The Effects of Awareness-Raising Training on Oral Communication Strategy Use

YASUO NAKATANI Nakamura Gakuen Junior College 2-3-5-906 Atagohama Nishi-ku, Fukuoka 819-0013 JAPAN Email: [email protected] This study examines current patterns of oral communication strategy (OCS) use, to what degree these strategies can be explicitly taught, and the extent to which strategy use can lead to improvements in oral communication ability. In a 12-week English as a Foreign Language (EFL) course based on a communicative approach, 62 female learners were divided into 2 groups. The strategy training group (n = 28) received metacognitive training, focusing on OCS use, whereas the control group (n = 34) received only the normal communicative course, with no explicit focus on OCSs. The effects of the training were assessed by 3 types of data collection: the participants pre- and postcourse oral communication test scores, transcription data from the tests, and retrospective protocol data for their task performance. The ndings revealed that participants in the strategy training group signicantly improved their oral prociency test scores, whereas improvements in the control group were not signicant. The results of the transcription and retrospective protocol data analyses conrmed that the participants success was partly due to an increased general awareness of OCSs and to the use of specic OCSs, such as maintenance of uency and negotiation of meaning to solve interactional difculties.

INITIAL RESEARCH INTO LEARNING strategies sought to identify the characteristics of good language learners (Rubin, 1975; Wong-Fillmore, 1979). The researchers assumed that procient learners might be using special techniques that differed from those of less procient learners. Since then, numerous Second Language Acquisition (SLA) studies have identied the particular strategies of effective learners (e.g., Bialystok, 1978; Brown & Palincsar, 1982). Questions have also been raised about how to introduce the strategies of good learners to students in foreign (FL) or second language (L2) classrooms (e.g., Ellis & Sinclair, 1989; Oxford, 1990). Recent research in SLA has argued that use of specic strategies plays an important role in
The Modern Language Journal, 89, i, (2005) 0026-7902/05/7691 $1.50/0 C 2005 The Modern Language Journal

learning the target language (e.g., McDonough, 1995; Oxford, 1996). In particular, many scholars believe that metacognitive strategies, which focus on raising the learners awareness of the learning process, might enhance L2 skills (Cohen, 1998; OMalley & Chamot, 1990; Wenden, 1991). It has also been suggested that learners communicative skills can be improved by developing strategies for communication (e.g., Cohen, Weaver, & Li, 1998; D rnyei, 1995). It seems reasonable to o hypothesize that raising learners awareness of strategies that they might use to solve potential communication problems could develop their oral prociency. Despite agreement over the effectiveness of metacognitive strategies, little concrete work has been done to provide a method of metacognitive strategy training that would equip students to use communication strategies appropriately and effectively. Until recently, research on strategy

Yasuo Nakatani training for communication has been limited almost exclusively to cognitive strategy applications for vocabulary learning tasks (e.g., Kitajima, 1997; Poulisse, 1990). Only a small amount of research has attempted to conrm how strategies are used for oral production as learners engage in interactional tasks. Furthermore, there are no studies that examine how learners actual strategy use contributes to their English as a foreign language (EFL) oral prociency. By combining several assessment methods, such as analysis of learners conversation test scores, transcription data from the tests, and retrospective verbal reports, this study investigates the effect on EFL learners of awareness training about strategy use in conversational interaction and examines the relationship of strategy use to EFL oral prociency. BACKGROUND Strategic Competence and Metacognitive Strategy Training Although it is generally agreed that strategy training should enhance learners strategic competence, there is little agreement over the denition of this concept. For instance, Canale and Swain (1980) regarded strategic competence as the ability to use verbal and nonverbal strategies in order to avoid communication breakdowns that might be caused by a learners lack of appropriate knowledge of the target language. Canale (1983) modied this view and dened strategic competence as the skills underlying actual communication. He expanded this concept to include both the compensatory characteristics of communication strategies and the enhancement characteristics of production strategies. Bachman (1990) separated strategic competence clearly from language competence (knowledge of and about a language) by regarding strategic competence as a capacity that projects language competence into real communication contexts. The concept was further broadened by Bachman and Palmer (1996) who dened strategic competence as a set of metacognitive components, or strategies, which can be thought of as higher order executive processes that provide a cognitive management function in language use (p. 70). They proposed a clear model of strategic competence that included three components: goal-setting , assessment, and planning . According to this model, learners need to make a conscious effort to set task goals, assess what is needed to work with the task, and plan how to use their topic and language knowledge. The model emphasizes the importance of knowing how to

77 manage the language as well as language knowledge itself. The underlying concept of this model is that FL learning requires awareness of metacognitive strategy use. It is evident that the term strategic competence is used differently by different scholars. This article follows the argument of Bachman and Palmer (1996), which is based on the cognitive theory that language learning involves many conscious decisions at both the cognitive and the metacognitive levels. For the purposes of this study, strategic competence is dened as the ability to manage communication not only during an interaction, but also before and after the interaction, in order to achieve an intended interactional goal. Strategic competence is the ability to use metacognitive strategies consciously in order to solve languagerelated difculties in communicative situations. Metacognitive strategy training has been reported to be effective (Cohen, 1998; OMalley & Chamot, 1990; Wenden, 1999). Because language or FL learning involves complex cognitive skills, learners are heavily engaged in conscious internal mental activity. The production of unfamiliar FL speech is particularly difcult. In order to cope with difculties that arise in oral communication in the FL, learners need to use a variety of communication strategies. The question then becomes: How can they come to use strategies effectively in order to learn the FL? Schmidt (1990) focused on the role of the learners consciousness as one of the important factors of target language acquisition. He argued that the frequency and salience of target language input may be essential conditions for extracting meaning from that input. The more frequently learners recognize specic features of input, the better chance they have of accessing that input. The more perceptually salient the features of input, the greater the chance they can inuence conscious learning. Thus we may say that the degree to which target language data are integrated into a learners schematic system depends on the learners awareness of the learning process. However, when generating the target language, learners have to recognize a communicative goal and manage their inadequate interlanguage system to achieve collaborative interaction by intentionally using strategic and contextual knowledge. The EFL learners in this study tended to be unskilled at channeling their attention and at deciding on which part of an utterance to focus. They had little idea when to apply knowledge of specic contexts in order to solve problems during FL communication. They were not familiar with using strategies to generate the target language functionally. Accordingly, it can be said that they

