanese high schools since 1992.Observers note that the beliefs of the teachers have not have been taken into account in The Course of Study (LoCastro, 1996;Pomatti, 1996;Wada, 1994).There is evidence of this in the JET program as well.According to McConnell (1995), the decision to request ALTs for schools is often made at the prefectural level for political reasons.At the local level then, the day-to-day supervision of ALTs is often left to Japanese teachers of English, who resent the extra workload (Gillis-Furutaka, 1994;McConnell, 1995;Uehara, 1992).The traditional style of reform done by the Ministry of Education is well described by Markee's notion of the center-periphery model of innovation diffusion, in which teachers "merely implement the decisions that are handed down to them " (1997: 63).
This lack of regard for teachers' beliefs about language teaching may be a fatal omission.In contexts in which educational innovations are being implemented, teachers' attitudes take on tremendous importance.Teachers' attitudes and beliefs are the single strongest guiding influence on teachers' instruction (Cuban, 1993;Doyle, 1992;Fang, 1996;Freeman, 1989Freeman, , 1998;;Reynolds & Saunders, 1987;Thompson, 1984).This article reports Japanese high school English teachers' approval of communicative and non-communicative activities through empirical data resulting from a recent nationwide survey of 876 Japanese EFL high school teachers in nine randomly selected prefectures.The article also describes teachers' perceptions of the circumstances in which they operate, and discusses what effects these circumstances likely have on teachers' approval of communicative activities.This juxtaposition of attitudes and circumstances is suggested by Ajzen (1988), who was concerned about the links between personal attitudes, intentions, circumstances, and personal action; and Markee (1997), who was concerned about the effects of an educational culture on teachers' acceptance of a language education innovation.The presentation and discussion of the data will be used to characterize, from the teachers' point of view, the current state of Japanese EFL education in high schools during a period of time in which sweeping, nationally applied policies have been instituted.

Understanding Teachers' Attitudes: Limitations
Because this study explores teachers' attitudes towards various types of instruction, it is necessary to clarify the relationship between teacher attitudes and actual behavior.For this purpose, Ajzen's model (1988) was adopted.Use of Ajzen's model in EFL/ESL research contexts has been reported in Kennedy and Kennedy (1996).According to Ajzen, an attitude is a person's "evaluative reaction" to some object of interest (1988, p. 23).Ajzen suggested that attitudes then "predispose" the person to creating a cognitive response (a belief) about the object, and a potential to act on the object (an intention).However, positive attitudes towards communicative activities and even positive intentions to do them in the classroom may be influenced by what Ajzen called "subjective norms" and "perceived behavioral control" (p.133).Ajzen defined "subjective norms" as an influence on intentions arising from a person's "perception of social pressure to perform or not perform the behavior under consideration" (p.117).Thus, for Japanese high school English teachers, sources of subjective norms would be their students, or colleagues.
Ajzen defined "perceived behavioral control" as "the extent to which people have the required opportunities and resources" to do something (p.127).Thus, teachers may be hindered in doing communicative activities by "internal" and external" factors of perceived behavioral control (pp. 128-130).Examples for Japanese high school English teachers would be adequate training in communicative methodologies, or textbooks that aided them in creating communicative activities.According to Ajzen's model, then, teachers' attitudes may not be predictive of their behavior.Even though they say they approve of particular types of activities, they may not actually do them in their classrooms.Thus, any data on teachers' attitudes must be interpreted carefully in terms of the realities of teachers' every day work.

The Realities of Japanese High School English Education
There are several aspects of current Japanese high school English education which constitute potential impediments to teachers' acceptance of communicative activities, and thus, the policies of Japanese educational authorities.These are: yakudoku, an entrenched traditional method of instruction; high stakes university entrance exams, and inadequate pre-and in-service teacher education programs.
Yakudoku, a traditional method of foreign language instruction, focuses almost exclusively on the translation of English literary texts into Japanese, and direct grammatical instruction in Japanese (Bamford, 1993;Bryant, 1956;Gorsuch, 1998;Henrichsen, 1989;Hino, 1988;Law, 1995).Yakudoku has been characterized as an impediment to earlier efforts to change EFL instruction (Henrichsen, 1989, p. 104).In two yakudoku classrooms, Gorsuch (1998) observed strongly teacher-centered instruction focused largely on the translation of a difficult English text into Japanese.Both teachers in the study reported that they did not ask the students to produce their own original spoken or written English utterances or sentences, because it would be too "difficult" for students.Clearly, students' abilities to communicate in English could not be developed in such classrooms, in that one of the cornerstones of communicative activities is to create semi-realistic situations in which students can express intended meanings in the second language (Hatch, 1992;Richards & Rodgers, 1986;Terrell, Egasse, & Voge, 1982).
