視覺小說融入低成就學習者之英語閱讀補救教學：行動研究 - 政大學術集成

全文

(1)國立政治大學英國語文學系碩士班碩士論文. 指導教授：黃怡萍博士 Advisor：Dr. Yi-Ping Huang. 立. 政治大. ‧ 國. 學 ‧. 視覺小說融入低成就學習者之英語閱讀補救教學：行動研究. y. Nat. Using visual novels in remedial English reading instruction for low-achieving EFL learners:. n. er. io. al. sit. An action research approach. Ch. engchi. i n U. v. 研究生：宋欣倫撰 Name：Hsin-Lun Sung 中華民國一〇八年七月 July, 2019. DOI:10.6814/NCCU201900597.

(2) USING VISUAL NOVELS IN REMEDIAL ENGLISH READING INSTRUCTION FOR LOW-ACHIEVING EFL LEARNERS: AN ACTION RESEARCH APPROACH. A Master Thesis Presented to Department of English,. 立. 政治大 National Chengchi University. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. i n U. v. In Partial Fulfillment of the Requirements for the Degree of Master of Arts. engchi. by Hsin-Lun, Sung July, 2019. DOI:10.6814/NCCU201900597.

(3) Table of Contents. Chinese Abstract. v. English Abstract. vi. Chapter One: Introduction. 1. Chapter Two: Literature Review. 3. 政治大 Game-based Language Learning 立 Remedial English Instruction. 3 5. Research Questions. 9. ‧ 國. 學. Chapter Three: Methodology. ‧. Context and Participants. er. io. sit. Intervention. Data Collection. al. n. Data Analysis. 11 12. y. Nat. Cycle Plan. 11. Ch. Chapter Four: Results. n U engchi. iv. 15 18 21 23. Week 1: Needs Analysis. 23. Cycle 1: Week 2~7. 28. Cycle 2: Week 8~13. 35. Chapter Five: Discussion. 41. Research Question 1. 41. Research Question 2. 44. Research Question 3. 45. iii. DOI:10.6814/NCCU201900597.

(4) Chapter Six: Conclusion. 47. Pedagogical Implications. 47. Limitation and Suggestions for Future Research. 48. References. 50. Appendices. 55. Appendix A: Needs Analysis Plan. 55. Appendix B: Survey for Needs Analysis. 57. Appendix C: Guiding Questions for Written Reflection. 62. 政治大. Appendix D: Semi-structured Interview Protocol. 立. 63. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. iv. DOI:10.6814/NCCU201900597.

(5) 國立政治大學英國語文學系碩士班碩士論文提要. 論文名稱：視覺小說融入低成就學習者之英語閱讀補救教學：行動研究指導教授：黃怡萍博士研究生：宋欣倫論文提要內容：. 立. 政治大. 視覺小說是一種互動式的數位遊戲，通常以包含文字、聲音、視覺藝術的. ‧ 國. 學. 多媒體形式來敘述一個故事。它的故事及視覺藝術吸引到玩家來閱讀。因此，視覺小說可以在英語閱讀補救教學中被選用，作為吸引低成就 EFL 學習者閱讀. ‧. 英語的方法。少有研究使用數位遊戲進行英語閱讀補救教學。本研究採用行動. y. Nat. sit. 研究的方式，來探索使用視覺小說進行英語閱讀補救教學，是否能增進學習者. n. al. er. io. 的閱讀能力，並提升英語閱讀動機。本行動研究有兩個循環，並從三位台灣國. i n U. v. 中低成就學習者身上獲取資料，包括錄音及影像紀錄、課堂觀察紀錄、學生的. Ch. engchi. 書面反饋、訪談。研究結果顯示(1)學習者對於閱讀英語視覺小說抱有正面的態度；(2)閱讀視覺小說的過程中，過多的未知單字是主要的閱讀困難；(3)學習者能夠藉由閱讀策略克服困難；(4)視覺小說能激起學習者的閱讀動機。本研究的最後提供英語補救教師及視覺小說開發者兩個要在英語閱讀補救教學中使用視覺小說的建議。. 關鍵詞：視覺小說、補救教學、數位遊戲式語言學習、低成就學習者. v. DOI:10.6814/NCCU201900597.

(6) Abstract A visual novel as an interactive digital game usually tells a story or a novel with multiple media, such as words, sound, and anime-style art. Its story and visual arts attract players to read. Therefore, visual novels can be adopted in remedial English reading instruction in order to engage and motivate those low-achieving EFL learners to read. Few studies have examined the use of digital games in remedial English. 政治大. reading instructions. To fill the gap, this study adopted an action research approach to. 立. explore using visual novel in remedial English reading instructions to enhance the. ‧ 國. 學. learners’ reading ability and raise their motivation to read in English. This action research study conducted two action cycles to collect data from three junior high low-. ‧. achieving EFL learners in Taiwan. Data included audio- and video-recording,. y. Nat. sit. classroom observation field notes, the students’ written reflection, and interviews. The. n. al. er. io. results showed that (1) the learners hold positive attitude towards reading English. i n U. v. visual novels, (2) having too many unknown words was their major reading difficulty. Ch. engchi. in reading the visual novel, (3) the learners were able to cope with the difficulty by using reading strategies, (4) the visual novel motivate the learners to read. At the end of this study, two suggestions are provided for English remedial teachers and visual novels developers on ways to give remedial reading instruction with visual novels.. Key words: Visual novel, remedial instruction, digital game-based language learning, low-achieving learner. v. DOI:10.6814/NCCU201900597.