78 lacked the metacognitive skills needed to learn the FL through interaction. Therefore, in the training program in this study, the focus was on metacognitive strategies that could enhance the learners creative use of language to get meaning across in contexts of authentic interactions despite their target language deciencies. In particular, the current study examined the applicability of awareness training in order to develop communication strategies for interaction. Studies on Strategy Training for Communication In view of the importance of metacognitive strategy training, let us consider how previous, representative studies have evaluated its usefulness in actual learning programs. Only a small amount of research has attempted to conrm how strategies for oral production are used as learners engage in actual tasks. OMalley, Chamot, Stewner-Manzares, Russo, and Kupper (1985) examined whether metacognitive, cognitive, and social/affective strategies could be taught successfully in an English as a second language (ESL) classroom context. They integrated tasks involving listening, speaking, and vocabulary training and found that their strategy training had a signicant effect on performance in a speaking task, but that it had no effect on performance in vocabulary and listening tasks. This mixed nding on strategy training in the classroom setting suggests that, although learning strategies can be taught, the success rate of the instruction is not always predictable. In general, OMalley et al.s results indicate the usefulness of metacognitive training that aims at raising students awareness of strategies to help them deliver meaningful messages in speaking tasks. However, the researchers only dealt with training students to deliver a monologue to an audience. There was no task-training for conversation with others in the target language. Therefore, it is worthwhile in this study to examine the effects of metacognitive strategy training on conversation tasks. Unlike OMalley and his colleagues, D rnyei o (1995) looked at speaking skills in conversation. He examined the teachability of communication strategies by focusing on whether the training of a specic strategy enhanced the quantity and quality of learners strategy use. His study was conducted in high school EFL classes in Hungary over a period of 6 weeks. He used three types of communication strategy: topic avoidance and replacement, circumlocution, and using llers and hesitation in order to remain in the con-

The Modern Language Journal 89 (2005) versation and gain time to think. The focus was on strategies for solving a learners own performance problems, which did not require interaction with others. A signicant improvement was found among the participants in the strategy training group in the quality and quantity of strategy use and in their overall speech performance. In addition, participants in this group showed positive attitudes towards their training. These results indicate that strategy training might be effective for developing conversation skills. However, the study excluded the types of negotiation behaviors used when learners carry out trouble shooting exchanges cooperatively. Also, D rnyeis study did o not seem to address clearly the question of how to integrate metacognitive strategy training directly into a strategy-based program in order to develop learners strategic competence. An initial attempt to introduce metacognitive strategies for target language communication training was made by Cohen et al. (1998). In their study, they taught students specic strategies for oral communication, such as preparation, self-monitoring, and self-evaluation. In order to evaluate the effect of the training, the researchers analyzed the results of pre- and posttraining speaking tests and checklists that the students lled out to document their strategy use. The students who received strategy training partially improved their test scores on the posttests. Yet the results of the strategy checklists suggested that high-prociency students did not always use more strategies than lower-prociency students. Although Cohen et al. attempted to improve the learners target language communication ability, their instruction model did not seem to introduce efciently such interaction skills as negotiation of meaning between interlocutors as a crucial component for learning the target language. Their results reect their approach to the evaluation of speech performances. By using semidirect, one-way audiotaped recordings they did not pay attention to the interactional aspects of oral production. In summary, the preceding discussion has highlighted several important points. First, research suggests that pairing communication strategies with appropriate metacognitive strategy training could enhance learners awareness of strategy use and develop their communicative skills. Second, strategy training studies have, to date, excluded strategies for negotiation aimed at achieving communicative goals in the target language. It is necessary to examine whether EFL learners can be guided to communicative success through the effective use of strategies for interaction.

Yasuo Nakatani Denitions of Oral Communication Strategies In order to avoid using terms that may cause the confusion in this research area, the term oral communication strategy (OCS) is used instead of communication strategy. Oral communication strategies specically focus on oral interaction and interlocutors negotiation behavior for coping with communication breakdowns. As D rnyei (1995) pointed out, opinions dio verge on what constitutes a communication strategy. In particular, two different types of denition have evolved. Focusing on the range of problemsolving activities open to the individual has come to be regarded as the psycholinguistic view (e.g., Bialystok, 1983; Kitajima, 1997; Poulisse, 1990). Focusing on the interaction between interlocutors and negotiation of meaning has come to be recognized as the interactional view (e.g., Rost & Ross, 1991; Willems, 1987). The position adopted in this article is close to that of the interactional view because I am particularly interested in EFL learners strategy use during interaction with their communication partners in classroom tasks. Many researchers have stated that learners can comprehend the target language and assist in their acquisition of it through negotiation of meaning with an interlocutor (e.g., Gass, 1998; Nakahama, Tyler, & van Lier, 2001; Pica, 1996). It has also been suggested that the use of communication strategies could improve learners skills for interpersonal communication (Bejarano, Levine, Olshtain, & Steiner, 1997; Clennel, 1995). Therefore, it is worthwhile to examine whether attempts by EFL learners to solve communicative problems that occur during interaction could be shaped into important components of a strategy-based program for communication performance. However, there is little research that investigates learners use of strategies for communication by using their actual discourse data, and there remains the unsettled question of taxonomies of strategies for communication. The following section reports on how the EFL learners use of OCSs was analyzed in this study.

79 1. How does explicit instruction in oral communication strategies affect students speaking prociency? 2. What kind of impact does strategy training have on students discourse? 3. How do the students perceive their test performance and strategy use in their retrospective verbal report protocols? Participants The participants consisted of 62 female students enrolled in mixed-level 12-week EFL classes (90 minutes per week) at a private college in Japan. They ranged in age from 18 to 19 years old. Each student had completed 6 years of English study prior to entering the college. The students chose these classes because of class availability at particular times, which means that they were not randomly assigned to groups. This is a potential weakness of the study. Given that female-only participant groups were chosen, the results of the current study might not relate to populations that do not share similar characteristics. The strategy training group comprised 28 students who received the OCS training. Another 34 students served as a control group. Both groups were taught by the same instructor using a basic communicative approach, including information gap activities. In addition, the strategy training group received explicit strategy training. Though the participants were not randomly assigned to groups, there was no signicant difference between the two groups in the results of the oral communication pretest (t = 0.497, p = .614). Therefore, it can be said that the participants were evenly distributed between the two groups according to the results of the pretest. Strategy Training Group For the strategy training group, explicit strategy instruction was introduced to help the learners become aware of their own learning processes. In order to develop their metacognitive skills, specic oral communication strategies that might enhance skills for managing interaction actively during spontaneous communication were selected and described on an oral communication strategy sheet, which was delivered to students at the beginning of the course. The sheet listed examples of achievement strategies (see Achievement Strategies section for detailed explanation) that the students could use in each lesson. From these lists, the students could locate strategies that they believed useful for interaction in

METHOD Research Design The present study investigated how students with low speaking ability changed their use of OCSs after undergoing strategy training. These changes were then related to the students gains in oral communication ability after the course. The study asked the following three questions.