There are historical reasons why yakudoku remains firmly in place.In postwar Japan during the late 1940s and early 1950s, English language education in secondary schools was marked by a real shortage of English teachers who could speak English and who had sound pedagogical training (Henrichsen, 1989).As a result of post-war teacher education policies designed to quickly increase the number of certified teachers in all fields, large numbers of college graduates who were not proficient in spoken English were made English teachers at secondary schools as a "stop gap measure" (p.163).Such teachers likely used yakudoku, because this is what they knew, and did not have to speak English in order to teach it, a trend which continues today (Kawakami, 1993;Pomatti, 1996;Wakabayashi, 1987).
University entrance exams in Japan are high stakes, and affect the lives of Japanese high school students in many school settings.Many observers have noted strong effects of university entrance exams on classroom instruction in Japan (Eckstein & Noah, 1989;National Institute for Educational Research, 1991;Rohlen, 1983), including English language instruction (Brown & Yamashita, 1995a, 1995b;Gorsuch, 1998;Hildebrandt & Giles, 1983;Kawakami, 1993;Kodaira, 1996;Koike & Tanaka, 1995;Law, 1994Law, , 1995;;Miller, 1998;Yukawa, 1994) and on teachers' attitudes towards communicative activities (Gorsuch, 1999a).Reportedly, Japanese high school English teachers feel they are expected to prepare students for university entrance exams by having students translate English passages into Japanese, taking vocabulary quizzes, and focusing their instruction on developing students' linguistic knowledge at the expense of linguistic skills (Law, 1995;Miller, 1998).Many students at academic high schools seem to believe that the purpose of high school English education is university exam preparation (Kodaira, 1996;McConnell, 1995;Pomatti, 1996).Students may influence teachers' instruction through their expectations that teachers are supposed to prepare them for the exams, a phenomenon noted in Japan (Gorsuch, 1999a;Hildebrandt & Giles, 1983), and in other contexts in which high stakes tests are in place (MacDonald & Rogan, 1990;Madaus, 1988;Morris, 1985).
Inadequate pre-service teacher education programs are a third impediment to teachers' acceptance of activities designed to develop students' communicative skills.Current EFL pre-service teacher education programs lack vision and depth of instruction in teaching methodology, and do not provide sufficient teaching practica experiences (Kawakami, 1993;Kizuka, 1997).Many would-be teachers get teaching certificates from universities that do not have an education faculty.Such programs may have little actual interest in teacher preparation (Kizuka, 1997;Kobayashi, 1993).In these programs for EFL teachers at "course approved" universities, would-be teachers need only take a minimum numbers of courses related to English, such as English literature or linguistics.They do not get enough courses which bridge "English language theory and practice" (Kizuka, 1997;National Institute of Educational Research, 1989).The result is a pre-service teacher education system that is inadequate to the task of supporting the development of fundamental changes in instruction implied by policies presented in The Course of Study and the presence of ALTs in high schools.
Inadequate in-service teacher education programs are a fourth impediment.On the face of it, it does not seem likely that Japanese in-service programs can produce teachers who have the tools to analyze and change their own teaching, as proposed by Combs (1989), Lortie (1975), andKanu (1996).Government mandated in-service teacher education in Japan consists of first year induction for new teachers, and very limited in-service courses for experienced teachers.Responsibility for the planning and execution of these programs along Ministry of Education guidelines is left in the hands of prefectural and municipal Boards of Education (Kobayashi, 1993).This has two implications.First, in-service teacher education varies widely in frequency and content from prefecture to prefecture.And second, first year induction and in-service programs are generally provided for public high school teachers, but not for private high school teachers.
"Instructional technique" training for new high school English teachers in Kyoto consists of thirty days of "TEFL training" (Gillis-Furutaka, 1994, p. 34).In Fukui Prefecture, new English teachers at public schools have their teaching observed once by a "High School English Teacher's Consultant," who gives the new teacher "feedback and guidance."In addition, new teachers must undergo a two day seminar in which teachers "learn about game and activity design, motivational strategies, and teaching communicatively" (male Japanese prefectural English faculty in-service program coordinator, personal communication, December 4, 1997).
Public high school English teachers are also required to undergo limited in-service training at later points in their careers.In-service programs can potentially promote the use of communicative activities in Japanese classrooms among senior teachers who may not have had the opportunity to receive training otherwise, and who are "farther away" from their university pre-service training than junior teachers.Indeed, Cohen and Spillane (1992) note that teachers' length of career can influence their attitudes towards instruction.In-service training, if effective, may change senior teachers' attitudes.
Unfortunately, at least one observer, a high school EFL teacher herself, questioned the quality of board of education sponsored in-service education programs, and noted that such programs are offered only for short periods of time (Okada, 1997).Data provided by teaching consultants in Fukui, Nagano, Shizuoka, and Yamaguchi prefectures suggested programs that run from one to three days.The brevity of in-service training for Japanese teachers runs counter to the suggestions of Cohen and Spillane (1992) and MacDonald and Rogan (1990), who stated that effective in-service teacher education should be extended for long periods of time, and conducted while teachers continue their usual teaching schedule.
