1. Introduction
As part of a series in Language Teaching dedicated to making research in periphery countries available to a wider international readership, this study analyzes research in second/foreign language teaching and learning conducted in Taiwan between 2004 and 2009. Since, in Taiwan, ‘foreign language education’ generally means ‘English education’ (Su Reference Su2000), this study is limited to a review of local research in English teaching and learning. ‘Local research’ in this paper refers to research projects conducted in Taiwan and published in Taiwanese journals or the proceedings of conferences held in Taiwan, with Taiwanese students and teachers as the main subjects of study, written either in Chinese or English. Research conducted in Taiwan but published in foreign journals (e.g. TESOL Quarterly) will not be included in this review, except in the case of papers initially published locally and subsequently appearing in international journals (perhaps with revisions).
Local researchers who dedicate themselves to English teaching and learning network through three major English teaching associations: the English Teaching and Research Association, which sponsors the annual International Conference on English Teaching and Learning in Taiwan; the English Teachers’ Association of the Republic of China (ETA–ROC), which sponsors the annual International Symposium on English Teaching; and the Taiwan ESP Association, which publishes a journal of ESP-oriented papers. In addition to the publication sources sponsored by these associations, the main arenas for the publication of English teaching and learning research are national journals or conference proceedings sponsored by individual universities.
With the aim of providing an informative and representative review of the literature, papers from both journals and conference proceedings are selected on the basis of the following criteria: (i) the selected journals should be published regularly, peer-reviewed, and included in the referred journal lists suggested by Wang & Huang (Reference Huang, Chern and Lin2006), except for those that were first issued later than 2006; and (ii) the selected conference proceedings should contain full papers presented at regularly held conferences, which may not necessarily go through a peer review process prior to publication.
The journals chosen for this review are of three types:
(1) English teaching journals: English Teaching & Learning and Taiwan Journal of TESOL. These are ranked nationally as the first and the third in their category respectively (Wang & Huang Reference Wang and Huang2006). Both publications are well-established, good barometers of research in English teaching and learning, and have a wide local readership.
(2) Interdisciplinary journals: traditional Taiwanese journals publishing papers in a range of fields. These journals are accessible to non-specialists, but are essentially institution insider-oriented publications which provide a forum for faculty to publish papers. We have chosen two journals: Soochow Journal of Foreign Languages and Literatures, ranked nationally as the second among journals in this category (Wang & Huang Reference Wang and Huang2006), and National Normal University Journal of Humanities. Both journals are well-established and published by their associated universities.
(3) English for specific purposes (ESP) journals: Taiwan International ESP Journal (TIESPJ) is the only one of its kind in the ESP area, and thus is chosen for this category. The inclusion of this journal reflects the current trend towards the development of ESP in Taiwan.
The conference proceedings chosen for this review are: (1) Proceedings of the International Conference on English Teaching and Learning in the Republic of China; (2) Selected Paper from the International Symposium on English Teaching; and (3) Proceedings of the International Conference and Workshop on TEFL and Applied Linguistics. The first two conferences are sponsored by the English Teaching and Research Association and ETA-ROC, respectively, these being the oldest major national English teaching associations in Taiwan. Partially under the auspices of the Ministry of Education (MOE) and National Science Council of Taiwan, the third conference is organized and supported by Ming Chuan University.
A total of 641 articles published in the journals and proceedings mentioned above were originally deemed relevant to this review. Based on the topics and issues they discuss, these articles were divided into three categories: English-in-education policy/curriculum, research on skills development, and assessment. When the research on English skills development was further classified into reading, writing, speaking and listening skills, it was found that writing skills development has drawn the most attention (n = 222) from researchers, followed by reading skills development (n = 138), then speaking skills development (n = 116), and, finally, listening skills development (n = 47). The marginalization of listening among the four skills is a trend repeatedly reported in the literature (Feyten Reference Feyten1991; H. Cheung Reference Cheung2006; Vandergrift Reference Vandergrift2007; Field Reference Field2008; Lynch Reference Lynch2011).
Over one hundred papers were then carefully selected and reviewed. Although some personal preferences were inevitably involved in making this selection, we decided to highlight studies that we saw as representative of current research interests.
In the following sections, we will first provide some background on English education in Taiwan. This is followed by an analysis of the research in terms of the three categories mentioned earlier: English-in-education policy and curriculum, research on skills development, and assessment. After the analysis, the unique characteristics and shortcomings of research in Taiwan are discussed, and finally, the implications for future research are discussed.
2. English education
Taiwan is a multilingual country where Mandarin Chinese is the official language, and Southern Min, Hakka, and various Austro-Polynesian aboriginal languages, closely related to Mandarin, are in daily use by large sections of the population. As part of ‘the expanding circle’ of World Englishes (Kachru Reference Kachru and Kachru1992: 356), English is categorized as a foreign language (EFL) in Taiwan. Although English is used by only about 2–4% of the total population on a daily basis, it is widely perceived as very useful for one's education and subsequent career (S. C. Chen Reference Chen2010).
The significance of English in Taiwan can be partly attributed to the government's language policy. English was privileged over other foreign languages in formal education in 1862 when China's first public foreign-language school, Tung Wen Kuan, was established in Beijing by the Government of the Ching dynasty. After several changes of government and relocation to Taiwan, the language policy already implemented in China was transmitted and adopted in Taiwan, English being formally listed as the obligatory foreign language to be taught from the beginning of middle school (Taga Reference Taga1976). In 1968, English gained further status after being made an obligatory subject in middle school, with the word ‘English’ being parenthesized after the term ‘foreign language’ in official documents (Ministry of Education 1972, 1973, cited in Tse Reference Tse1987: 87). The curricular standards at this time focused on grammar-translation, with accuracy emphasized over fluency.