80 specic tasks. The students also used a strategy diary (see Appendix A) to make plans, monitor, and evaluate their performance. I did not analyze the diary data for the current research project. The strategy diary was specically used for selfreective training by the learners. The strategy training consisted of a ve-phase instructional sequence: review, presentation, rehearsal, performance, and evaluation. In the review phase, the students reected on the previous lesson and repeated its simulation task at the beginning of each new lesson, which enabled them to warm up for a new task. In the presentation stage, according to an instructors guideline, the students recognized the goals and procedures of the new task and discussed through brainstorming sessions basic dialogues that they were asked to create and the possible OCSs for doing so. During the next stage, the students rehearsed once with their peers and used their diaries to make plans for using specic OCSs. When they then performed the tasks, they monitored their own performance according to the guidelines of the strategy diary. They were encouraged to use OCSs intentionally during the task. During the evaluation stage, the students checked and reected on their own learning in order to develop their metacognitive awareness. They reected on their strategy use and analyzed their self-assessment of their performance assessment by using the strategy diary. Control Group The control group took part in communicative tasks based on materials similar to those used for the strategy training group but without any specic strategic focus. They were exposed to a conversation-training supplement that was similar in length to the strategic supplement of the strategy training group (usually 15 to 20 minutes). Hence, they engaged in communication activities, such as pair work and group work, for a longer period than did the students in the strategy training group. When reviewing their lessons, the instructor summarized the contents, and the students did not self-evaluate. The students in the control group focused more on conversation with peers and spent more time practicing speaking than did the strategy training group. Therefore, we can say that the students in the control group were expected to learn English by using the target language as much as possible in authentic interaction but that they had much less time for reection. Test Items and Assessment Scale Speaking Tasks. All participants were asked to complete simulated authentic conversation tasks

The Modern Language Journal 89 (2005) on both a pretest and posttest to determine whether they were able to improve their speaking ability over 12 weeks. Different tasks (see Appendix B) were used for the pretest and the posttest to avoid improvement of scores through familiarization with the test content. The difculty of these two tests was examined in the pilot study and no signicant difference was found between them. The tasks were similar to daily classroom activities. Students were given a card describing a hypothetical situation that they might encounter while traveling alone in a foreign country. They were given 5 minutes to prepare a role-play in which the student test takers assumed the role of a customer and the interviewer was a clerk. The student and interviewer engaged in a simulated conversation derived from a situation described on a card. The interviewer in the role-play tasks did not carry out any assessment during the conversation; instead, the interaction was recorded on videotape. Assessment Procedures. I used the Oral Communication Assessment Scale for Japanese EFL Students (see Appendix C), which was established by an action research project at the college (Nakatani, 2002). This scale consists of seven different levels and focuses on the learners uency, ability to interact with the interlocutor, and exibility in developing dialogue. Two independent assessors, who were native speakers of English, did the scoring. Neither was involved in the tests. Each rater was asked to watch the video of the tasks and to score the rst 6 minutes of each participants conversation. The raters were not given any information about the candidates English prociency so there was no halo effect. The interrater reliability of the pretest, estimated by Cronbachs alpha, was .896, a high degree of coefciency. DISCOURSE DATA All the videotaped pretests and posttests were transcribed and analyzed. The transcripts consisted of the 62 participants pre- and posttest discourse. A third trained English-speaking observer reviewed the transcripts while watching the videotapes and focused specically on the segmentation and content of each utterance. A detailed analysis of the discourse in the oral prociency tasks was conducted as follows: (a) The quantity of speech production was measured by the number of words per c-unit (an utterance, such as a word, phrase, or sentence, that gives referential or pragmatic meaning to interaction [see Brock, 1986]) in each participants transcript; and (b) the degree to which the participants

Yasuo Nakatani exhibited different patterns of achievement and reduction strategy use was analyzed on the pretest and posttests. Speech Production The participants speech production rates were counted by the number of words per c-unit, which indicates how many words the students used for an utterance. C-unit analysis was useful for assessing the Japanese EFL students performance because their discourse consisted of many 1-word utterances and incomplete sentences. The participants false starts, slips, and unnecessary selfrepetitions for buying time were excluded from the number of words because they did not seem to have any pragmatic meaning. For example, llers whose use was not appropriate in English but was rather inuenced by Japanese such as Ee? Really? were excluded. Strategy Use In the evaluation of the participants strategy use, the focus was on how strategies were used for the purpose of communication and on how this use represented the extent of discourse in the oral prociency tests. According to previous research dealing with transcription data analysis, strategies for communication have been categorized into achievement and reduction strategies (see, e.g., D rnyei, 1995). The general consensus is that the o former present learners active behavior in repairing and maintaining interaction, and the latter reect learners negative behavior as they try to avoid solving communication difculties, which is a common behavior among low-prociency learners. These two types of strategies observed in the transcription data of this study were further subcategorized into several groupings based on previous representative studies (e.g., Bialystok, 1983; D rnyei & Scott, 1997; Faerch & Kasper, 1983; o Tarone, 1983). A detailed explanation of these strategies and of the examples collected in the current research follows. Achievement Strategies The following categories were classied as achievement strategies: help-seeking, modied interaction, modied output, time-gaining, maintenance, and self-solving strategies. Help-Seeking Strategies. The help-seeking strategies were of two types: an appeal for help and asking for repetition. The former was used when seeking

81 an interlocutors assistance in solving problems caused by the lack of target language knowledge. The latter was used when the participant did not hear or understand what the partner had said. These strategies are exemplied by the following student utterances.
Appeal for help: Im sorry, I dont understand. Asking for repetition: I beg your pardon?