Finally, due to budget constraints, some prefectures may not offer any specialized EFL in-service teacher education, as in the case of Toyama Prefecture, which discontinued their "English Teacher's Workshop" in 1997 (male Japanese prefectural English faculty in-service program coordinator, personal communication, February 25, 1998).It is apparent that specialized in-service teacher education for EFL teachers is not uniform at the national level.Data from this study may indicate whether teachers' length of career has an effect on their approval of communicative, or other activities, and whether teachers at different stages in their career report that participation in in-service programs influences their instruction.

Diversity in Japanese High School Education
The Japanese high school education system is surprisingly diverse, and The Course of Study, a broad national policy, and the JET program, a national level program, are being applied to it.In the research project used to generate the data for this article, teachers at both public and private academic and public vocational and night high schools were surveyed, in order for the data to be generalizable to the population of high school English teachers in Japan.Combined teachers' lists for the nine prefectures revealed that Japanese English teachers at public vocational schools constituted a sizable minority, 783 (12.7%) of all 6,167 teachers in the nine prefectures.Private high school English teachers accounted for 21.8% (1,345) (Gorsuch, 1999a).
From the prefectural teachers' lists, it is apparent these high schools are located in urban areas, and are university-preparation oriented.There is essentially no literature extant focusing on EFL instruction in private academic high schools as specific contexts.
There is more literature extant on public vocational and night high schools, although still virtually nothing on EFL programs and teachers specifically.Unfortunately, what there is describes a system of schools which currently have no clear purpose, and where the students have been labeled "low ability."While vocational education at the upper secondary level has been historically intended to fill the labor needs of commerce and industry, vocational and night high schools later became the territory of students who could not successfully compete for admission into colleges or universities (Cantor, 1985;James & Benjamin, 1988).Of direct relevance to high school teachers, Cantor stated "vocational courses find it difficult to recruit good, well qualified teachers" and "both teachers and students suffer from low morale" (p.71).James and Benjamin (1988) painted an equally stark picture, suggesting that the Ministry of Education creates guidelines (The Course of Study) that keep high school curricula "hard" and fast paced.The guidelines thus act as a screening mechanism to place high school age students in secondary schools appropriate to their academic abilities, as defined by their ability to score well on examinations.The effect of applying a difficult, unitary set of guidelines on a whole population of students with varying abilities in test taking is that high schools in which "low ability" students are concentrated "are given little leeway to address the needs of these students" (39).This may also be true for EFL teachers in vocational high school settings.The data presented in this article may indicate whether such teachers constitute a unique group which responds to the needs of a specific group of students.The data may also indicate whether The Course of Study is really applicable to students in vocational and night high schools.

Assistant Language Teachers: The JET Program
The overt purpose of the JET program is to have the assistant language teachers (ALTs) and Japanese teachers of English (JTEs) interact in English and raise JTEs' awareness of English as a communicative medium (Wada & Cominos, 1994b: 1).As such, the JET program offers a powerful potential for instructional change among Japanese teachers of English.Yukawa (1992Yukawa ( , 1994) ) documented changes in the teaching of a male JTE at a high school as a result of team teaching with an ALT.Generally, the JTE stopped using the traditional yakudoku translation method and began using communicative methods in class.When the JTE and ALT's teaching relationship ended, however, Yukawa found that the JTE reverted back to teaching in traditional ways.It is possible that the JTE, without the support of the ALT, "disconfirmed" his previous decision to use an educational innovation, in this case, communicative activities (Markee, 1997).Futher research on the persistence of the effects of ALTs on JTEs' instruction seems in order.
It should be noted that team teaching with ALTs is not universally available, or applied.ALTs in the JET program are sent only to schools which formally request them (male Ministry of Education JET functionary, personal communication, September 26, 1997).This means that teachers in some prefectures have more opportunities to teach with ALTs than in others.For example, heavily populated Kanagawa Prefecture has 62 English speaking ALTs in the JET program, while less populous Shizuoka Prefecture has 152 (Ministry of Education, 1997).In addition, schools schedule ALTs for classes in quite different ways, with some schools sending ALTs to a new school every day ("one-shot visits"), to schools that have JTEs and ALTs maintain a regular thrice weekly team teaching schedule in one classroom.

Purpose/Research Questions
The Ministry of Education Course of Study has been applied at a national level to Japanese high school EFL teachers at different stages in their careers in very different types of schools, and with variable access to ALTs.It is important to document teachers' responses to the communicative ethos of The Course of Study in light of these three variables, and to learn more about their attitudes towards activities associated with other language learning approaches known to be in use in Japan.The research questions are: What teaching activities associated with communicative, audiolingual, and yakudoku approaches to foreign language instruction will Japanese high school English teachers report as being appropriate or not appropriate for English I and II courses?Will teachers' responses differ according to teachers' length of career, type of school, or level of involvement with an ALT?
In addition to documenting teachers' attitudes towards various language learning activities, it is necessary to document teachers' perceived circumstances.Elements of teachers' circumstances would include: teachers' perceptions of the strength of influence of university entrance exams, students' expectations, colleagues' expectations, pre-and in-service teacher education programs, etc. (For a full description of postulated influences in teachers' instruction see Cohen & Spillane, 1992;and Gorsuch, 1999a).In order to compare these data effectively with the results of research question #1, teachers' responses will also be examined in the light of the three variables of teachers' length of career, type of school, and level of involvement with an ALT.