English education in Taiwan continued in much the same way through to 1993, when some high schools were permitted to offer other languages as second foreign languages, and some colleges were allowed to choose other languages to fulfill the foreign language requirement. These changes were made in response to a large-scale educational reform driven by the internal democratic process and external economic demands (Law Reference Law2004). Accordingly, the English curriculum for secondary school levels was revised in 1994 and 1995, with the ‘communicative approach’ adopted as the central guiding principle to enhance students’ communicative competence. Meanwhile, the English curriculum at college level shifted from a focus on the improvement of students’ reading skills to an integration of the four language skills.
While the change in the English curriculum as a result of the 1993 mandate allowed both teachers and students to enjoy more freedom in offering and taking English courses, in reality, teaching remained much the same as over the previous twenty years (Tsao Reference Tsao, Kaplan and Baldauf2000). One reason was that English was the only foreign language to be tested in the entrance examinations for the higher level of schooling, so classroom teaching remained test-driven and focused on grammar.
During the 1990s attempts were made to further internationalize Taiwan by advocating English education from an earlier age and promoting improvements in communicative competence. This English language movement was initiated by the government's attempt to promote Taiwan as an Asia-Pacific business center in the global economy (S. C. Chen Reference Chen2006). The importance of English was further promoted by Challenge 2008, a six-year National Development Plan (Government Information Office 2002) which ran from 2003 to 2008, resulting in the establishment of basic English conversation classes for people from all walks of life and attempting to promote internationalization, in which the recruitment of international students in college was emphasized.
3. English-in-education policy and curriculum
In response to the growing status and more widespread use of English in Taiwan, recent research on English teaching and learning has been developed in the context of a new English-in-education policy set in motion by the Ministry of Education (MOE) and implemented in 2001. This policy involved English teaching and learning in elementary schools, an English language exit requirement (stating that college students should obtain the certificate of proficiency in English prior to graduation), and English as a medium of instruction (EMI) for courses in higher education. Discussion of research on these three aspects of the policy will be presented in the following sections.
3.1 English education in elementary schools
English education was officially introduced to elementary schools in 2001, starting in the fifth grade for at least one class period (40 minutes) each week, and then extended to the third grade in 2005 for two class periods per week, although some local governments had already introduced English at lower grade levels in 2001 (Tsao Reference Tsao2004a). English education in Taiwanese elementary schools has the following characteristics:
(1) The implementation of English education at the elementary level is a decentralized policy in which local governments or even individual schools can decide the allocation of resources and process of implementation.
(2) The central guiding principle is the development of students’ oral proficiency over reading and writing skills.
(3) Textbook selection is operated in a free market, so publishers or language institutions, rather than the MOE, create teaching materials to meet the needs of the market.
(4) The goals of the English curriculum include: (i) developing basic communicative skills in English, (ii) cultivating learning interest and positive attitudes, and (iii) promoting the understanding of local and foreign cultures (Ministry of Education 2003).
After several years of implementation, government officials and scholars began to revisit the English-in-education policy for elementary schools and advocated its revision. Tsao (Reference Tsao2004b), focusing on language-planning theory, highlighted one problem of EFL in Taiwan: students have few opportunities to practice English outside the classroom. In his opinion, instead of focusing on the development of oral skills, as specified in the elementary-school English curriculum, students should develop all four language skills (listening, speaking, reading, and writing) simultaneously, in order to lay a solid foundation for English competence. He also suggested that a shift should be made towards the development of students’ reading skills and skills of autonomous learning at high school level and beyond. Tsao's modified model of English skills development threw light on the need to revise the curriculum and stimulated a number of related empirical studies.
Focusing on beginning readers, Zapparoli & Su (Reference Zapparoli and Su2007) observed the English curriculum implemented for first-graders in Hsinchu elementary schools and found that the alphabet and simple vocabulary were learned through traditional pedagogy, which they regarded as inappropriate for the first-grade learners. Based on the results of an experimental study they conducted with these students, they suggested that a story-based curriculum, using stories to play with the rhythms of language and so develop students’ phonological awareness and alphabetic knowledge, may be more effective, attractive, and beneficial to young learners, regardless of their previous English experience. However, their suggestion, while useful, was not widely adopted as there was little support for this change in the curriculum (cf. M. C. Wu & Chen Reference Chen2006).
In a pivotal work, W. C. Chang et al. (Reference Chang, Yeh, Joe, You, Chern and Liaw2007) drew on the English-in-education policy as a framework to identify the problems and difficulties encountered by school teachers, administrators, and students at all levels. Students’ English abilities showed a bimodal distribution, with a low-score peak and a high-score peak within each stage of formal education; this was confirmed in their study as one of the most serious problems in this context, and one which was identified as early as 2002 for junior high school students (C. H. Chang & Yu Reference Chang and Yu2004). Furthermore, the lack of qualified teachers and educational resources, the wide gap in English proficiency between urban and rural students, and the difficulties related to teaching in multilevel classes were all identified as serious problems, leading to a claim of ‘chaos in English education in Taiwan’ (W. C. Chang et al. Reference Chang, Yeh, Joe, You, Chern and Liaw2007: 673). More specifically, this condition was attributed to the lack of a unified English-in-education policy to connect goals in different stages of formal education. After illustrating the lack of coordination between various agencies at different stages of education (the Departments of Elementary Education, Secondary Education, Higher Education, and Technological and Vocational Education), the authors called for a specialized language planning agency that could bridge the gap between the government's English-in-education policy and the understanding of that policy by school administrators and language practitioners.