Modied Interaction Strategies. The modied interaction strategy was the process whereby the students sent signals for negotiation in order to overcome communication difculties. This process included conrmation checks, comprehension checks, and clarication requests. Conrmation checks are used to conrm that the speaker has understood something correctly, for example My reservation no? No bargain? Comprehension checks are used to see if the listener has understood correctly, for example, I have a little money, so change to double room. Do you see? Clarication requests ask for an explanation when the speaker does not entirely comprehend something, for example, Why? What kinds of tour? Modied Output Strategies. When using this type of strategy, the participants rephrased an utterance in response to their conversation partners signals for negotiation. The students were given opportunities to produce specic grammar points in creative and complex ways when speaking in the target language, which could lead them to improve their interlanguage. Modied output is exemplied as follows.
Customer (student): 10 oclock? I heard 9 oclock. Travel agent (interviewer): Which one? Pardon? Customer: I heard the ight time is 9 oclock.

Time-Gaining Strategies. When the speakers had difculties expressing an idea, they used these strategies to give themselves time to think and to keep the communication channel open. The conscious use of llers such as Well, let me see . . . and lled pauses such as Oh . . . enabled them to keep the conversation going. Time-gaining strategies are exemplied as follows.
Travel agent (interviewer): When will you start? Customer (student): Let me see . . . tomorrow.

Maintenance Strategies. Maintenance strategies consisted of two types: providing active response and shadowing . The former entailed making positive comments or using other conversation gambits such as I know what you mean and Sounds

82 good. The latter type presented exact, partial, or expanded repetitions of the interlocutors preceding utterance in order to show the listeners understanding of important issues. Therefore, shadowing was functionally different from other types of repetition such as false starts and selfrepetitions. The following exemplies providing active response.
Customer (student): Really? Customer (student): I see, OK .

The Modern Language Journal 89 (2005) They sometimes paused for a long time without appealing to the interlocutor to help nish the utterance. In the worst case, they kept silent without any response. These can be seen in the following example.
Travel agent (interviewer): . . . Also we request our customer to pay beforehand. Customer (student): . . . before . . . [long pause]

Shadowing can be seen in the next example.

Travel agent (interviewer): We have a bargain tour for four days. Customer (student): Four days. Ah . . . OK.

First-Language-Based Strategies. These strategies consisted of interjections in Japanese for a lexical item when the learner experienced communication difculties. The students occasionally used Japanese either intentionally or unintentionally. Interjection of Japanese words:
Hotel clerk (interviewer): . . . Anything else? Customer (student): How can I go . . . [pause] minato (harbor) . . . yotto (yacht) . . .

Self-Solving Strategies. When the learners encountered difculties caused by their own insufcient linguistic resources, they used these strategies to solve the problems without their interlocutors help. They tried to nd relevant linguistic items or expressions by using paraphrase, approximation, and restructuring . Paraphrasing took the form of exemplication or circumlocution for describing characteristic properties or functions of the intended term. In using approximation the learners used an alternative expression that had semantic features similar to those of the intended term. In restructuring, the learners changed to another expression in order to communicate the intended message when they realized their problem in completing a sentence.
Paraphrase: Trying to explain the word harbor Customer (student): the place for ships . . . like bay (instead of harbor ) Approximation of the word accept Customer (student): Do you available travelers check? Restructuring of request Customer (student): May I see . . . sorry, can I use travelers check?

Interlanguage-Based Reduction Strategies. When the learners faced communication problems due to a lack of linguistic resources, they sometimes coped by using their interlanguage system to reduce intended utterances and avoided using certain language structures or specic topics. By cutting out some intended elements, they occasionally produced inappropriate word order based on their interlanguage system. For example:
Travel agent (interviewer): . . . and a standard 3-day tour costs $200. Customer (student): More more cheaply. Travel agent (interviewer): The ight arrives at L.A. at 10 oclock. Customer (student): I . . . I heard leaves L.A. at 9 oclock.

Reduction Strategies The following strategies were categorized as reduction strategies: message abandonment strategies, rst-language-based strategies, interlanguage-based reduction strategies, and false starts. Message Abandonment Strategies. The students used these strategies to avoid engaging in communication when they faced problems in the target language. When they were not able to nd appropriate forms or rules, they stopped speaking in midsentence and left a message unnished.

False Starts. False starts referred to occasions in the conversational discourse when the learners ran into difculties in executing their utterance and repeated one or more of the preceding words. This repetition caused disruptions in their plans for producing the intended utterances accurately. The learners sometimes used false starts with pauses, and occasionally they used them when they realized that there were problems with the expression they were using, such as: I . . . I don t . . . I dont breakfast . . . I have . . . I dont have . . . Retrospective Verbal Protocol Verbal protocol analysis has been recognized as an essential method in investigating learners intentions behind their strategy use during

Yasuo Nakatani communicative tasks (e.g., Clennel, 1995; Cohen, 1998; OMalley et al., 1985). This method can provide researchers with useful information regarding how and why learners choose specic strategies. When this information is compared with learners actual discourse data, it can provide researchers further insight into the EFL learners conscious strategy use. In this study, all students were asked to review their performance on the pre- and posttest tasks by listening to the audiotape recording made during the tests. They were instructed to record their thoughts, in Japanese, on another tape while listening to their performance. They were supposed to report what they were thinking during the tests when they encountered interaction difculties and what their reactions were to their communication problems. These retrospective verbal reports were transcribed and used to understand the students reasons for their strategy use and personal reactions to them. The verbatim transcripts were coded for the appearance and incidence of OCSs by using the taxonomy presented in the preceding section of this article. An additional, independent reader who was asked to examine the transcripts independently veried the transcriptions and provided similar results on the coding. I have translated these retrospective data into English in this article. RESULTS AND DISCUSSION Research Question 1: The Effect of Strategy Training on Speaking Prociency Paired-samples t tests (two-tailed) were introduced to examine whether there was a signicant difference in score gains within each group. These results are presented in Table 1. The improvement in the students speech scores was signicant in the strategy training group (mean gain: 1.38, t = 4.11, p < .01). By contrast, there was no signicant change in the control group scores (mean gain: 0.25, t = 1.05). The interrater reliability of the posttest, estimated by Cronbachs alpha, was .92, a high degree of coefciency, which clearly
TABLE 2 Comparison of the Two Groups Production Rate on Pre- and Posttest by t tests Strategy Training Group (n = 28) M Pretest Posttest 2.49 2.95 SD 0.43 0.41 Control Group (n = 34) M 2.35 2.45 SD 0.63 0.5 t 1.1 4.3