What influences on instruction will Japanese high school English teachers report as being strong or weak?Will teachers' responses differ according to teachers' length of career, type of school, or level of involvement with an ALT?

Method Participants
The participants for this study were 876 Japanese high school English teachers at public academic, public vocational, and private academic high schools in nine randomly selected prefectures (Fukui, Kanagawa, Nagano, Saga, Shizuoka, Tokushima, Toyama, Yamagata, and Yamaguchi).Teachers' names were sampled using a systematic random sampling procedure from nine teachers' lists obtained from prefectural boards of education, and from high school teachers in the prefectures.The number of 876 represents a 85% return on the target sample size of 1,035.340 of the respondents were public academic high school teachers, 277 were public vocational and night high school teachers, and 259 were private academic high school teachers.

Materials
The main data collection instrument providing data for this article was a Japanese-language questionnaire (for the English-language version see the Appendix).The questionnaire had four subsections.Subsection A was designed to capture teachers' attitudes towards classroom activities associated with communicative, audiolingual, and yakudoku approaches to foreign language instruction.All three approaches are known to be in current use in Japanese high schools.Teachers were asked to respond to twelve activities in terms of their appropriateness for English I and II courses they were currently teaching by circling a score from 1 ("strongly disagree") to 5 ("strongly agree") under each questionnaire item.To develop the construct validity of the items in this section, eight EFL educator panelists (four of them Japanese, four of them native speakers of English) were asked to categorize a list of 30 activities into the three approaches.Only those items which the panelists were able to unanimously categorize were included in the questionnaire.
Subsection B was designed to establish the grouping variables for the study: teachers' length of career, type of school, and level of involvement with ALTs.Teachers responded to the items by checking one category for each item that fit their situations.For length of career (B1), the three categories were 0-8 years of experience, 9-16 years, and 17+ years.For type of school (B2), the categories were public academic high school, public commercial or industrial high school, public night high school, and private academic high school.Teachers' responses to public commercial, industrial, and night high schools were combined and treated as one category (public vocational high schools).For level of involvement with ALTs (B3), the three categories were teaching English I or II with an ALT at least once a week, less than once a week, and not at all.These grouping variables and their categorical breakdowns were suggested by the literature (Cohen & Spillane, 1992) and a pilot survey conducted by the author (Gorsuch, 1999a).
Subsection C provided the researcher with additional information about the teachers, including their educational experiences.Subsection D was designed to capture teachers' perceptions of the strengths of various influences on their instruction in English I and II classes.On seventeen items, teachers were asked to rate their agreement that a given influence influenced their instruction on a scale from 1 to 5, with 1 indicating "strong disagreement" (a weak influence) and 5 indicating "strong agreement" (a strong influence).The items were inspired by Cohen and Spillane's (1992) notion of "instructional guidance," a model designed to enumerate all possible influences acting on teachers' instruction.The items included in the main questionnaire were items that displayed an adequate degree of construct validity through the earlier pilot survey.
The five page questionnaire was mailed out to teachers in the nine prefectures in three successive waves during spring and summer, 1998, about three weeks apart.Included in each of the first wave of questionnaire envelopes were the questionnaire, a postage paid addressed return envelope, and the gift of a pencil.Teachers were not asked to provide their names when returning the questionnaire.Teachers' responses to items were coded and the data were entered into a MacIntosh PowerBook 5300cs computer on a statistical program, StatView 4.5 (1995).All analyses were conducted using StatView 4.5.Questionnaires with missing data were not included in subsequent analyses.

Analyses
Descriptive statistics for all items in questionnaire subsections A (activities) and D (influences on instruction) (k = 29) were calculated including means, standard deviations, skewness coefficients, minimum/maximum scores, and modes.Descriptive statistics for each item split by the three grouping variables (teachers' length of service, type of school, level of involvement with ALTs) were also calculated.Factoral ANOVAs were calculated for each of the 29 comparisons per grouping variable with statistical significance set at p < .0017(.05 divided by 29) to check for significant differences in mean scores on subsection A and D items based on teachers' group memberships.Cronbach's alpha was used to estimate the reliability (internal consistency) of subsection A and D items.

Results
Descriptive statistics for Subsection A are in Table 1.They have been reported from highest mean to lowest.Teachers gave centered responses on the data.The highest mean score (item A12) was 3.893 and the lowest was 3.080 (item A4).Such centered scores above a "3" indicate a very mild approval of all twelve activities presented to teachers.Teachers in general dwelled in the area between "don't know" (3) and "approve" (4), a conservative and cautious place in which to be.All of the items had a negative skew, which indicated that teachers' responses tended to be bunched up towards the upper end of the distribution created by their scores.This, taken with mode of 4 ("approve") on all items, suggests that as a group, teachers responded in quite similar ways on each item.