Other scholars attempted to solve the same problems by promoting remedial instruction. Studies on remedial instruction for underachievers in elementary schools include Chu et al.'s (Reference Chu, Yu, Chang, Ting, Yu and Hu2007) two-year longitudinal study comparing the effectiveness of two phonological remediation programs for 95 elementary school children with poor word recognition skills: systematic phonics instruction (e.g. Wasik & Slavin Reference Wasik and Slavin1993) and a combination of phonics instruction with phonological awareness training. The two groups produced statistically different results, but these did not reach the significance level set, although the participants’ overall word-decoding ability was enhanced by the intervention.
Juan & Cheng (Reference Juan and Cheng2008) explored the effectiveness of extensive reading in a remedial program for elementary-school underachievers. The results showed an increase in students’ letter and vocabulary knowledge, although their performance was still below average. In line with the research on underachievers, Liou et al. (Reference Liou, Chen, Yu and Chang2009) found that students can gain some benefit from remedial programs, which were found to improve aspects of their learning, but to maximize the learning outcome of underachievers, the teachers’ supportive attitudes should not be ignored (Y. K. Hsu & Chen Reference Hsu and Chen2007).
Although remedial instruction programs were first developed for use in elementary schools, they have been extended to secondary schools. For example, H. L. Huang & Chen (Reference Chen2009) explored the effects of storytelling on underachievers’ motivation to learn English in a junior high school. Similarly, C. F. Yu (Reference Yu2005, Reference Yu2008) and Lee et al. (Reference Lee, Leu, Shih and Yu2009) investigated online remedial programs for college students in a freshman English class. All of them found that remedial programs can help the underachievers improve their English. An interest in English education for underachievers across all educational levels has thus become one of the characteristics of current English education in Taiwan.
3.2 English language exit policy
Due to the demands of Challenge 2008, an English exit policy – a gate-keeping device requiring students to take a specified examination in English language (e.g. TOEIC or TOEFL) before graduation – has been promoted by the MOE through projects such as Aim for the Top University and Teaching Excellence. This is currently in operation in many colleges and universities, each of which can set its own standards (M. Chen & Johnson Reference Chen and Johnson2004). Some colleges and universities, for instance, oblige students to pass the first stage of the intermediate General English Proficiency Test (GEPT), and others, the first stage of the high-intermediate GEPT, although this can be replaced by the equivalent level of any other standardized test. A student who fails the specified test cannot graduate until they retake and pass it. While the English exit policy has been widely implemented, it is not taken very seriously. In practice there are many ways to meet the exit requirements, one of which is for students to take a remedial course offered by their individual department or college.
Research has been carried out on the implementation of the English exit policy. For example, S. W. Chen & Liu (Reference Chen and Huang2007) compared undergraduates’ motivation to learn English before and after the adoption of the policy. They divided student into two groups: one of 4,926 students taking the required freshman and sophomore English courses, and the other of 227 students taking any other foreign language courses (e.g. German, Russian, or Spanish). Chen & Liu examined archived teaching evaluations for the years 2002–2005, including itemized teacher evaluations, overall teacher evaluations, and students’ self-evaluations. The results revealed that the English exit policy fostered more positive attitudes toward the process of learning English, English teachers, and language courses. However, it remains unclear whether one group showed significantly greater changes in their attitudes than the other, so further research is required on this issue.
In another study, C. M. Shih (Reference Shih2008) interviewed college students to examine their attitudes toward having to pass the GEPT before graduating. About 60% of the interviewees were found to endorse the exit requirement, but they suggested that the test environments and types of test questions should be improved, and that efforts should be made to ensure that the test be internationally accepted.
To enhance the pass rate, self-access centers and programs have been established to provide a supportive autonomous environment (C. Chao Reference Chao2004), so that students can invest more time and effort in studying English, particularly for those who have so far failed to meet the exit requirement. Principally funded by the MOE but also with support from the universities, these self-access centers have increased rapidly in recent years. The issue of what exactly is available in them has drawn much attention from academics. One of the earliest observational and descriptive research studies in this area was carried out by C. Chao (Reference Chao2005), who analyzed the types of resources available by visiting the campus websites of 107 universities, conducting interviews and observing participants. The results showed that the elements most frequently provided by self-learning centers were: learning resources (e.g. online learning, books, and other facilities), self-assessments of learning outcomes (e.g. contests or simulated tests), one-on-one tutoring (e.g. teacher tutors and peer-tutoring services), simulated naturalistic environments for English learning (e.g. English corners), and learning communities (e.g. student associations and toastmasters groups).
W. W. Cheng & Lee (Reference Lee, Leu, Shih and Yu2009) evaluated university students’ attitudes toward the self-access program offered in a university Multimedia English Learning Center. Drawing on 217 students’ responses to questionnaires, the authors concluded that while the self-access program satisfied the users, other learning-related services should be offered, such as counseling services or assistance from staff or teachers. The results showed that, even in a self-access center, Taiwanese students need teachers and other personnel as well as technology in order to get more involved in self-directed learning. This implied that students’ needs go beyond what such centers can provide.
Despite the popularity of self-access centers, little empirical research has been conducted to investigate either the feasibility of learner-centered study within the highly teacher-oriented and test-oriented educational system that operates in Taiwan, or the actual experience of learners, including the time they spend and the strategies they use to improve learning in such centers. Further studies should attempt to address these areas.
3.3 English as a medium of instruction in higher education
English as a medium of instruction (EMI), also called academic internationalization (Coleman Reference Coleman2006), has been popular in Taiwan as a way of enhancing students’ English ability, promoting the internationalization of the local education system in order to recruit foreign students, and increasing the competitiveness of Taiwanese educational institutes. Like the English exit policy above, the EMI policy was promoted by the MOE through various projects, such as Aim for the Top University, and Teaching Excellence. Academic internationalization, among others, is now one of the evaluation criteria for funding from the MOE. In general, EMI is implemented by providing professors with nominal financial incentives to teach in English. In addition, at some universities (e.g. National Taiwan University), English is the medium of instruction if there are international students in the class.