83

p ns <.01

demonstrates that raters were able to differentiate between the two groups with enough consistency. In short, the students in the strategy-training group improved their prociency in the oral communication tests signicantly more than those in the control group. Instruction based on OCSs seemed to facilitate target language development during the simulated tasks. The lack of a signicant improvement in the control group indicates that simply offering students communication practice was not sufcient to develop their speaking ability. Research Question 2: The Impact of Strategy Training on Students Discourse Production Rate. Table 2 shows the results of independent two-tailed t tests comparing the two groups production rates on the pre- and posttests. As can be seen in the table, there was no signicant difference between the two groups on the pretest. In contrast, on the posttest, the strategy training group produced more words per cunit than the control group (p < .01). The results indicate that, through explicit strategy training, the students in the strategy training group learned to make longer utterances that enhanced their abilities to negotiate meaning and maintain the conversation ow than the control group. Changes in Strategy Use. In order to examine the differences between the two groups, a repeatedmeasures two-way ANOVA with one betweensubject factor (training) and one within-subject

TABLE 1 Results of t tests on Test Score Gains between the Two Groups Group Strategy Training Group (n = 28) Control Group (n = 34) df 26 32 Pretest M (SD) 2.23 (1.25) 2.41 (1.54) Posttest M (SD) 3.61 (1.59) 2.66 (1.35) Gain 1.38 0.25 t 4.11 1.05 p .01 ns

84

The Modern Language Journal 89 (2005)

TABLE 3 Means and Standard Deviations of Strategy Use on Pre- and Posttests Strategy Training Pretest M Achievement Strategies Help-Seeking Modied Interaction Modied Output Time-Gaining Maintenance Self-Solving Total Reduction Strategies Message Abandonment First-Language-Based Interlanguage-Based Reduction False Starts Total 0.64 2.46 0.18 0.14 3.46 0.96 7.84 18.4 1.21 5.82 3.93 29.36 SD 0.9 2.9 0.4 0.5 4 0.8 5.5 7 2 2.9 3 9 Posttest M 0.96 4.82 1.79 2.21 8.61 1.14 19.53 9.57 0.5 4.82 3.11 18 SD 1.1 2.8 1.1 2 6.3 1.2 8.7 4.5 0.9 2.1 2.4 6.1 Pretest M 0.79 2.06 0.09 0.56 2.53 1.44 7.47 19.9 1.85 3.5 2.59 27.84 SD 1.0 2.4 0.3 1.3 3.5 1.6 7.2 8.6 2.4 2.4 3.1 9.9 Control Posttest M 0.71 1.76 0.29 0.44 3.53 1.38 8.11 20.4 2.15 3.5 3.09 29.14 SD 0.9 2.2 0.6 0.8 3.4 1.4 6 6.6 2.8 2.6 2.2 8

factor (test) was conducted for each category of strategies. Table 3 presents the means and standard deviations for each achievement and reduction strategy used on the pre- and posttests by both groups respectively. Table 4 presents a summary of the results of ANOVAs for each category of these strategy types. (See Appendix D for the full results of ANOVAs for each category.) Among the achievement strategies, the results of the ANOVAs revealed that there was a signif-

icant effect for training (p < .01), a signicant effect for test (p < .01), and signicant interaction between test and training (p < .01) in the following strategies: modied interaction, modied output, time-gaining, and maintenance strategies. These results indicate that the strategy training group signicantly increased their use of these strategies compared to the control group. No signicant difference was found in the use of helpseeking strategies or self-solving strategies.

TABLE 4 Summary of ANOVA Results for Categories of Strategy Use F ratio Training Achievement Strategies Help-Seeking Modied Interaction Modied Output Time-Gaining Maintenance Self-Solving Total Reduction Strategies Message Abandonment First-Language-Based Interlanguage-Based Reduction False Starts Total Note. df = 1. p < .05. p < .01. 0.08 9.72 45.21 8.3 9.27 1.62 15.83 15.72 5.64 13.55 1.42 29.63 Test 0.52 8.57 65.5 19.39 33.97 0.12 60.83 26.05 0.57 1.52 0.18 216.36 Interaction 1.6 14.16 39.12 24.34 15.45 0.47 44.04 32.28 3.27 1.52 3.03 28.95

Yasuo Nakatani Regarding the use of reduction strategies, the results of the ANOVAs indicate that the strategy training group decreased the number of reduction strategies, especially message abandonment strategies, compared to the control group (p < .01). No change was found in the use of rst-language-based, interlanguage-based, or false start strategies, which suggests that the use of these strategies could not be signicantly changed by the strategy training. Examination of the transcription data indicates that the strategy training helped learners improve their ability to develop discourse by applying achievement strategies and avoiding reduction strategies. In particular, the students in the strategy training group signicantly increased their use of strategies for negotiation in order to solve communication difculties and avoid misunderstanding. For instance, they came to use more conrmation and comprehension checks after the training. It can be assumed that they actively conrmed the interlocutors utterances in order to improve their comprehension. These strategies enabled the students to receive precise, modied input concerning the interlocutors previous utterance. The students also learned to modify their utterances actively when they received negotiation signals from the interlocutor. They appeared to make use of these opportunities to try out new forms and modify other forms, thereby adjusting their utterances to achieve mutual comprehension. Moreover, these students reduced the number of conversation gaps in their discourse by using strategies for taking the oor, buying time to think of an appropriate expression, and responding exibly, tactics that enabled them to maintain the conversation ow. It seems reasonable to say that this explicit strategy training can enhance EFL learners OCS use and thus help to develop their target language interaction. Research Question 3: Student Perceptions of Test Performance and Strategy Use The focus of the verbal protocol analyses presented in this article is on qualitative examples of the strategies that provide evidence of the learners awareness of strategy use. The following verbal protocol data for the pre- and posttests present what participants in each group thought when facing communication difculties and how they reacted to their communication problems in the conversation tests. The presentation of these results consists of two sections. First, pretest verbal protocol data are used to examine how the students perceived their task performance before the

85 training. Then, their posttest data are analyzed in order to compare the types of strategies used by the two groups. Before the Training . When reviewing their pretests, the participants in both groups clearly reported their communication problems. The results indicate that before the course they sometimes lacked sufcient linguistic knowledge for spontaneous communication or that they lacked strategic knowledge to maintain their interaction before the course, or both. They felt under pressure to produce the FL accurately but could not always nd ways to respond to unpredictable situations. They may have been unable to respond appropriately because they had rarely used English to make real decisions about what they wished to achieve, and they had rarely adjusted their language according to such decisions. As seen in the following pretest verbal protocol data, neither the strategy training group nor the control group reported using any achievement strategies; they used only the message abandonment strategy. Strategy Training Group
Ive learned English for 6 years, but it is still very difcult for me to make myself understood in English. In this task, there were many unexpected questions and responses from the hotel clerk. I often lost track of what I was saying. I attempted to use appropriate forms again and again but I got confused and I didnt. I spoke disjointedly with many pauses. I was frustrated because I couldnt say what I wanted to say.