Relative approval ratings between items associated with communicative, audiolingual, and yakudoku approaches were not entirely clear cut, although teachers were less approving of yakudoku activities than expected.However, when items were grouped by level of control of teachers over the language used by students, a more unambiguous pattern emerged.The yakudoku items (A1, A4, and A10) aside, teachers approved of controlled activities more than they did activities involving student generation of extemporaneous (non-scripted) language.If items were ranked by mean score from 1 (highest mean score) to 12 (lowest mean score), the six "high teacher/language control" items all rank 6 or above (items A11, A12, A3, A5, A6, and A8), indicating higher approval by teachers.The three "low teacher/language control" items (A2, A7, and A9--all of them communicative items) were ranked at 9, 10, and 11, indicating lower approval by teachers.
Descriptive statistics for Subsection D are in Table 2.These are ranked from highest mean score to lowest.Teachers' responses were more varied and less centered for subsection D items than on subsection A items.The highest mean score was M = 4.318 (students' abilities in English) and lowest was M = .587(membership in an academic organization).For whatever reason, teachers saw no reason to restrict their responses to 3 and 4 on the one to five point Lickert scale as they largely had on subsection A items.Negatively skewed items indicated that teachers' responses tended to be concentrated around the upper end of the distribution created by teachers' scores, while positively skewed items indicated that teachers' responses tended to be concentrated around the lower end of the distribution.
The highest mean score items were D16 (M = 4.318, mode = 4) (students' abilities in English) and D12 (M = 4.026, mode = 4) (class size).Both indicated strongly that teachers felt these influences in their instruction.Both items represent very "local" influences, which would act directly upon the teachers inside their classrooms.The third, fourth, and fifth highest ranked mean scores belonged to items D2 (M = 3.905, mode = 4) (university entrance exams), D15 (M = 3.855, mode = 4) (students' expectations), D3 (M = 3.701, mode = 4) (textbook), all of which indicated still fairly strong perceptions overall that these influenced teachers' instruction.The sixth and seventh highest mean score items D17 (M = 3.620, mode = 4) (teachers' English speaking ability) and D6 (M = 3.558, mode = 4) (teachers' experiences learning English as students) indicated moderate agreement that these influence teachers' instruction.
Between the sixth and seventh highest ranked mean score items and the eighth, ninth and tenth highest mean scores is a rather large break of nearly half a point, down to items D7 (M = 3.094) (colleagues), D11 (M = 2.986) (locally written English I and II syllabuses), and D1 (M = 2.961) (Ministry of Education Course of Study).These three items were very centered (mode = 3), indicating neither agreement nor disagreement that these influence teachers' instruction.
The eleventh, twelfth, and thirteenth highest mean score items were also in a league of their own, numerically.Items D14 (M = 2.634, mode = 2) (expectations of students' parents), D5 (M = 2.462, mode = 3) (in-service teacher education), and D4 (M = 2.379, mode = 2) (pre-service teaching license program) all represented rather "distant" influences, distant either through time or proximity.Teachers' responses indicated mild disagreement with the notion that these influence instruction.
On the teacher's length of career grouping variable, six mean scores on Subsection A (activities) and D (influences) items were significantly different by group at p < .0017.See Table 3. br>

5.633
There were some significant differences between teachers on the basis of their length of teaching career.The most senior group of teachers with 17+ years of experience were more likely than the most junior teachers (0-8 years) to approve of a traditional yakudoku reading activity (A1).The same senior teachers were less likely to approve of a communicative information gap activity than the most junior teachers (A3).In terms of instructional influences, the junior teachers reported being more strongly influenced by their own language learning experiences, colleagues, and the expectations of students than the senior teachers did (D6, D7, D15).Finally, the middle group of teachers with 9-16 years of experience reported being more strongly influenced by teacher development courses they took privately than the junior group of teachers (D9).
On the type of school grouping variable, eleven mean scores on Subsection A (activities) and D (influences) items were significantly different by group at p < .0017.See Table 4.Both public vocational high school English teachers and private academic high school English teachers emerged as singular groups, implying that teachers in these groups have quite different priorities.In terms of influences on instruction, public vocational high school teachers indicated that they were less influenced by university entrance exams than both public academic and private academic high school teachers (D2).Public vocational teachers also reported less influence from their English I and II syllabuses than public academic high school teachers (D11).Finally, public vocational teachers reported being less influenced by students' parents' expectations than private academic high school teachers (D12).
The differences that set private academic high school English teachers apart from teachers in the public sector were more numerous, and point to Japanese private academic high schools as being unique environments.In terms of activities, private academic teachers were more approving of traditional yakudoku reading activities than public vocational high school teachers (A1) and public academic and vocational teachers combined (A4).However, private academic high school teachers were less approving of a communicative information gap activity than public academic high school teachers were (A3).Perhaps related to private academic high school teachers' attitudes towards activities is the fact that such teachers reported being less influenced by prefectural in-service teacher education programs than both public academic and vocational high school English teachers (D5).This may imply that such public funded in-service programs are simply not available to private high school teachers.If that is the case, then private high school teachers may have fewer opportunities for professional development, and do not learn about activities such as the communicative information gap activity.