(7) CHAPTER 1 INTRODUCTION. Remedial education aims to assist students in order to achieve expected competencies in core academic skills such as literacy and mathematics. It is designed for students who are not reaching their target levels, whether they have special needs. 政治大 academic skills regardless of what the cause may be. These learners demonstrate 立. or not. The students who enroll in remedial courses are mostly under-prepared in the. consistently low performance in languages but they may not have behavior problems. ‧ 國. 學. or an intellectual disability (Firkins, Forey, & Sengupta, 2007). Even students of high. ‧. intelligence can be under-prepared in a certain academic skill. In Taiwan, English as a. y. Nat. foreign language is often one of the three main academic subjects in remedial. er. io. sit. education for junior high school students. The English remedial course here aim to help students reach the basic level. One basic and important ability for low-achieving. al. n. v i n EFL learners to develop is toCread fluently and comprehend what is read. The ability hengchi U to read in English is important for these learners to get information or communicate. with others. Therefore, it is crucial to provide reading instruction in remedial courses. Prior research has shown that low-achieving EFL learners have difficulties in reading fluency and reading comprehension. These difficulties include having many unknown words, misidentifying contextual clues for prediction, and insufficient background knowledge (Yang, 2010), to name a few. As such, teachers need to take into consideration a wide variety of reading difficulties when designing remedial courses. In addition, low-proficiency learners generally have low motivation, and they tend to do the minimum required work (Thompson & Cochrane, 2012). They lack 1. DOI:10.6814/NCCU201900597.

(8) self-confidence and have relatively low self-expectation for their own performance, which may make them not willing to participate in class. Although previous research has adopted teaching methods different from those in regular courses for remedial instruction in English reading, there is oftentimes still a gap between students’ English proficiency levels in remedial courses and their counterparts’ levels in regular classes (Juan & Cheng, 2008). As the prior research has indicated that most low-achieving EFL learners have difficulties in improving their reading ability and learning motivation in remedial. 政治大 problem when I gave remedial courses. Therefore, I am looking for a way to motivate 立. courses for returning to their regular English classes, I have also encountered the same. low-achieving students and to enhance their reading abilities. Furthermore, prior. ‧ 國. 學. research has employed computer-assisted remedial reading intervention in L2 reading. ‧. instruction, for example online simulated-test English remedial courses (Sheu, 2011). y. Nat. and online reciprocal teaching and learning systems (Huang & Yang, 2015; Yang,. er. io. sit. 2010). However, the instruments in these studies tend to use online chatrooms or forums which allow the learners to discuss on the system when encountering reading. al. n. v i n Cgames difficulties. Few studies adopt digital reading instruction for lowU h e ningremedial i h c achieving EFL learners.. The purpose of this study is thus to enhance low-achieving EFL learners’ reading ability and to raise their motivation to read in English by using digital gamebased reading instruction through an action research approach. The results are expected to be useful for English remedial teachers and low-achieving EFL learners to have an alternative method of remedial reading instruction.. 2. DOI:10.6814/NCCU201900597.

(9) CHAPTER 2 LITERATURE REVIEW. In this chapter, previous research on two topics is reviewed: remedial English reading instruction and game-based language learning. The purpose of reviewing studies on remedial English reading instruction is to understand how language. 政治大 are, while previous research on game-based language learning is reviewed for 立. learners succeed in reading and what pedagogical implications from previous research. understanding how digital games can contribute to language learning.. ‧ 國. 學 ‧. Remedial English Reading Instruction. y. Nat. In order to design reading instruction for improving low-achieving EFL. er. io. sit. learners’ reading fluency and comprehension, there is a need to understand what fluent reading comprehension is. To provide a more accurate picture of reading, Grabe. al. n. v i n Cfluent and Stoller (2011) argues that comprehension is a set of processes U h e nreading i h gc rather than a single defined process. For example, fluent reading comprehension. involves rapid, efficient, strategic, purposeful, comprehending, and linguistic process to name a few (See Grabe & Stoller, 2011, p. 11~13, for detailed definition). That is, a successful reader is strategic enough to monitor and regulate one’s own reading comprehension so that they can cope with various reading problems with corresponding reading strategies. The development of various reading strategies has been found to be effective in improving students' reading comprehension (Baker & Brown, 1984). In addition, reading is a purposeful process because readers read a text in different ways according to their reading purposes, and the reading purposes will 3. DOI:10.6814/NCCU201900597.

(10) trigger their motivation to read. Learners will be more engaged in English reading if they are intrinsically motivated to read (McGeown, 2013). Prior research has adopted some teaching methods for remedial instruction in English reading. Many studies focus on enhancing the reading comprehension as a purposeful process and linguistic process. For example, Chou (2013) used Reader’s Theater as remedial instruction for 25 freshmen from a technical university in Taiwan, while other 24 freshmen received regular remedial instruction acted as the control group in the study. The results showed that the learners received Reader’s Theater as. 政治大 their control group counterparts, though they did not show significantly lower English 立 remedial instruction performed significantly better and had a higher motivation than. learning anxiety.. ‧ 國. 學. On the other hand, Juan and Cheng (2008) carried out a study using extensive. ‧. reading as a remedial program for three elementary school EFL students in Taiwan.. y. Nat. They aimed to explore whether extensive reading can improve the learners’. er. io. sit. performance and attitude towards English learning. The results showed that the three learners’ attitude changed positively that they became more engaged and confident in. al. n. v i n C hhad improvement inUEnglish letter and English learning. All the three learners engchi. vocabulary knowledge, and their reading speed and accuracy rate also increased. However, only one of them met the average level in their regular English class. For the other two learners, there was still a gap between their English proficiency levels and their counterparts’ in regular classes. Another study conducted by Tan, Pandian and Jaganathan (2016) focused on enhancing reading comprehension as a purposeful process. They provided graded readers, which are simplified for learners to comprehend without difficulty, as reading materials for 28 low proficiency undergraduates. The learners read for general comprehension and also increased their confidence through successful reading 4. DOI:10.6814/NCCU201900597.