Control Group
I couldnt catch what the speaker said because he spoke too fast, and so I couldnt respond. I just waited until the speaker gave me some help. I took too much time to think how to make English sentences. I feel strongly that I lack knowledge of English vocabulary and grammar. When I was asked difcult questions, I had a very hard time thinking what kind of language I should use in such unexpected situations.

After the Training . Students in the strategy training group reported that they had learned to use achievement strategies on the posttest. In particular, they showed awareness of using modied interaction, modied input, time-gaining, and maintenance strategies in order to solve their communication difculties.

86 The following examples illustrate ways in which they negotiated meaning with their interlocutors by using modied interaction and modied output. The rst student example shows the use of clarication requests that could help mutual understanding. The second student example shows her positive attitude toward providing comprehension checks to maintain discourse. The third student example illustrates the use of conrmation checks learned during the strategy training. The fourth and fth student examples illustrate the use of modied output in order to adjust to an appropriate way of speaking when receiving signals for negotiation. Modied Interaction
S1: When I did not understand what the speaker said, I tried to request his explanation. I could manage to carry on my conversation by signaling my difculties to the partner. S2: I often checked whether I could make myself understood. I paid attention to whether the listener followed my speech. Such behavior seemed to help us maintain the conversation. S3: When I heard a difcult expression, I didnt let it pass. I attempted to check the meaning by repeating a part of speakers utterance. This is a useful technique that I have learned during lessons.

The Modern Language Journal 89 (2005)

tion breakdowns as they were easy to use. I had still problems making myself understood in English.

It might therefore be difcult to conclude that strategy training affects all students positively because such training might not t some students general approaches to learning. There were also instances when the students seemed to realize that they could better maintain and develop their conversation by using maintenance strategies. In these instances, the rst student reported that she attempted to express things in her own way. The second student indicated that she made efforts to keep the conversation owing by reacting smoothly and signaling her understanding of the important point. The third student clearly noticed the effectiveness of shadowing. Maintenance Strategies
S1: I frequently signaled my understanding and tried to react smoothly. I realized that I could express myself in my own words to some extent. S2: I signaled that Id understood by nodding and giving positive responses such as Yes, Ah. I often used shadowing in order to conrm that Ive understood. S3: I think shadowing is a very useful strategy. By using it, I could interact well and make our conversation uent.

Modied Output
S4: I tried to change my utterances when the clerk couldnt understand my intention. After the training, I got used to helping my communication partner understand what I wanted to say. S5: I attempted to repeat or change the speakers previous utterance in order to check whether my understanding was correct.

With respect to gaining time to think of an appropriate expression, the following data indicate that students in the strategy training group used llers and lled pauses. They became aware of how to avoid communication breakdowns by using time-gaining strategies. However, the third student reported that with the use of some strategies, it was not always easy to improve the form of her utterances. Time-Gaining Strategies
S1: When I needed time to think, I said well and let me see instead of keeping silent. S2: I tried not to make communication gaps. I intentionally used llers. S3: It was not so easy to use some strategies spontaneously. I could only use llers to avoid communica-

As seen in the following examples, some students in the control group indicated an improvement in their attitude on the posttest. They seemed to be able to control affective factors such as anxiety. They had been involved in communication-centered lessons, and they appeared to have gained condence expressing themselves in English. However, they reported using few strategies for maintaining interaction when they had communication difculties. They did not seem to recognize any particular strategies for solving problems, such as negotiation of meaning or gaining time to think. They reported that they had no choice but to use message abandonment strategies when facing conversation problems.
S1: Last time I was totally confused and I couldnt say what I wanted. So this time, I tried to relax during the conversation and I spoke more smoothly. S2: Generally, I feel much better about speaking English, but I sometimes stopped my conversation. S3: I couldnt speak well in the test. When I didnt understand the speakers utterances, I lost my words. I got into a sort of panic.

Yasuo Nakatani
S4: When I didnt understand a word I paid too much attention to it, and consequently I couldnt say anything. S5: When I had trouble understanding the interlocutors utterance, I couldnt respond and became silent. I paused a lot in unexpected situations, which made my conversation awkward.

87 fewer reduction strategies, such as message abandonment. Although we must recognize that training does not always improve learning for all students, the retrospective protocol data indicated that the students at least became aware of specic strategies that they could use to improve their discourse. It can be concluded that training focused on conscious practice in using OCSs is likely to improve Japanese female EFL learners communication during simulated tasks. The ndings of the present study suggest that EFL learners who lack metacognitive skills need to learn to recognize and analyze specic linguistic and sociolinguistic cues in order to comprehend and integrate input into their schemata. They should consciously use their interlanguage system to control their performance and to maintain interaction. In order to achieve these goals, learners strategic competence can be developed through raising their awareness of managing and supervising specic strategy use. Therefore, if the goal of instruction is to offer opportunities to students to acquire independent learning skills, they can develop metacognitive strategies in order to make plans, monitor, and evaluate their interaction for future target language learning contexts. Given that this type of strategy research is still in its initial stage, further in-depth investigations should be pursued to add to the ndings of the present study. In particular, the current study was conducted with a rather small number of participants within a short period of time. For example, because the research was conducted in real classroom settings within 12 weeks, it was not possible to conduct a delayed posttest that could provide information concerning the longitudinal effects of the strategy training on the students oral prociency. It is important, therefore, to examine whether the training groups advantage lasts for a long time, and whether the OCSs that they learned are accessible for their future target language study beyond the classroom. Another vital area of future research should focus on students target language learning processes in the strategy training lessons. Detailed and precise information on what actually occurs in the classroom is important for validating the research results. It might be useful to make videos of student performance in classrooms and collect ethnographic data. Furthermore, this case study dealt mainly with meaning-focused strategy training and thus there is room for further investigation regarding the impact of strategy training on the forms of utterances.