In terms of the influence of human agents on instruction, private academic high school teachers reported being less influenced by their colleagues than public academic high school teachers (D7).However, private academic high school teachers reported being more influenced by their school principals than teachers at either public academic or vocational schools (D8).Finally, private academic high school English teachers reported being much less influenced by ALTs in English I and II courses than either public academic or vocational teachers (D13).This can mean two things: First, private high schools may not have ALTs, and second, private high schools may not use ALTs to team teach in their mainstream English I and II courses and are instead assigned to "oral communication" classes which are less widely offered (the latter has been strongly suggested in Gorsuch, 1999a).
On the level of involvement with an ALT grouping variable, only two mean scores on Subsection A (activities) and D (influences) items were significantly different by group at p < .0017.See Table 5.

380.547
Teachers teaching with ALTs more than once a week, or less than once a week approved of the communicative information gap activity more than teachers with no ALT contact (A3).And, not surprisingly, teachers teaching with ALTs more, or less, than once a week reported being much more influenced by ALTs than teachers with no ALT contact (D13).
Cronbach's alpha internal consistency coefficient for subsection A and D items was only .6878,which was only moderate.Subsection A and D items purportedly measure several different constructs, which will depress internal consistency estimates.In addition, teachers' responses to subsection A items (activities) were very centered (values all around "3").Such homogeneous values will probably depress internal consistency estimates.In addition to the constructs the researcher intended to measure, there was some measurement error, as indicated by the moderate reliability coefficient.

Discussion
What activities do teachers approve of?The results indicated that teachers have generally positive attitudes towards communicative language teaching (CLT) activities.However, teachers seemed to prefer the more highly controlled, passive skill, CLT activities over CLT activities that called for students to engage in extemporaneous (non-scripted, non-memorized) speech and writing.Teachers' greater preferences for controlled CLT activities were matched by strong preferences for the audiolingual activities, which involved the students' use of memorized speech in pattern practice drills or dialogs.Thus, the teachers seemed to indicate that CLT activities were alright, as long as the teachers could control students' language while using them.The teachers seemed to be responding in a cautious, although positive, way towards communicative activities.Gorsuch (1998) described the two high school English teachers she observed as being overwhelmingly concerned with student accuracy.There may be perfectly justifiable reasons for teachers' desire for control.Japanese classes typically have at least 40 students in them (Gorsuch, 1998;Kawakami, 1993).With such a large class, it would be easy to "lose control" of students during a communicative speaking activity.In addition, teachers night feel hard pressed to effectively monitor 20 or more pairs of talking students.Yet The Course of Study specifically mentions helping students develop a positive attitude towards communication.If students are to do so, they have to be allowed and encouraged to communicate in class.The reasoning behind this is, how can students develop a positive attitude towards communication if they do not actually experience communication?In the end, teachers may have to learn to give up a measure of control over students' use of English, and demand smaller classes.
The communicative information gap activity A3 seemed to be a kind of litmus test for approval or non-approval of CLT activities based on group membership.Teachers who approved of A3 more highly were younger teachers, teachers at public academic high schools, and teachers who had at least some contact with ALTs.Teachers who did not approve of A3 as much were older teachers, teachers at private academic high schools, and teachers with no contact with ALTs.Concerning teachers' length of career, more senior teachers may not approve of A3 because they have been out of pre-service teacher education programs longer than junior teachers.This, coupled with what seems to be a real lack of in-service teacher education programs, and a lack of interest on the part of teachers in taking professional development courses privately or belonging to academic organizations (Table 2) may imply that senior teachers have not had sufficient training to feel comfortable trying out an activity like A3 for themselves.
Most interesting, though, was the greater approval of A3 by teachers teaching at least once a week or less than once a week with ALTs than teachers not teaching with an ALT at all.Perhaps teachers who have regular contact with ALTs find it easier to model CLT pair work activities for students.It could also be that when an ALT is in the classroom, students expect to do something different than highly controlled language practice.There may also be a link with teachers' self-perception of English speaking skill--in a separate analysis of teachers' self ratings of English speaking skill, it was found that teachers teaching with ALTs at least once a week rated their English speaking skills significantly higher than teachers who had less or no contact with ALTs (Gorsuch, 1999a).Whether a causal factor or not, presence of an ALT is linked with greater approval of A3 and higher self reports of teacher English speaking ability.
There was one difference on teachers' approval of yakudoku item A4 due to group membership.Teachers at public academic and vocational high schools were less likely to approve of having students recite their Japanese translations in class than private academic high school teachers.One possible reason is that private academic high school teachers seem to be largely excluded from in-service teacher education offered by prefectural or municipal boards of education, where they may receive training in other methodologies.