(11) experiences on authentic and simplified texts. In so doing, the learners thought the graded readers were encouraging, and the results showed that suitable reading materials can be beneficial for learners to develop a more positive attitude towards English reading. Last, some other studies gave remedial reading instruction with assistance of computer and technology, for example online simulated-test English remedial courses (Sheu, 2011) and online reciprocal teaching and learning systems (Huang & Yang, 2015; Yang, 2010). These studies focused on enhancing reading comprehension as a. 政治大 comprehension and their use of reading strategies. The instruments in these studies 立 strategic process. The results showed the improvement of learners’ reading. required the learners to discuss on the system when encountering reading difficulties. ‧ 國. 學. so that they could observe and learn from the teachers’ or their peers’ use and. ‧. explanation of reading strategies. These studies showed that computer-assisted. y. Nat. language learning can assist low-achieving learners to cope with their learning. er. io. sit. difficulties. On top of that, some studies focused on game-based language learning in order to motivate the learners and help they learn the language. Many empirical. al. n. v i n C h instruction can motivate studies indicated that game-based and engage language engchi U. learners (e.g., Jackson, Dempsey, & McNamara, 2012). It is also found that gamebased learning could contributed to low-proficiency learners’ reading ability in the long term (e.g., Saine, Lerkkanen, Ahonen, Tolvanen, & Lyytinen, 2011).. Game-based Language Learning Scholars defined game-based language learning differently. Moreno-Ger, Burgos, Martínez-Ortiz, Sierra, and Fernández-Manjón (2008) defined game-based learning as “any initiative that mixes videogames and education” (p. 2531). The videogames here can range from an AAA commercial games, which are often 5. DOI:10.6814/NCCU201900597.

(12) developed and marketed with high budgets, to interactive multimedia wrappers, such as a flash video. Prensky (2001) highlighted the enjoyable elements of digital games and learning as he defined game-based learning as “precisely about fun and engagement, and the coming together of and serious learning and interactive entertainment into a newly emerging and highly exciting medium — Digital Learning Games” (p. 5). Reinhardt and Sykes (2012) further differentiate educational games from vernacular games. Educational games, also known as serious games, refer to those games which are specially designed for learning purposes, while the vernacular. 政治大 for purely commercial and entertainment purposes. Reinhardt and Sykes defined 立. games, called commercial off-the-shelf (COTS) games, refer to those games designed. game-based language learning as using educational games in language learning. ‧ 國. 學. processes. Studies on game-based language learning focused on how game designs. ‧. offer environments for language learning. On the other hand, Reinhardt and Sykes. y. Nat. gave a new term, game-enhanced language learning, to refer to language learning. er. io. sit. taking place in commercial off-the-shelf games. Studies on game-enhanced language learning concerned how language learning occurs ‘in the wild’ and how commercial. al. n. v i n C hpurposes. ThoughUthe scholars have different games can be converted for educational engchi definitions of game-based learning, there is something in common: the enjoyment is an important element in language learning processes that involves games. In this study, game-based language learning is defined as using any digital games for language learning while game-enhanced language learning specifically highlights the use of commercial off-the-shelf games. One of the main reasons for the turn towards adopting digital games in language learning is the need to re-engage under-motivated and under-achieving learners who have been turned away from learning in formal educational contexts (Thomas, 2012). Game-based language learning is beneficial for low-achieving 6. DOI:10.6814/NCCU201900597.

(13) learners because digital games may motivate these learners, create an informal atmosphere that enhances learner receptiveness (Richard-Amato, 1988), and supports collaborative learning. Wichadee and Pattanapichet (2018) conducted a quasiexperimental study on 77 undergraduates to examine what impact a digital game had on learners’ performance and motivation. The experimental group of 38 students was taught using the digital game Kahoot while the control group of 39 students was taught with the conventional method. They found that the experimental group obtained significantly higher scores and higher motivation than the control group.. 政治大 English learners examined the correlation between L2 English proficiency with the 立. Another empirical study conducted by Sylvén and Sundqvist (2012) on 86 young L2. frequency of gaming and types of games played. The results showed that frequent. ‧ 國. 學. gamers outperformed moderate gamers who, in turn, outperformed non-gamers. It. ‧. also suggested that there is a positive correlation between L2 gaming and incidental. y. Nat. and informal L2 learning. Also, Hoda, Henderson, Lee, Beh and Greenwood (2014). er. io. sit. conducted a multi-disciplinary study adopting an action research approach to examine how teachers and students interact with each other and with their developing digital. al. n. v i n C the game. The results showed that was engaging, promoted collaborative h egame ngchi U. gameplay and supported reciprocal teaching. As the prior research suggests, gamebased language learning can improve learner motivation, create an informal atmosphere, and supports collaborative learning, which is beneficial for lowachieving language learners. Among all the digital games, a visual novel is an interactive digital game and, to be more specific, a subgenre of adventure games. As the name suggests, it usually tells a story or a novel with multiple media, such as words, sound, and anime-style art. While the visual novels heavily rely on written language, the visuals and verbal language also play a crucial role in reading the visual novels (Green, 2017). 7. DOI:10.6814/NCCU201900597.