In summary, only students in the strategy training group students became aware of how to use achievement strategies and avoid reduction strategies. Accordingly, the students reports about strategy use were almost consistent with their actual performance in the discourse data. Given that EFL learners tend to face many communication breakdowns, they need to acquire such skills in order to maintain and develop their conversational interactions. During the interactions carried out in the training tasks, the students often struggled to produce the target language, and they were given numerous opportunities to overcome their communication difculties by using strategies. As seen in the retrospective verbal protocols, some students in the strategy training group readily recognized the usefulness of OCSs, which led them to use the strategies consciously. Therefore, the success of the strategy training group could be attributed in part to their conscious use of strategies during conversation. They increased the number of strategies they used to maintain conversation ow and solve potential communication problems. It is clear that this increase was the effect of metacognitive strategy training. Through such training, the students learned not only which type of strategy to use but also how to use it appropriately.

CONCLUSION This study investigated the impact of oral communication strategy training on the discourse of Japanese EFL learners, who are believed to be relatively ineffective speakers of English in comparison with some other EFL learners. The ndings indicate that the students in the strategy training group signicantly improved their oral test scores as compared to the students who did not have the training. The transcription analysis of the discourse also revealed that the students in the strategy training group came to make longer utterances and use more achievement strategies, such as modied interaction, modied output, timegaining, and maintenance strategies than the control group. At the same time, this group used

88
ACKNOWLEDGMENTS I wish to express my gratitude to Mr. Charles Owen and Dr. Jeannette Littlemore of the University of Birmingham for their insight, comments, and suggestions. I would also like to thank Dr. Ernesto Macaro of University of Oxford for his helpful comments on my research. I am solely responsible for any remaining errors and omissions.

The Modern Language Journal 89 (2005)

Ellis, G., & Sinclair, B. (1989). Learning to learn English. Cambridge: Cambridge University Press. Faerch, C., & Kasper, G. (1983). Strategies in interlanguage communication. London: Longman. Gass, S. M. (1998). The role of interaction in native speaker comprehension of nonnative speaker speech. Modern Language Journal, 82, 308319. Kitajima, R. (1997). Inuence of learning context on learners use of communication strategies. JALT Journal, 19, 723. McDonough, S. H. (1995). Strategy and skill in learning a foreign language. London: Edward Arnold. Nakahama, Y., Tyler, A., & van Lier, L. (2001). Negotiation of meaning in conversation and information gap activities: A comparative discourse analysis. TESOL Quarterly, 35, 377405. Nakatani, Y. (2002). Improving oral prociency through strategy training . Unpublished doctoral dissertation, The University of Birmingham, England. OMalley, J. M., & Chamot, A. U. (1990). Learning strategies in second language acquisition. Cambridge: Cambridge University Press. OMalley, J. M., Chamot, A. U., Stewner-Manzares, G., Russo, R. P., & Kupper, L. (1985). Learning strategies applications with students of English as a second language. TESOL Quarterly, 19, 557584. Oxford, R. L. (1990). Language learning strategies: What every teacher should know. New York: Newbury House. Oxford, R. L. (1996). Language learning strategies around the world: Cross-cultural perspective. Honolulu, HI: Second Language Teaching and Curriculum Center. Pica, T. (1996). The essential role of negotiation in the communicative classroom. JALT Journal, 18, 241 268. Poulisse, N. (1990). The use of compensatory strategies by Dutch learners of English. Enschede, the Netherlands: Sneldruk. Rost, M., & Ross, S. (1991). Learner use of strategies in interaction: Typology and teachability. Language Learning, 41, 235273. Rubin, J. (1975). What the good language learner can teach us. TESOL Quarterly, 9, 4151. Schmidt, R. (1990). The role of consciousness in second language learning. Applied Linguistics, 11, 129 158. Tarone, E. (1983). On the variability of interlanguage systems. Applied Linguistics, 4, 143163. Wenden, A. (1991). Learner strategies for learner autonomy. New York: Prentice-Hall International. Wenden, A. (1999). An introduction to metacognitive knowledge and beliefs in language learning: Beyond the basics. System, 27, 435441. Willems, G. (1987). Communication strategies and their signicance in foreign language teaching. System, 15, 351364. Wong-Fillmore, L. (1979). Individual differences in second language acquisition. In C. Fillmore, D. Kempler, & W. Wang (Eds.), Individual differences in language ability and language behavior (pp. 203 228). New York: Academic Press.

REFERENCES Bachman, L. F. (1990). Fundamental considerations in language testing . Oxford: Oxford University Press. Bachman, L. F., & Palmer, A. S. (1996). Language testing in practice. Oxford: Oxford University Press. Bejarano, Y., Levine, T., Olshtain, E., & Steiner, J. (1997). The skill use of interaction strategies: Creating a framework for improved small-group communication interaction in the language classroom. System, 25, 203213. Bialystok, E. (1978). A theoretical model of second language learning. Language Learning, 28, 6983. Bialystok, E. (1983). Inferencing: Testing the hypothesis-testing hypothesis. In H. W. Seliger & M. H. Long (Eds.), Classroom oriented research in second language acquisition (pp. 104123). Rowley, MA: Newbury House. Brock, C. (1986). The effect of referential questions on ESL classroom discourse. TESOL Quarterly, 20, 47 59. Brown, A. L., & Palincsar, A. S. (1982). Inducing strategies learning from texts by means of informed, self-control training. Topics in Learning and Learning Disabilities, 2, 117. Canale, M. (1983). On some dimensions of language prociency. In J. W. Oller, Jr. (Ed.), Issues in language testing research (pp. 333342). Rowley, MA: Newbury House. Canale, M., & Swain, M. (1980). Theoretical bases of communicative approaches to second language teaching and testing. Applied Linguistics, 1, 147. Clennel, C. (1995). Communication strategies of adult ESL learners: A discourse perspective. Prospect, 10, 420. Cohen, A. D. (1998). Strategies in learning and using a second language. Essex, England: Longman. Cohen, A. D., Weaver, S. J., & Li, T.-Y. (1998). The impact of strategies-based instruction on speaking a foreign language. In A. D. Cohen (Ed.), Strategies in learning and using a second language (pp. 107156). Essex, England: Longman. D rnyei, Z. (1995). On the teachability of communicao tion strategies. TESOL Quarterly, 29, 5585. D rnyei, Z., & Scott, M. L. (1997). Communication o strategies in a second language: Denitions and taxonomies. Language Learning, 47, 173210.