Teachers' responses to all of the activity items in the questionnaire were centered around "3" (Table 1).When "significant" differences in level of approval or disapproval are discussed above, such differences were very subtle, sometimes representing half a point or less of difference on a five point scale.This was a disappointing result, yet not altogether unexpected, given the general conservatism of educators in Japan.The Course of Study is asking teachers to do something quite new--develop students' communicative abilities--and teachers are responding cautiously, and obviously only within the bounds of their understanding of what both spoken and written communicative activities entail.
What influences teachers?Teachers responded to items in subsection D in non-centered fashion.Perhaps they felt less cautious and constrained when asked to respond to "safer," less ideologically laden, items.Unfortunately, teachers' responses indicated that there were powerful impediments working against their acceptance of CLT activities, such as the strong influences of university entrance exams and students' expectations, and the surprisingly weak influences of pre-and in-service teacher education programs, and privately undertaken courses.
With the exception of the entrance exams item (D2), teachers generally agreed that students' English abilities (D16), class size (D12), students' expectations (D15), the textbook (D3), teachers' English speaking abilities (D17), and teachers' English learning experiences (D6) exerted powerful influence on their instruction.Some of these may prevent teachers from teaching communicatively.It is not surprising that teachers consider their students' abilities to be a crucial factor in planning instruction.No teacher wants to go into a classroom with a lesson plan that is too easy or too difficult for the students.Activities of the first type will bore them, and the second type will stymie and then bore them.Either case implies teachers' loss of control over the class, something Japanese teachers have indicated through their activity preferences as undesirable to them.Unfortunately, Japanese teachers seem to consider communicative activities to be "difficult," even for students in top ranked high schools (Gorsuch, 1998).If other teachers with less able students share this perception, then teachers will likely not use communicative activities, regardless of their cautious approval suggested in this study.
As noted above, class sizes are large (40+).Teachers are likely concerned whether they will be able to control such a large group of students.This perception, coupled the high influence rating teachers gave to the student expectations item (Table 2), gives the feeling that teachers may be very sensitive to losing control of the students by going against students' expectations.Recall the observations of scholars cited earlier that the majority of students expected their English class work to prepare them for entrance exams.In such a climate, teachers are unlikely to feel they can comfortably use communicative activities in class.
In terms of teachers' ratings of the influence of textbooks, current Ministry of Education approved English I and II textbooks largely focus on developing students' intensive reading skills for entrance exam preparation, and do not provide aid to teachers in developing communicative activities (Gorsuch, 1999b).This does not bode well for communicative activities, in that appropriate textbooks are necessary to successful implementation of educational innovations (MacDonald & Rogan, 1990).
There really is no escape from the influence of university entrance exams, apparently.Not only did teachers give exams a high rating, exams make their influence known through students' expectations, and through textbooks.There was one difference on the grouping variable B2 (type of school) on the university entrance exam items, however.Public vocational high school teachers were less likely to report that university entrance exams influenced their instruction than teachers at public and private academic high schools.Vocational public high schools may be the perfect venue in which to introduce programs with genuinely communicative aims.Because teachers (and, possibly the students) in these schools feel less influenced by the need to prepare their students for university entrance exams, teachers could, with concerted help, develop English courses making use of suitable communicative activities.If well designed, such activities can be motivating to students who traditionally have little desire to learn English, especially in the traditional exam preparation oriented way (yakudoku).Rather than being seen as the sad realm of students who cannot compete academically in the prevailing educational culture, the public vocational high school sector could be an important venue for meaningful instructional change that can later be adapted to the public and private academic high schools.This view of public vocational high schools is in accord with recent efforts to revitalize vocational high school education in Japan ("Vocational school curriculum urged to include scuba diving," 1998).
Teachers reported "colleagues," "locally written syllabuses," and The Course of Study as having a neutral influence on their instruction (Table 3).However, the youngest group of teachers (0-8 years of experience) reported colleagues as being more influential than middle (9-16 years) and senior teachers (17 years years) did.Given junior teachers' newness to teaching in specific contexts, it is not surprising that they need the help of more senior teachers to show them the ropes.Whether this help centers on actual teaching in English classrooms is not known.
Providing yet another argument for the adoption of alternative language programs in public vocational high school, teachers at those schools reported that locally written syllabuses influenced their instruction less than teachers in academic high school contexts did.With students who cannot compete to enter universities, vocational schools are left behind in terms of their locally written syllabuses, which are local tokens of The Course of Study.A syllabus may be written, but teachers will not, or cannot follow them, perhaps due to students' low academic interests and abilities.
One of the most distressing findings of this study was the low influence status accorded by all teachers to pre-service, in-service, and privately undertaken teacher education courses (Table 2).Either in-service or private courses are not available to teachers, or teachers do not avail themselves of them.Pre-service courses may simply not be attuned to current and future teachers' needs.These circumstances are a negative indictment of foreign language education in Japan.Without adequate pre-service and continuing teacher education, teachers cannot learn about the theoretical bases of different language learning approaches, nor get guided experiences in using them.In this non-teacher-development climate, it is difficult to see how teachers can realistically try communicative activities.However, there was a ray of hope in that teachers with 9-16 years of teaching experience were more likely to report that privately undertaken teacher education courses were influential than the youngest teachers (Table 3).It may be that these middle-aged teachers represent a group of potential users of communicative activities in that they may have confidence in their teaching seasoned by experience, yet feel they want further knowledge and variation in their working lives.The Ministry of Education and local boards of education may wish to develop more intensive and flexible in-service programs aimed specifically at this group of teachers.