Using questionnaires, Y. Y. Chang et al. (Reference Chang, Liao, Welsh, Wu, Liu, Venhuizen, Chen and Chen2006) examined the effect of EMI on students’ English proficiency and learning motivation. Although the students complained that they had difficulty understanding English lectures and textbooks, they felt that EMI was helpful in improving their English proficiency. While this research addressed the connection between EMI and English proficiency development, it ignored the other goals of EMI set by the MOE, such as the internationalization and competitiveness of Taiwan's higher education system.
Based on their investigation of teacher beliefs and teaching strategies, S. C. Teng & Wang (Reference Teng and Wang2009) concluded that EMI can help to increase students’ English proficiency. They did not objectively assess students’ actual progress, but were concerned mainly with students’ perceptions of how EMI contributed to their English proficiency. Overall, this line of research did not explore the major goal of EMI as set out by the MOE; that is, whether the policy has enabled the educational system to attract more foreign students. In particular, the connection between EMI and changes in language functions and social identity were left unaddressed. Thus, in many respects, the effectiveness of EMI remains to be explored.
4. Research on skills development
Research into skills development in English is relatively well-established in the area of literacy development in Taiwan, but less emphasis is placed on listening and speaking (H. Cheung Reference Cheung2006). While reading and writing skills are regarded as inextricably linked in the theory of language learning, these two language skills are generally investigated separately in Taiwan. This review mirrors this tendency by discussing the research on reading and writing separately in the following subsections.
4.1 Reading skills
As reading requires both word recognition and comprehension skills (Leipzig Reference Leipzig2001), research on reading skills development can be divided into two corresponding strands: research on vocabulary acquisition through extensive reading, and on the use of reading strategies.
The first strand comprises experimental, classroom-based research, focusing on vocabulary acquisition as it relates to extensive reading (Shen & Yuan Reference Shen and Yuan2005). This type of study has become popular due to the fact that vocabulary knowledge has long been deemed a major determinant of ESL/EFL reading comprehension (Laufer Reference Laufer, Coady and Huckin1997). With financial support from the National Science Council of Taiwan, H. T. Huang & Liou's (Reference Huang and Liou2004) project developed an online extensive reading syllabus for college students to enhance word recognition and extensive reading. While the results were positive, this syllabus has not been extensively adopted by other Taiwanese learners or researchers, probably because the online extensive reading system was discontinued after the completion of the project, so was not really accessible to many Taiwanese. Preliminary results of this study were published in Taiwan, but a revised and more comprehensive report was published in an international journal (H. T. Huang & Liou Reference Huang and Liou2007).
To improve children's word recognition, M. C. Wu & Chen (Reference Chen2006) compared two word-recognition training programs for fourth-grade poor readers: one focusing on training phonological awareness at the onset/rime-level, the other on phoneme-based phonics instruction. In practice, this means that the former teaches students phonetic symbols, while the latter teaches them phonics. Based on the performance of their pre- and post-intervention word recognition tests, phoneme-phonics instruction yielded significantly better results, a finding that contradicts the conclusions in Chu et al. (Reference Chu, Yu, Chang, Ting, Yu and Hu2007), which reported that training in phonological awareness (learning to manipulate individual or larger units of phonetic symbols for pronunciation purposes) can help children pronounce real and pseudo words.
In research on more efficient and effective approaches to vocabulary learning, Min & Hsu (Reference Min and Hsu2008) conducted a quasi-experimental study to compare the effectiveness of two pedagogies: reading plus vocabulary-enhancement activities (RV) vs. narrow reading (NR) (reading thematically-related articles). The results showed that RV was more efficient and effective for the learning and retention of vocabulary by senior high school students – a finding that corroborates Laufer's (Reference Laufer2003) results in Israel.
The second strand of reading skills development is concerned with reading strategies. Although reading strategies have been recognized in the L2 literature as an important way of increasing reading comprehension (Anderson Reference Anderson1991), there has until recently been little research on reading strategy instruction in Taiwan. Y. L. Chang (Reference Chang2005) investigated the strategies used in timed reading by college students majoring in accounting. The participants were introduced to various reading strategies at the beginning of a semester. Later, after reading three simulated intermediate GEPT passages, they were then asked to answer comprehension questions and recall what strategies they had used. In spite of little correlation between strategy use and task performance, ‘guessing meaning from the context’ and ‘rereading’ were the most frequently used strategies, while ‘reading aloud’ and ‘identifying sentence structure’ were the least. The author concluded that the students were unable to use reading strategies in an effective manner because of their low motivation in learning how to do so, a lack of lexical repertoire, and an over-dependence on dictionaries.
To compare reading strategies by students at different EFL proficiency levels, H. C. Huang, Chern & Lin (Reference Lin2006) designed an online reading program with 15 clickable strategy links which were categorized into four strategy types: global strategy, problem-solving strategy, support strategy, and socio-affective strategy. Using each reader's clicking behavior to record the use of strategies, the authors found that support strategies were used the most and problem-solving strategies the least. Additionally, high-proficiency students were more skillful in relying on such strategies. The authors also suggested that teachers should integrate reading-strategy instruction into classes, so that students will learn to use such strategies spontaneously. However, taking practical limitations into account (time constraints, unwillingness to make extra efforts, lack of required knowledge, and so on), most teachers may find it difficult to carry out the suggestions. To date this practice has rarely been adopted in class.