Yasuo Nakatani APPENDIX A Strategy Diary

Date: Review: What did you learn last week? Objectives of this lesson 1. Useful expressions and grammar 2. Words and phrases Metacognitive Training 1. Before the task: Preview your performance. Goal and procedures: Required linguistic resources: Planning how you should react to your partner: Advance organization: Utilize your knowledge of the context. 2. During the task: Interact with your partner. (a) Attending to the oral communication strategies: Appeal for help: Asking for repetition: Conrmation checks: Comprehension checks: Clarication requests: Using llers: Maintenance: Offering assistance: (b) Monitoring: Monitor your comprehension and production. 3. After the task: Self-assessment. Review your performance and evaluate your oral communication strategy use. Evaluate your task objectives. 4. Assimilation: Reect on what you have learned.

89

APPENDIX B Oral Communication Tasks for Pre- and Posttest

Pretest: Hotel Check-In Role A: You are on a trip to Australia. You need to check into a resort hotel. Two months ago, you booked a single room from Japan by telephone for two nights. You did not ask the price of the room at that time but your travel guidebook has information about the prices, which states that your desired room costs 20 Australian dollars per night. Please use a travelers check for payment and ask about the time of breakfast tomorrow morning. You intend to join the yacht tour, which your guidebook states will start at 8:00 a.m. Role B: You are the hotel clerk at the Oasis Inn. The hotel has undergone many changes recently and the hotel clerical staff have been replaced; thus, there is no previous information about the customers. You should not accept checks from foreigners. As this is a resort area, breakfast will be served relatively late at 8:00 a.m. The hotel has increased the room rates. Rates: Single room $30, Double room (no shower) $40, Twin room (shower) $45, Suite room $80. (All single rooms are booked for tonight.) Posttest: Travel Agency Role A: You are visiting a travel agency in San Francisco. One month ago you booked a cheap tour to Los Angeles for three days starting tomorrow. This tour was advertised in a newspaper ad at $150. You have come here to get the travel voucher. Please use a travelers check for payment and ask about the ight schedule for tomorrow. You intend to go to the Disneyland tour, which your guidebook states will start in L.A. at 9:00 a.m. You are on a tight budget. Role B: You are working at a travel agency. You are a new employee and do not know how to access the customers data, which the clerk before you used. You can only accept cash or a credit card. All bargain 3-day tours starting tomorrow are booked. The following tours are available: Bargain tour for 4 days: $200, the ight leaves at 10 a.m. and arrives at 11 a.m. Standard tour for 3 days: $220, the ight leaves at 10 a.m. and arrives at 11 a.m.

90 APPENDIX C Assessment Scale

The Oral Communication Assessment Scale for Japanese EFL Students Level 7

The Modern Language Journal 89 (2005)

Almost always communicates effectively in the task Speech is generally natural and continuous. Can interact in a real-life way with the interlocutor. Can generally develop the dialogue spontaneously with few errors. Generally communicates effectively in the task Is not quite uent but interacts effectively. Can generally react exibly. Makes a positive contribution to the dialogue. Communicates reasonably effectively in the task Is sometimes uent but with hesitancies. Can interact fairly comfortably and gain exibility. Makes some contribution to the dialogue. Communicates moderately effectively in the task Makes some pauses but fairly intelligible. Shows some exibility. Is somewhat independent of the interlocutor in the dialogue. Communicates modestly in the task Makes frequent pauses but somewhat intelligible. Shows little exibility. Can maintain dialogue but in a rather passive way. Communicates marginally in the task Makes numerous pauses, at times long ones. Still depends on the interlocutor but begins to interact a little with him/her. Given help, communicates quite basically. Requires some tolerance from the interlocutor. Communicates extremely restrictedly in the task Can answer simple questions but with numerous long pauses. Depends on interlocutor with only partial contribution to dialogue. Some questions have to be repeated or rephrased.

Level 6

Level 5

Level 4

Level 3

Level 2

Level 1

Note. From Nakatani (2002).

APPENDIX D Results of Repeated-Measures Two-Way ANOVAs

Help-Seeking Strategies Source Training (between groups) Test (within groups) Training Test Modied Interaction Strategies Source Training Test Training Test Modied Output Source Training Test Training Test SS 19.21 25.24 15.07 df 1 1 1 MS 19.21 25.24 15.07 F 45.21 65.50 39.12 p .000 .000 .000 (Continued) SS 92.03 32.68 53.97 df 1 1 1 MS 92.03 32.68 53.97 F 9.72 8.57 14.16 p .003 .005 .000 SS 0.09 0.42 1.29 df 1 1 1 MS 0.09 0.42 1.29 F 0.08 0.52 1.60 p .777 .475 .211

Yasuo Nakatani
Time-Gaining Strategies Source Training Test Training Test Maintenance Strategies Source Training Test Training Test Self-Solving Strategies Source Training Test Training Test Achievement Strategy Total Source Training Test Training Test Message Abandonment Strategies Source Training Test Training Test First-Language-Based Strategies Source Training Test Training Test Interlanguage-Based Strategies Source Training Test Training Test False Start Source Training Test Training Test Reduction Strategy Total Source Training Test Training Test SS 1950.48 9786.87 1309.67 df 1 1 1 MS 1950.48 9786.87 1309.67 F 29.63 216.36 28.95 SS 14.18 0.79 13.41 df 1 1 1 MS 14.18 0.79 13.41 F 1.42 0.18 3.03 SS 101.88 7.68 7.68 df 1 1 1 MS 101.88 7.68 7.68 F 13.55 1.52 1.52 SS 40.11 1.36 7.81 df 1 1 1 MS 40.11 1.36 7.81 F 5.64 0.57 3.27 SS 1185.12 530.83 657.79 df 1 1 1 MS 1185.12 530.83 657.79 F 15.72 26.05 32.28 SS 1123.82 1222.69 885.14 df 1 1 1 MS 1123.82 1222.69 885.14 F 15.83 60.83 44.04 SS 3.94 0.11 0.43 df 1 1 1 MS 3.94 0.11 0.43 F 1.62 0.12 0.47 SS 277.55 289.71 131.77 df 1 1 1 MS 277.55 289.71 131.77 F 9.27 33.97 15.45 SS 14.14 29.31 36.79 df 1 1 1 MS 14.14 29.31 36.79 F 8.30 19.39 24.34

91

p .005 .000 .000

p .003 .000 .000

p .208 .730 .495

p .000 .000 .000

p .000 .000 .000

p .021 .454 .076

p .000 .223 .223

p .238 .673 .087

p .000 .000 .000