(14) Furthermore, visual novels often supply players with affective and graphic stimuli capable of engendering immersive experiences (Cavallaro, 2009). Oftentimes, players can make choices on what to do and/or what to say during gameplay. That is, a visual novel is a text-based gameplay which players creatively control decisions and plot turning points. The ending of a visual novel varies deepening on players’ choices so that it allows players replay the game several times. Prior research has examined the effectiveness of learners’ mastery on English expression and English conversation. Amalo, Agusalim, and Murdaningtyas (2017) conducted an experimental study with. 政治大 visual novels in enhancing students’ English-expressions mastery. They found that the 立 research and development method on 55 college students to measure effectiveness of. experimental group had significantly higher achievement in English-expressions than. ‧ 國. 學. their counterparts in the control group. Another research and development study. ‧. conducted by Agusalim (2015) on developing a visual novel game aimed to create. y. Nat. independent learning, increase motivation, and improve the ability and learning. game is attractive and useful for the learners.. al. er. io. sit. achievement for learners of distant learning. The results show that a visual novel. n. v i n Csuggest It is clear that previous studies novelsUcan be beneficial for h e nvisual h gc i. language learners. However, there are just a few studies on visual novels and language learning. Moreover, both of the studies reported that adopted game-based language learning to developing visual novels specifically for language learning purposes. Prior research has not paid attention to using commercial off-the-shelf visual novels to promote game-enhanced language learning for learners, not to mention for lowachieving learners. There is a need to explore the use of commercial off-the-shelf visual novels for low-achieving leaners. Therefore, the current study aims to adopt visual novels as digital game-enhanced reading instruction for junior high lowachieving EFL learners in Taiwan to enhance their reading ability and raise their 8. DOI:10.6814/NCCU201900597.

(15) motivation to read in English. Research Questions The following research questions will guide my study: 1. How do the learners in the remedial course in a junior high school in Taiwan interact with the reading tasks requiring them to play visual novel in the two cycles? 2. What kinds of reading difficulties, if any, do the learners encounter while playing the visual novels in each cycle? How do they cope with the. 政治大 How does playing the visual novel motivate the learners to read in each cycle? 立 difficulties?. 學 ‧. ‧ 國 io. sit. y. Nat. n. al. er. 3.. Ch. engchi. i n U. v. 9. DOI:10.6814/NCCU201900597.

(16) 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. DOI:10.6814/NCCU201900597.

(17) CHAPTER 3 METHODOLOGY. This study adopts an action research approach, defined as a spiral of steps, “each of which is composed of a circle of planning, action and fact-finding about the result of the action” (Lewin, 1948, p. 206). As the name suggests, action research. 政治大 enacts a planned intervention in response to a perceived problem while collecting and 立 involves both action and research. That is to say, the researcher in action research. analyzing data systematically to reflect on or evaluate the planned intervention for. ‧ 國. 學. further action. The purpose of action research studies is to “find out more about what. ‧. is going on in one’s own local context in order to change or improve current practice. y. Nat. in that situation” (Burns, 2009, p.115). Thus, since the current study aims to address. er. io. sit. and find a solution to low-achieving EFL learners’ reading difficulties, an action research approach will be used.. al. n. v i n Cdescribe The following sections methodology used in this action research. U h e nthe i h gc. The context and the participants of the study are specified first. Preceded by the background information, the next two sections explain the cycle plan and the intervention designed for the study. To reflect on and evaluate the planned. intervention, the last part of this chapter describes what and how data is collected and analyzed.. Context and Participants This study took place in a junior high school in a city in Northern Taiwan. The school offers remedial courses for the 7th and 8th graders on Chinese, math, and 11. DOI:10.6814/NCCU201900597.

(18) English. Those who have a percentile rank lower than 35 in one of the three subjects and further failed a test, based on the guidelines for remedial instruction by the Ministry of Education, can sign up for the after-school remedial classes. Each class takes one period, 45 minutes, per week and around 13 weeks in a semester. There are often three to five students in a class for remedial teachers to take care of each students’ needs more easily. About half of the remedial teachers are full-time teachers in the school, while the other half are part-time teachers who only teach remedial classes in the school. All the remedial teachers are given full control of course design. 政治大 The target course of the present study was an English remedial course for the 8 立. and encouraged to adopt innovative teaching methods.. th. graders in the fall semester, 2018. The researcher served as a part-time remedial. ‧ 國. 學. teacher in the class. Five students signed up for the class, and three of them. ‧. voluntarily participated in the study. According to the researcher’s observation, the. y. Nat. participants had nearly no opportunity and motivation to learn English after class. All. er. io. sit. of the three participants planned to go to vocational senior high school in the future so that they did not pay much attention to learning English for academic purposes but for. n. al. general or specific purposes.. Ch. engchi. i n U. v. Cycle Plan There were three stages in the current study, one planning stage and two cycles (see Figure 1). The planning stage aimed to conduct a needs analysis to identify specific problems and design the intervention (see Appendix A for the needs analysis plan). The participates filled in a self-report survey and took a diagnostic reading test on the first week of the course. The target reading strategies and syllabus were decided according to the results of the needs analysis. The two cycles started from the second week of the course. Each following 12. DOI:10.6814/NCCU201900597.