In Shang (Reference Shang2008), a total of 127 college students majoring in English participated in an English reading-strategy training program over one semester, in which they received instruction in four reading strategies: cognitive, meta-cognitive, compensation, and test strategies. The training effect was evaluated by means of pre- and post-reading comprehension tests, pre- and post-training questionnaires, and the students’ journal entries. The results demonstrated that reading-strategy training helped students in the use of such strategies and enabled them to better comprehend the texts. Similarly, using a think-aloud protocol, I. J. Chen & Huang (Reference Huang and Liou2007) found that high-intermediate proficiency learners used 17 strategies to comprehend different reading texts, and that high-proficiency EFL learners were more familiar with ways of enhancing their reading skills.
Taking a non-experimental approach, Sun, Liu & Hung (Reference Sun, Liu and Hung2009) investigated the effects of skimming, scanning, guessing, and careful reading strategies with the use of a survey. Students majoring in English at college were classified into four groups based on the four strategies that they claimed to use most often. The results showed no significant differences in the TOEIC reading scores between the four reading strategy groups. While this study may be informative, in practice no one uses a single strategy in the reading process.
The discussion above shows that studies on reading skills in Taiwan are broadly in line with those conducted elsewhere, such as Winstead (Reference Winstead2004). However, the target of such research has mostly been limited to English-major classes taught by the researchers at the college level; work using non-English majors or primary or secondary school students is comparatively rare. This focus may have widened the gaps between the recommendations of researchers and the actual behavior of practitioners.
4.2 Writing skills
General and academic English writing in Taiwan has begun to receive more attention in the last decade following the implementation of a number of related educational reforms (W. C. Chang et al. Reference Chang, Chang and Hsu2009). Some scholars began to analyze students’ written output, while others were interested in how the writing process or the integration of modern technologies into the process (the use of CALL) can empower student writers.
Inspired by process-based theory, F. C. Chang, Chang & Hsu (Reference Chang, Joe, Liaw, Yeh, Chern, You and Huang2009) investigated the effect of self- and peer-proofreading on the EFL writing produced by 121 second graders. The analysis was concentrated on variables measuring the students’ attitudes toward proofreading, awareness of writing mechanics and grammar, use of spelling strategies, and ability to revise and make corrections. Young writers made fewer mistakes when made aware of mistakes to be avoided. Similarly, motivated by the idea that peer feedback is more conducive to post-draft revision than teacher commentary (Hillocks Reference Hillocks1986), Min (Reference Min2007, Reference Min2009) compared students’ performance before and after peer reviews, and confirmed that peer feedback enabled college students to make greater progress in writing. In brief, this type of research favors comparing within-group performance before and after a peer-review activity. As Y. L. Cheung (Reference Cheung2011: 537) points out, ‘These studies failed to compare the effects of peer feedback and no peer feedback and did not give evidence to explain the usefulness of peer feedback’. Future research would benefit from the use of a control group to establish the effect of peer review.
Taking another approach, S. H. Lin & Liao (Reference Lin and Liao2009) assessed the effect of using interactive journal writing with junior high school students. The use of a dialogue journal was reported to yield several types of benefit: writing fluency, writer confidence, and interpersonal relationships. On the basis of the theory of holistic teaching and learning (Blanton Reference Blanton1992; Sauve Reference Sauve2002), H. C. Lin, Ho & Chien (Reference Lin, Ho and Chien2009) proposed a holistic approach to the teaching of writing in an advanced writing course at a national university. They argued that listening, speaking, and reading should be fostered simultaneously in a writing course in order to integrate students’ language expertise into their writing aptitude.
The approaches to writing instruction in the research projects mentioned above were all basically process-oriented, but imbued with product-oriented features and activities. While shown to be useful for writing skills development, they have yet to be widely put into practice at all school levels.
In contrast to research on the writing process of a single group, contrastive research on written products has shed light on how academic writing is approached. C. C. Yeh has been carrying out several between-group studies covering a range of topics related to writing, including a rhetorical analysis of hedging devices in academic writing. For instance, she revealed that graduate students used a greater number of hedging devices than expert writers (C. C. Yeh Reference Yeh2007). In another study, based on 18 student writing samples, she found a difference between experts’ and students’ citation behavior. Student writers displayed fewer citations, a reliance on long quotes, a lack of variety in citation forms and inaccurate citations (C. C. Yeh Reference Yeh2009). She therefore argued in favor of more explicit instruction in this area.
As argumentation is a fundamental writing style across a range of EAP and ESP writing tasks, F. W. Cheng & Chen (Reference Chen2009) used a genre-based approach to investigate how Taiwanese and American college students composed written arguments. Compared to the work by American students, Taiwanese students’ arguments were found to be both less extended and less complex, displaying a limited range and quantity of argumentative structures. However, the ‘time for task’ disadvantage of the Taiwanese participants (Zhao Reference Zhao2007) should have been taken into consideration for an appropriate comparison of the texts produced by native and non-native speakers.
CALL has been gradually integrated into the development of skills for process-based writing, as it can efficiently display related behaviors. M. J. Wang (Reference Wang2004) investigated students’ online learning processes by analyzing the time spent on online learning, learning styles, and peer feedback. In C. H. Chiu (Reference Chiu2009), students improved both their summary writing skills in content and reading comprehension after using the WriteToLearn web-based program, an online tutoring program designed for college-level writing classes.
In line with recent technological developments, Pai & Liou (Reference Liou, Chen, Yu and Chang2009) used blogs and wikis as platforms to understand how collaborative writing was carried out by students in different task environments. The participants were freshmen and junior students, the former working in pairs on blogs, while the latter paired up to write using the wiki platform. Each pair worked according to a similar pattern: after discussing the topic to be written about with the main author, the paired ‘co-author’ contributed a 70-word passage to co-construct a text. The main authors then determined how to incorporate the passage into their first drafts, and provided feedback for the co-authors to work on revisions, as needed. The results showed that the students favored the use of technology to facilitate collaborative writing.