Conclusion
It is clear that there is no one solution to enhancing teachers' approval of the communicative activities called for by The Course of Study and the continued presence of ALTs in Japanese high school EFL classrooms.This article has given empirical evidence suggesting that teachers mildly approve of communicative activities, yet the data also suggested there are potent impediments working against teachers actually using such activities in their classrooms.This article has also shown how these impediments may work on teachers, from the teachers' point of view.
It is a time of extraordinary change in Japanese high school EFL education.This article has provided an empirical snapshot of the perceptions of Japanese EFL high school teachers, and how these policy changes may potentially affect them.Needless to say, to track future change, further study aimed at gathering empirical data is needed from a variety of points of view.The author hopes that the Ministry of Education, and particularly high school teachers themselves, will undertake such research and take the results into account when planning future curriculum revisions, teacher education programs, or research projects.
A-2.The teacher has students look at a page that has a "picture strip story."Students can uncover only one picture at a time.Before uncovering the next picture, the students predict, writing the prediction in English, what will happen in the next picture.Students can then look at the next picture to confirm or disconfirm their predictions.I think the above is an appropriate activity for my English I or English II classes: A-3.The teacher has the students work face to face in pairs.One student sees a page that has some missing information.The other student sees a different page that has that information.The first student must ask questions in English to the other student to find the missing information.I think the above is an appropriate activity for my English I or English II classes: A-4.The teacher asks students to translate English phrases or sentences into Japanese in preparation for class.Then in class, the teacher calls on individual students to read their Japanese translation of an English phrase or sentence, and the teacher corrects it if necessary and gives the whole class the correct translation with an explanation.I think the above is an appropriate activity for my English I or English II classes: A-5.The teacher has students chorally repeat word pairs such as sheep/ship and leave/live.I think the above is an appropriate activity for my English I or English II classes: A-6.The teacher has students memorize and practice a short English sentence pattern.The teacher then gives the students a one word English cue and has the students chorally say the sentence pattern using the new word.I think the above is an appropriate activity for my English I or English II classes: A-7.The teacher pairs off students.Then the teacher asks the students to write a letter in English to their partner.I think the above is an appropriate activity for my English I or English II classes: A-8.The teacher has students memorize an English dialog and then has the students practice the dialog together with a partner.I think the above is an appropriate activity for my English I or English II classes: A-9.The teacher has pairs or small groups of students ask each other and then answer questions in English about their opinions.I think the above is an appropriate activity for my English I or English II classes: A-10.Students read a sentence in Japanese, and then see an equivalent English sentence below where the words been scrambled up.The students must then rewrite the English sentence in the correct order suggested by the Japanese sentence.I think the above is an appropriate activity for my English I or English II classes: A-11.On one page students see a picture.Underneath the picture are several short English stories.Students have to choose which story they think best matches the picture.I think the above is an appropriate activity for my English I or English II classes: A-12.On a page, students see an English paragraph in which the sentences have been scrambled.The teacher then asks the students to put the sentences into order so the paragraph makes sense.I think the above is an appropriate activity for my English I or English II classes:A-13.What activity do you feel is most effective for your students in your English I or II class?Please write a brief description here:(Optional)    Please answer the following questions by writing a check next to the most correct answer.Choose only one response.B-1.How many years have you been teaching in high school?_____0-8 years_____ 9-16 years_____ 17+ years B-2.What kind of high school are you currently teaching in?_____ public academic high school_____ public commercial or industrial high school_____ public night high school_____ private academic school B-3.Are you currently teaching English I or English II with an ALT (Assistant Language Teacher)?_____Yes, at least once a week._____Yes, but less than once a week._____No, I do not teach English I or English II with an ALT Please read the sentences below and write a check in the blank that best describes your level of agreement.Choose only once response.C-1.My English speaking ability is good enough for me to use in class.C-2.As a student I studied English primarily through translating English stories, essays, or literary works into Japanese.C-3.I think the pace we have to teach English at my high school is:much too fast____ fast____ about right____ slow____ much too slow____ C-4.The average size of my English I or English II classes is:over 50____ 40-49____ 30-39____ 20-29____ below 19____ Please read the sentences below concerning your current instruction in English I and II classes and write a check in the blank that best describes your level of agreement.Choose only one response.D-1.The Monbusho guidelines for English I and English II influences my classroom practice.D-2.College and university entrance exams influence my classroom practice.D-3.The textbook my students are using influences my classroom practice.D-4.The teaching license program I completed at university influences my current classroom practice.D-5.In-service teacher education specifically designed for English teaching offered by my prefectural or municipal board of education influences my classroom practice.In-service teacher education for English teaching is not available from the Board of