(19) cycle lasted for 6 weeks with the first five weeks to take action and the final week to review and reflect on the results of the action. There were some changes made in the second cycle according to the results of the first cycle.. Figure 1. Cycle Plan Stage 3: The 2nd Cycle (Weeks 8~13). Stage 1: Planning (Needs analysis). Stage 2: The 1st Cycle (Weeks 2~7). 學. ‧ 國. 立. 政治大. First Cycle: Weeks 2~7. The first cycle began on the second week of the course. The planned. ‧. intervention took place in the first five weeks of the cycle. From Week 2 to Week 4,. y. Nat. sit. the learners were given some pedagogical tasks that aimed to help them learn some. n. al. er. io. reading strategies. In the following two weeks, the learners were asked to do the real-. i n U. v. world task, playing the visual novel. During this period of time, the researcher. Ch. engchi. observed how the learners perform and interact with the given reading tasks. On the last week of this cycle, the researcher and the learners reflected on the outcomes of the planned intervention in the first cycle (for the timetable for the 1st cycle, see Table 1).. 13. DOI:10.6814/NCCU201900597.

(20) Table 1. Timetable for 1st cycle 1st Cycle. Week 2 3. Pedagogical tasks: Reading strategies. 4 5 Real-world tasks: Visual novel 6 7. Reflect on outcomes. Second Cycle: Weeks 8~14. 立. 政治大. ‧ 國. 學. The second cycle started from the eighth week of the course, and there were some changes made according to the results of the first cycle. Instead of having both. ‧. pedagogical tasks and real-world tasks, the second cycle only adopted the visual novel. sit. y. Nat. as reading materials for the learners to acquire reading skills. Similarly, the researcher. io. week (see Table 2 for the timetable for the 2nd cycle).. n. al. Table 2. Timetable for 2nd cycle Week. Ch. engchi. er. and the learners reflected on the outcomes of the planned intervention on the last. i n U. v. 2nd Cycle. 8 9 10. Reading the visual novel. 11 12 13. Reflect on outcomes. 14. DOI:10.6814/NCCU201900597.

(21) Intervention The intervention of this study was a task-based syllabus featuring visual novels gameplay. The task-based syllabus aimed to increase learners’ motivation to read and to give them a chance to do a possible real-world reading task. Nunan (2004) defined real-world tasks as “uses of language in the world beyond the classroom” (p. 1). In the current study, reading a visual novel was the real-world task which the learners may do for the purposes of entertainment in their life. Scholars have proposed many different types of tasks, and most of the. 政治大 one of the methods of classifying tasks is to categorize the strategies. In addition to 立. typologies are derived from the types of language use. Nunan (2004) pointed out that. typologies for developing oral language skills, there were also strategy-based. ‧ 國. 學. typologies for developing reading skills (e.g., Grellet, 1981). By combining strategies,. ‧. texts and reading purposes, the learners may improve their reading comprehension. y. Nat. ability and increase their reading speed (Lai, 1997). Therefore, the task-based syllabus. er. io. sit. was adopted for this study.. In the first cycle, to enable the learners to do real-world tasks, real-world tasks. al. n. v i n need to be transformed into C pedagogical tasks in the h e n g c h i Uclassroom first. That is, the. learners were given pedagogical tasks of reading short passages before reading the visual novel. The pedagogical tasks provided the learners opportunities to practice comprehending short passages and learn some reading strategies. The results of the first cycle showed that the teacher needed to model how to do the real-world tasks before the students did the tasks on their own in the second cycle. In the modelling stage, the teacher showed the students how to find the key words to comprehend the texts effectively, analyzing underlying intentions of the speakers based on the texts, and making predictions about what the interlocutor may say or do next. By so doing, the students understood how to read the visual novel and how to use the reading 15. DOI:10.6814/NCCU201900597.

(22) strategies. Another feature of the current study is to use a visual novel as the reading materials for real-world tasks. The visual novel adopted is Carpe Diem, released in 2015 (see Figure 2 & 3 for the game screenshots). Carpe Diem is a casual indie visual novel, which is made by an individual developer and has simple gameplay. It talks a story about two people going on a date. The students/players act as the male character and interact with the female character on the screen. The players can choose where to go on a date.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. Figure 2. The script is presented line by line.. 16. DOI:10.6814/NCCU201900597.

(23) 政治大 Figure 3. Players can make 立 their own decision on where to hang out.. ‧ 國. 學. There were three criteria for choosing the visual novel: learners’ proficiency. ‧. level, relevance, and estimated game time. To meet the learners’ proficiency level, the. sit. y. Nat. visual novel should use simple English which may not give the learners much trouble. al. er. io. to comprehend the texts. The target proficiency level was estimated by the guideline. v. n. for remedial instruction and adjusted by the teacher/researcher’s teaching experience. Ch. engchi. i n U. with low-achieving learners. Second, the topic of the visual novel would be better to be relevant to the learners’ life experience. By so doing, it is more likely to increase the learners’ motivation to read the visual novel. Last, because of the limited time in class, estimated game time of the visual novel could not be too long. The visual novel may take regular players 20 to 30 minutes for one round. It was expected that the learners were able to read the visual novel in 60 to 90 minutes, which may take two or three classes.. 17. DOI:10.6814/NCCU201900597.