Local research on CALL has stressed the importance of asynchronous online feedback, which helps students sharpen their L2 writing skills. Peer feedback has also been reported as beneficial to EFL writers in the literature (Tsai Reference Tsai2009). However, despite the usefulness and the convenience of automatic feedback, it was found to suffer from a number of drawbacks. For example, some programs (e.g. My Access!) assigned higher scores than human raters to students’ writing (H. J. Chen Reference Chen2009). H. H. Chen, Chiu & Liao (Reference Lin, Ho and Chien2009) also reported that some feedback was either misleading or failed to pinpoint students’ specific writing problems arising from modals, tenses, collocations, and word choice. This indicated that to increase the potential value of this technology, a more manageable CALL system is required.
4.3 Listening skills
According to Feyten's (Reference Feyten1991) criteria, of the four skills, listening provides the largest percentage (more than 45%) of our total communicative competence, followed by speaking (30%), reading (16%), and writing (9%). However, as we mentioned in our Introduction, listening has received less attention in the literature than the other three language skills (Buck Reference Buck2001; Vandergrift Reference Vandergrift2007). In Taiwan, investigations of listening skills have also had a low profile (H. Cheung Reference Cheung2006), reflected in the fact that English listening laboratory courses at some universities are taught by teaching assistants who are graduate students, rather than by lecturers (H. N. Chou Reference Chou2009). The lack of interest in listening skills may be due to the fact that English listening is not tested for school entry.
One line of research addresses the topic of listening difficulties. Y. G. Chao & Cheng (Reference Chao2004) explored the listening difficulties encountered by high-achieving and low-achieving high school students and collected data using the listening section of the intermediate GEPT and a listening-difficulty questionnaire. The results showed that listening difficulties were due to the students’ limited vocabulary, insufficient practice, and inability to follow rapid speech. In particular, low EFL achievers encountered more difficulties than high EFL achievers.
Following Vogely's (Reference Vogely1998) coding scheme for the sources of listening anxiety associated with audio/video viewing, Y. S. Cheng (Reference Cheng2005) examined the sources of, and changes in, 23 EFL college students’ listening anxiety over one semester. Additionally, the descriptive statistics of self-rated listening comprehension anxiety levels at the beginning, middle, and end of the semester were surveyed. One of the findings was comparable to the results in Vogely (Reference Vogely1998): video viewing activities provoked less anxiety than audio activities. Furthermore, listening anxiety was found to be negatively correlated with both course grades and self-perceived competence. To reduce listening anxiety, listening-strategy instruction should thus be taken into account.
By integrating CALL into listening skills development, Lo, Liou & Vongpumivitch (Reference Vongpumivitch2007) developed an online listening syllabus to enable students to undertake self-access practice. This system was constructed on a MOODLE platform integrated into a listening and speaking class, with an aim of helping future English teachers to design an online listening syllabus based on useful criteria with regard to instructional materials and online learning history data. However, to have more influence on the language learning of the Taiwanese this website would have had to be open to the general public.
Adopting an experimental approach, J. Y. Hsu & Hsu (Reference Hsu and Chen2007) investigated direct teaching of lexical collocation. A total of 34 college English majors were divided into two groups to receive three different types of instruction before completing a listening comprehension test which had been adapted from the TOEFL examination. The participants were divided into three groups, each receiving one of the treatments: single-item vocabulary instruction, lexical collocation instruction, and no instruction. The group receiving lexical collocation instruction scored the highest, while no significant difference was found between the other two groups. However, the research design was focused on the pre-listening phase of instruction and the product of listening, rather than the listening processes that help language learners improve overall as listeners.
Drawing on the listening styles profile proposed by Watson, Barker & Weaver (Reference Watson, Barker and Weaver1995), H. C. Teng (Reference Teng2007) investigated the relationship between listening styles and listening proficiency. The data was collected using three methods: (i) a listening-comprehension section of the intermediate GEPT, (ii) a questionnaire about four distinct listening styles identified by Watson et al. (Reference Watson, Barker and Weaver1995) – people-oriented, action-oriented, content-oriented, and time-oriented, and (iii) an interview. The author found that Taiwanese college students most often used people-oriented listening styles, implying that, in the process of listening, Taiwanese tend to find areas of common interest with others and respond empathetically to them. They also found a positive relationship between respondents’ use of people-oriented listening styles and their scores on the EFL listening test, and suggested that EFL listeners should be instructed to employ these styles more frequently to improve their comprehension.
More research into other aspects of listening skills development is needed, including the factors that influence listening outcomes, the strategies used by learners, technology- and network-based multimedia used for listening practice, and the pedagogical approaches that can be employed to help learners improve overall, as advocated by Field (2007). In addition, a more comprehensive research framework is yet to be formulated, in either a Taiwanese or more international context.
4.4 Speaking skills
Speaking has been identified by many scholars (e.g. Levelt, Roelofs & Meyer Reference Levelt, Roelofs and Meyer1999) as a highly demanding and complex cognitive skill that involves several different mechanisms. Despite its importance, speaking instruction has long been marginalized in Taiwan, and did not receive much attention until English education began to be implemented in elementary schools during the last decade. This is particularly true after the introduction of new TOEFL iBT test in recent years, which emphasizes listening, speaking, and the integration of different skills. Issues related to EFL learners’ oral proficiency have since been widely debated by Government officials, scholars, and the public. The growing perception of the importance of this skill has thus fostered more research on spoken English.
Correct pronunciation is deemed essential to successful communication. Y. L. Chen & Huang (Reference Chen and Huang2005) maintained that the prosodic features of pronunciation should be learned because mispronunciation harms communication. They therefore integrated computer-based pronunciation technology into their English class, so that students were exposed to both linguistic and communicative stimuli (e.g. segmental and supra-segmental features of English), and were given the opportunity to practice sound discrimination, sound/letter relationships, and contextualized sound contrasts. The results showed that students’ oral performance was improved and their attitude toward English pronunciation became more positive.
Instead of focusing on the microskill of pronunciation, S. J. Chang (Reference Chang2004) conducted a study from a broader perspective by investigating the effectiveness of a direct approach to English conversation instruction, with the aim of strengthening students’ conversational abilities. The direct approach views conversational competence as an end, focusing on the process, discourse, and strategies in casual conversation, and using games, paired (or group) discussion, and oral journals as the pedagogical activities. This approach was employed in an English conversation class with 35 students with English as their major or minor. The findings confirmed that the direct approach gave students an advantage in developing English conversational ability by raising students’ awareness of the discourse and microskills pertinent to conversation in English.
Y. F. Lin (Reference Lin2006) took a similar position towards speaking skills development in a small-scale study of 11 college students, making use of oral-written dialogue journals as writing pedagogy. Due to considerable academic pressure, large class sizes and the lack of a speaking test in the Joint College Entrance Examination (JCEE), the implementation of an oral-written dialogue journal as supplementary practice outside the classroom was considered a practical method of improving English speaking skills.
By using ‘tasks’ as the focus of a course design, Tseng (Reference Tseng2008) adopted a skill-integrated and metacognitive approach to investigate students’ speaking skills development. College students were chosen for the study; they chose topics and reading texts based on a selected textbook. They were then required to read the texts, listen to the teacher's lecture, lead the discussion in class, and write assignments. They were also required to record their speaking tasks performance for each topic. The students were then required to evaluate their peers’ and their own performance in terms of content and language use in the speaking performance. It was found that the integration of tasks related to all three skills (listening, reading, and writing) is effective in enhancing EFL learners’ overall oral proficiency. The findings of Tseng's study seem to validate the ‘task-dependency’ principle (Nunan Reference Nunan1999), which predicts that experience gained from receptive skills should be transferable to productive skills in pedagogical procedures.
As reported by Vongpumivitch (Reference Vongpumivitch2007a), local research on EFL speaking tasks concentrates mainly on the difficulty of speaking tasks, with related work mostly undertaken in a series of articles by R. W. Wu (Reference Wu2005, Reference Wu2007, Reference Wu2008). This suggests that more scholars should undertake speaking skills research from a variety of perspectives; for example, the effectiveness of speaking pedagogy needs to be extensively researched and properly assessed.
5. Assessment
Students’ language proficiency in Taiwan was found to be most frequently evaluated by ‘paper-and-pencil testing’ (H. N. Yeh Reference Yeh2000). Other types of assessment, such as listening or speaking, which supplement paper-and-pencil testing, have also become increasingly popular since the implementation of the communicative approach in the Nine-Year Joint Education curriculum, launched in 2001. There are currently three main types of assessment for language proficiency in Taiwan: English testing for college entrance examinations, English proficiency testing, and oral skills assessment.
English testing for college entrance examinations includes the Scholastic Aptitude English Test (SAET) and the Department Requirement English Test (DRET), both of which are administered once a year. The former is the basis for assessment of twelfth graders to see whether they are ready to go to college, the latter provides a second chance to those who are not satisfied with their SAET outcomes. The two tests were created as gatekeepers for admission to college. However, little research has been done on the validity, reliability, or potential washback of the tests (Fan & Chen Reference Fan and Chen2008; S. Y. Chou & Chen Reference Chou and Chen2009).
More research on established English tests, mainly the TOEIC, TOEFL, IELTS, and the GEPT, has recently been conducted in Taiwan. As one of the most widely administered English proficiency tests, the GEPT has gained much attention from local scholars (Fang Reference Fang2006), particularly with regard to its effects on students’ motivation to learn English. S. C. Teng et al. (Reference Teng, Kuo, Wang and Chang2006) investigated college teachers’ perceptions of the GEPT using a questionnaire, and found that the GEPT won favor because of its positive effect in encouraging EFL learners to study for the test. From the students’ perspective, Lu (Reference Lu2007) noted that students are being encouraged to meet the exit requirements by passing the GEPT. Generally speaking, the role of English proficiency tests in English education has been widely acknowledged.
However, the validity of the GEPT and its washback on English education is a matter of concern. Ma & Li (Reference Ma and Li2009) suggest a revision of the reading section of the high-intermediate GEPT in order to achieve satisfactory difficulty and discrimination indexes. Similarly, Shih & Wei (Reference Shih2008) highlighted the need to provide students with training in certain test items after they evaluated a specifically compiled GEPT question bank developed in-house at a university. Having analyzed the listening and reading comprehension sections of the advanced GEPT in 2007, R. W. Wu & Lin (Reference Wu2008) argued that measuring the construct of the GEPT was problematic. In their results, most test-takers’ English proficiency was found to be consistently below the benchmark set in the advanced GEPT. While the validity and reliability of the GEPT have been questioned, it has remained in use nationwide, as it was developed to meet the specific needs of people in Taiwan, students at various school levels, and adults in a range of professions.