(24) Data Collection To answer the three research questions, there were four sets of data collected: audio- and video-recording, classroom observation field notes, the students’ written reflection, and interviews.. Audio- and Video-recording The remedial classes were audio-recorded, while the learners’ gaming process of visual novels was videotaped. The visual recordings aimed to record the learners’. 政治大 say, the recordings documented how the learners interacted with the pedagogical tasks 立 interaction with the pedagogical tasks or visual novels, and with each other. That is to. and visual novels, what reading difficulties they had during the reading tasks, and. ‧ 國. 學. how they coped with the difficulties with other group members. Furthermore, the. ‧. recordings were used in the stimulated recall interviews as prompts to explore their. sit er. io. Observation Field Notes. y. Nat. concurrent thoughts during the gameplay.. al. n. v i n Observation field notes were C made by the teacher after h e n g c h i U each class in order to. record what happened in class. The teacher paid attention to how the learners. interacted with the reading tasks, what reading difficulties they had in the process of doing the tasks, and how they coped with the difficulties. Furthermore, weekly notes showed differences between how the learners interacted with the pedagogical tasks and the visual novels or between the interaction in each cycle, if there is any difference.. Students’ Written Reflection The learners were asked to write some reflection after playing visual novels as a 18. DOI:10.6814/NCCU201900597.

(25) learning activity. There were some guiding questions for the learners to write their reflection (Appendix C). The learners were asked to provide at least their observation, reflection, and insights into the intervention in their written reflection. The guiding questions were printed in English, and the teacher explained the meaning of each question in Chinese to make sure that the learners fully understand the questions. Though the learners were encouraged to write in English as much as possible, they can also choose to write in Chinese.. 政治大 There were several interviews with all the three participants to get access to 立. Interviews. their feelings, feedback, and opinions. Two types of interviews were used in this. ‧ 國. 學. study: stimulated recall group interviews and semi-structured individual interviews.. ‧. The interviews were conducted in Chinese and audio-recorded. The learners had. y. Nat. stimulated recall group interviews within a day after playing visual novels to explore. er. io. sit. their concurrent thoughts during the gameplay, while semi-structured individual interviews took place at the end of each cycle to explore their overall experience on. al. n. v i n the intervention. In order to C minimize the effect ofU h e n g c h i the power-relationship between the teachers and the learners, the learners were told that the purpose of the interview was to explore their thoughts of the intervention and to help them review what they had learned. There was no correct answer for the interview questions. The learners were interviewed in a group setting, which meant to make them relax and engage, to minimize the effect of the power-relationship between the teachers and the learners (Eder & Fingerson, 2001). The purpose of stimulated recall interviews was to understand the learners’ thoughts on the reading task of visual novels, to identify the specific reading difficulties they had in the gameplay, and to explore their metacognitive process when 19. DOI:10.6814/NCCU201900597.

(26) coping with the difficulties. The videotapes of the gaming process were replayed in the stimulated recall interviews so that the learners could share and discuss their concurrent thoughts during the reading tasks of visual novels. The following questions were asked in Chinese as prompts in the stimulated recall interviews: What were your thoughts of playing the visual novel? What were the difficulties that made you stuck here? What were you thinking when you had this difficulty? Why did you decide to adopt this strategy to cope with the difficulty?. 政治大 reflection on the experience and motivation to English reading with visual novels. The 立 On the other hand, the purpose of semi-structured was to explore the learners’. semi-structured interviews were able to access the learners’ deeper thoughts of the. ‧ 國. 學. interaction with the visual novels, their feelings of the gameplay and English reading,. ‧. and the feedback to the intervention (See Appendix D for the interview protocol).. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. 20. DOI:10.6814/NCCU201900597.

(27) Table 3. Timetable for data collection Week. Cycle. Syllabus. Data Collection. Ice-breaking, needs Problem identification 1. analysis, and a and plan intervention diagnostic test. 2. Audio & video-recording Pedagogical tasks:. Classroom observation field. 3 Reading strategy. 政治大 Real-world tasks:. observe outcomes. 立. 9. ‧ 國. Reflection & Stimulated recall. Reflect on outcomes. Semi-structured interviews. y. Classroom observation field. io. Reading the visual. n. al. observe outcomes 11. Midterm review. Audio & video-recording. Take action and 10. notes. Nat. 8. Classroom observation field. ‧. 7. Visual novel. Audio & video-recording. 學. 6. notes. sit. 5. Take action and. Ch. novel. engchi U. er. 4. v ni. notes Reflection & Stimulated recall. 12 13. Reflect on outcomes. Final review. Semi-structured interviews. Data Analysis After the data were collected, the four sets of data were analyzed systematically. The audio recordings and the interviews were transcribed, while the video-recordings were noted first. Then, the transcription was coded for the key topics related to the research questions so that initial themes were identified. The codes and themes were. 21. DOI:10.6814/NCCU201900597.

(28) further categorized into a few groups and were re-examined whether they are identified appropriately or not. Finally, the major categories of analysis were identified and developed. On the other side, the written data, observation field notes and the learners’ reflection, were noted first so that the themes were identified. Then, the themes were organized in order to summarize and reflect the learners’ behaviors and reactions. Finally, based on the themes, the patterns were gradually identified and developed so that the whole learning experience could be interpreted and clarified. To ensure the trustworthiness, triangulation strategy was adopted in the current. 政治大 triangulating the data sets, the interpretations of the data could be confirmed. For 立. study. The data were collected through multiple methods as mentioned above. By. example, the video-recordings ensured the observation field notes, and the. ‧ 國. 學. interpretations of the notes were confirmed by the interviews.. ‧. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. 22. DOI:10.6814/NCCU201900597.