Oral skills assessment has recently begun to receive more attention from scholars in Taiwan, and the range of research topics is quite varied. For example, C. S. Chang (Reference Chang2004, Reference Chang2005, Reference Chang2006) conducted a series of studies on pre-listening support: topical information, lexical support, and vocabulary preparation, respectively. Shang (Reference Shang2004) measured the effects of different response formats on listening comprehension with regard to a text's main ideas, its details, and the inferences that can be drawn from it. Y. G. Chao & Chien (Reference Chao2005) suggested adopting strategy-based instruction on the basis of a questionnaire presented to university students, asking them to recall the strategies they adopted or the difficulties they experienced when engaged on the listening section of a simulated TOEFL. Lower-proficiency learners, who encountered more difficulties in the listening process, were found to be in favor of translating English into Chinese. In contrast, higher-proficiency learners tended to rely on meta-cognitive strategies. A more comprehensive review of listening assessment was conducted by Vongpumivitch (Reference Vongpumivitch2007a). The author identified nine topics in listening-assessment research carried out between 2001 and 2006, ranging from the effects of pre-listening support and different types of listening tests on listening performance, through the impact of different response formats and modified listening input, an assessment of note-taking skills, the strategies used in listening tests and the influence of test anxiety on strategy use, diagnostic testing of listening problems, and portfolio assessment of listening ability, to teachers’ beliefs and attitudes toward listening tests. After examining previous work, Vongpumivitch called for more research on both existing and new topics, including the use of technology to assess listening and the use of photos, graphics, and videos to elicit a variety of linguistic output.
Studies on speaking assessment can be grouped under four headings: difficulty of assessing speaking, pausing patterns, self-assessment of speaking ability, and the use of technology to assess speaking (Vongpumivitch Reference Vongpumivitch2007b). Of these, the first has captured the most attention. For example, R. W. Wu (Reference Wu2005) used a picture description task as the central testing instrument to explore the relationship between familiarity with the pictures and speaking performance, and as might be expected, familiarity tended to produce better performance. Similarly, Li & Chen (Reference Chen and Huang2005), also using a picture description task, compared planned with unplanned performance in terms of fluency, complexity, accuracy, and errors in the process of picture description. Although there was no significant difference between planned and unplanned performance, high proficiency students spoke more fluently and demonstrated more complex and accurate output.
Apart from the studies just outlined, Taiwanese researchers have paid less attention to the assessment of speaking than to that of other language skills. We found some research reports, including those of Kuo & Chiang (Reference Kuo and Chiang2005), who took pausing patterns as an index of speaking ability, and C. Y. Chen (Reference Chen2006), who used self-reported results as a way of assessing speaking ability.
6. Conclusion
English education has recently been dubbed a ‘national movement’ in Taiwan, driving the country to become more competitive in the globalized world. This general, national interest in English has been reinforced by government policies such as the new English-in-education policy, Challenge 2008, the Aim for the Top University project, and the Teaching Excellence project. These policies have, in turn, contributed to the growing body of research into the diverse issues of English education which concern Taiwan. Although most of the studies are in line with the approaches illustrated in international, English-language publications in terms of their theories, topics, and methodologies, local research has certain unique characteristics. The main topics that have emerged in recent research, such as the evaluation of new policies, the revision of the curriculum, ways of implementing new techniques or activities, and an English exit policy, reveal not only current research trends in a Taiwanese context, but also the changing sociolinguistic profile in Taiwan, where English is now not just a subject at school, but also a skill for social and practical purposes. Taken as a whole, the work of local researchers responds to local needs and interests, but the implications of their classroom-based findings are yet to be widely applied to the Taiwanese context. Classroom practitioners are torn between the test-driven teaching styles that have been practiced since the beginning of English education in Taiwan, and a desire to adopt novel teaching approaches and/or use modern technology to develop students’ English proficiency for pragmatic purposes. In fact, most of the classroom-based findings in this review are the results of research on college students, particularly on students majoring in English, but there is more to be done at primary and secondary school levels as well as with adults, so that current insights can be applied to their classrooms. A more unified English-in-education policy might therefore be formulated in which the appropriate English proficiency levels for students at different stages of formal education can be successfully devised. This may contribute to the development of a more unified theoretical framework to explain and monitor changes in language behavior in Taiwan.
As might be expected, research in Taiwan varies greatly in quality. While some reaches an international standard for academic work in the field of English education, is published in high-quality national journals and contributes to the field, other work – presented in conference proceedings – is less readable and often contains little rationale and/or explanatory argument. English education is nevertheless a vibrant research field in Taiwan, but solid academic research published in conference proceedings can certainly improve the research network of English education in Taiwan.
In future, whether English education remains in the spotlight and continues to adequately represent foreign language education in Taiwan depends on the government's language-in-education policy. As stated by officials in the Ministry of Education, a good command of English is a must, but the ability to use a second foreign language in Taiwan is an asset for employment in multinational corporations (The China Post, 31 August 2011). In the next decade it is therefore likely that courses in foreign languages other than English will be more generally offered, and further research on the teaching and learning of those second foreign languages at different school levels will be beneficial.
Acknowledgments
The authors would like to thank three anonymous reviewers and Dr. Graeme Porte for their valuable feedback on the penultimate version of this review, but we are alone responsible for any omissions and errors that remain.
Suchiao Chen is Professor of the Department of Applied Foreign Languages at the National Taiwan University of Science and Technology, Taiwan. Her professional interests include TEFL, sociolinguistics, language planning, and intercultural communication. She has published books and research articles in international journals on an array of topics associated with sociolinguistics and TEFL. She currently serves as the editor of Taiwan International ESP Journal.
Yachin Tsai is Assistant Professor in the Department of Foreign Languages at the National Chiayi University, Taiwan and is the author of Asynchronous peer review and revision: On written argument (VDM, Saarbrücken, 2009). Her research focuses on L2 writing, listening, and educational technology.