(29) CHAPTER 4 RESULTS. The results of the two cycles are shown in this chapter. To address the participants’ reading difficulties and changes, the results are presented chronologically. At the beginning, the results of the needs analysis are provided. Then,. 政治大 gameplay of the visual novel are given. Last, the adjustments made for the classes in 立 the plans for the pedagogical tasks and the participants’ performance during the. the second cycle and the improvement made by the participants are shown.. ‧. ‧ 國. 學 Week 1: Needs Analysis. y. Nat. To understand the specific reading difficulties for the participants, the students. er. io. sit. were asked to fill in a survey for needs analysis and to read a short passage as a diagnostic test. The survey and diagnostic test included three major parts: reading. al. n. v i n C hreading, and uses U ability, attitudes toward English of reading strategies. engchi. According to the results of the survey, the participants self-reported that they. were not good at English reading with an average score of 2.13 out of a 4-point scale (1 means the participants strongly disagree with the item, while 4 stands for they strongly agree). Among the ten items in the survey (Table 4), the participants were better at reading signs (2.67), charts and graphs (2.67), making reasonable guesses (2.67). However, their performance in the diagnostic test showed that they could only identify the functions of signs, charts and graphs, such as timetable or map, instead of detail information on the signs, charts or graphs. Their major reading difficulty shown in the diagnostic test was their small vocabulary size. For example, they did not have 23. DOI:10.6814/NCCU201900597.

(30) sufficient vocabulary words to understand a sentence, not to mention the main idea of the passage.. Table 4. Self-reported English reading ability Item. Mean. 1.. I can recognize the words learned in class.. 1.67. 2.. I can read simple English signs.. 2.67. 3.. I can read simple charts and graphs.. 2.67. 4.. I can read simple daily conversation.. 政治大 I understand basic sentence patterns. 立. 1.67. 5.. 2.33. I understand main ideas of dialogues.. 2.33. 7.. I understand main ideas of short passages, messages, and letters.. 2.33. 8.. I understand main ideas and plots of stories.. 1.67. 9.. I can identify simple elements of stories, such as background,. ‧. ‧ 國. 學. 6.. y. Nat. 1.33. er. io. sit. characters, events, and endings. 10. I can make reasonable guess based on pictures and titles.. n. al. Ch. n U engchi. 2.67. iv. The results of the survey suggest that the participants have low motivation to read in English (Table 5). They also showed low motivation and slight resistance to read in English in the diagnostic test. The participants stared blankly at the paper during the test and waited for the end of the test. Even though the teacher asked them to try to read more, they would take a glance at the first sentence and immediately said they did not understand.. 24. DOI:10.6814/NCCU201900597.

(31) Table 5. Motivation to read Item. Mean. 11. Compared with other subjects, I am good at English reading.. 1.3. 12. I like the English reading classes in school.. 1.3. 13. I would like to take English reading courses.. 2. 14. I think reading in English is fun.. 2.67. 15. I do not want to read long and difficult articles.. 3.67. 16. Even though the contents are interesting, I do not want to read in English.. 立. 3. 政治大. ‧ 國. 學. Regarding use of reading strategies, even though they had reading difficulties in the diagnostic test, most participants still stared blankly at the paper without using any. ‧. reading strategies. Only one of the participants tried to guess the meaning of the. sit. y. Nat. passage; however, he simply made a wild guess rather than made any reasonable. io. er. guess from the text. Moreover, the needs analysis survey asked the learners to selfreport their familiarity of 15 reading strategies. Table 6 shows the learners’ degree of. al. n. v i n C hand table 7 showsUthe learners’ familiarity of familiarity of reading strategies, engchi. individual strategy. It is noted that while one represents that the learners have never heard the strategies, five means the learners know the strategies quite well and often use the strategies. The results of the survey included a total number of 45 strategies, that is, 15 strategies for each participant. However, the participants had only used three reading strategies (reaching the fifth level) and knew 20 strategies (reaching the third level and above) out of the total of 45 strategies (Table 6). That is, in general, the learners self-reported that they seldom used reading strategies and knew less than a half of the listed strategies. As for individual strategy (Table 7), only three strategies reached a mean score of three: “having a purpose in mind when I read” (3), 25. DOI:10.6814/NCCU201900597.

(32) “previewing the text to see what it is about before reading it” (3), and “getting back on track when getting sidetracked or distracted” (3.33). The rest got a mean score between two and three. That is, the participants had heard of the strategies, but they may simply know some of the strategies. The results of the survey show that the learners were not familiar with the reading strategies, not to mention used the reading strategies.. Table 6. Familiarity with reading strategies. 政治大 I have never heard of this strategy before. 立 Degree of familiarity. 1.. Freq. (N=45) 24.44% (11). I have heard of this strategy, but I don’t know what is means.. 31.11% (14). 3.. I have heard of this strategy, and I think I know what it means.. 8.89% (4). 4.. I know this strategy, and I can explain how and when to use it.. 28.89% (13). 5.. I know this strategy quite well, and I often use it when I read.. 6.67% (3). Nat. n. al. er. io. sit. y. ‧. ‧ 國. 學. 2.. Ch. engchi. i n U. v. 26. DOI:10.6814/NCCU201900597.

(33) Table 7. Individual strategy Strategy. Mean. a. Having a purpose in mind when I read.. 3.00. b. Taking notes while reading.. 2.67. c. Previewing the text to see what it is about before reading it.. 3.00. d. Reading aloud to help me understand what I’m reading.. 2.33. e. Checking to see if the content of the text fits my purpose for reading.. 2.67. f. Discussing what I read with others to check my understanding.. 2.67. 政治大 h. Underlining or circling important information in the text. 立. 3.33. i. Adjusting my reading pace or speed based on what I'm reading.. 2.67. j. Using reference materials such as dictionaries to support my reading.. 2.33. g. Getting back on track when getting sidetracked or distracted.. 2.67. ‧. ‧ 國. 學. k. Stopping from time to time to think about what I am reading.. 2.67. y. sit. io. al. er. information.. Nat. l. Using typographical aids like bold face and italics to pick out key 2.00 2.33. n. Re-reading to make sure I understand what I’m reading.. Ch. 2.33. o. Guessing the meaning of unknown words or phrases.. 2.67. n. m. Critically analyzing and evaluating the information read.. engchi U. v ni. To sum up, the results of needs analysis suggest that the participants had small vocabulary size, low motivation to read, and little use of reading strategies. These three major reading problems were identified and set as the teaching focuses for the first cycle.. 27. DOI:10.6814/NCCU201900597.

(34) Cycle 1: Weeks 2~7 The first cycle started from the second week of the English remedial course. The planned intervention took place in the first five weeks of the cycle, pedagogical tasks with reading articles from an English learning magazine in the first three weeks and gameplay of the visual novel in the following two weeks. In the final week of the cycle, the teacher and the participants reflected on the outcomes and experiences of the planned intervention in this cycle (Table 8).. Table 8. Timetable for the first cycle Syllabus. 立. Setting purposes of reading. Use reading strategies. Being aware of reading strategies. Pedagogical. Expand vocabulary size. Using reference materials. tasks: Reading. Pick the corresponding. y context.. 5 Real-world tasks. Discussing what is read with others to check their understanding. n. al. sit. io. 4. word meaning in the. er. Nat. strategies. ‧. 3. 學. 2. Teaching Focus. Have motivation to read. ‧ 國. Week. 政治大 Objectives. v. Cooperate with peers. i n U. Comprehend the texts. Reading visual novel. Ch. engchi. 6 Reflect on reading 7. Midterm review. Reviewing the learning experiences difficulties and solutions. Week 2 Having motivation to read is the first step for the students to learn reading in English. Therefore, the second class aimed to raise their motivation to read and make. 28. DOI:10.6814/NCCU201900597.

(35) them be aware of reading strategies. At the beginning of the class, the students were asked to reflect on their reading experience and how they faced reading difficulties. The students said that they would simply give up reading the texts and wait for the teacher to give them answers, which is the same as observed in the diagnostic test. To make students recall how they solve a problem, the teacher asked them what to do if they were lost. One of the students said she would ask somebody for direction. Another student said he would check the map on his smartphone. The teacher concluded that the students could also seek help and use reading strategies to cope. 政治大 In order to raise their motivation to read, the participants were asked to set 立. with their reading difficulties as they were lost.. purposes of reading as a reading strategy. One of the students identified his only. ‧ 國. 學. purpose of reading in English is answer comprehension questions. The teachers asked. ‧. the students what they want to know through reading an article about suspended. y. Nat. meals. The students gave many answers, such as what suspended meals were or what. er. io. sit. the origin of suspended meals was. However, it seemed that the reading purposes are looking for answers to the teacher’s question rather than having their own reading. al. n. v i n C hthe article about suspended purposes. After they were given meals, the students only engchi U read for the comprehension questions. They did not put any effort into reading the. paragraph about the origin of suspended meals, even though knowing the origin of suspended meals was one of their reading purposes. Therefore, it was concluded that the students tended to take meeting the teacher’s requirement to read or reading for comprehension questions as their reading purposes in the context of classroom. The participants as low-achieving learners only did the minimum required work to complete the class and showed low motivation to read for themselves.. 29. DOI:10.6814/NCCU201900597.

(36) Week 3 The results of needs analysis had identified a lack of vocabulary words as another major difficulty for the students to comprehend English passages. To help the students comprehend a text with their small vocabulary size, the students were asked to use reference materials to get the meaning of unknown words instead of staring blankly at given texts. All the students chose to type the whole text in Google Translation and hand in the translated text as their answers. However, not all the sentences were translated correctly. For example, Google Translation translated ‘More. 政治大 the air’, and the students could not understand or explain what ‘suspended 立. people buy suspended sandwiches’ as ‘More people buy sandwiches which hang in. the students fully understand the text correctly but text either.. 學. ‧ 國. sandwiches’ really means. This showed that the whole text translation could not help. ‧. To enable the students to comprehend reading passages correctly, the teacher. y. Nat. introduced Cambridge Online Dictionary to the students. Cambridge Online. er. io. sit. Dictionary provides English-Chinese translation for the students to comprehend words with the aids of their L1. With a dictionary, the students were expected to. al. n. v i n understand meanings of every wordC and further comprehend h e n g c h i U a sentence. After a brief introduction to the online dictionary, the students were asked to look up some words as practice in order to get familiar with the user interface. The students were able to use the online dictionary in a short time and comprehend the given text better than comprehend the text through the whole-text translation software. However, they had difficulty in understanding polysemous words. To be more specific, the students were unable to pick the corresponding word meanings in the context. For example, they had a hard time understanding the meaning of ‘change’ in the sentence ‘I need NT$40. Do you have any change?’ The word ‘change’ in the given sentence means ‘money in the form of coins.’ However, 30. DOI:10.6814/NCCU201900597.

(37) the students could only pick the first meaning of ‘change’ as a noun in the dictionary, ‘the act of becoming different, or the result of something becoming different.’ Though the students know how to use dictionaries, they were unable to pick the best word meaning for the given context.. Week 4 The results of the previous class showed that the students would only pick the first meaning in the dictionary for unknown words. In order to help the students to. 政治大 discuss what was read and check their understanding with each other. They were 立. pick the corresponding word meanings in the context, the students were paired up to. specifically asked to question the partner’s interpretation first and discuss what the. ‧ 國. 學. correct meaning may be.. ‧. Although the students were paired up for discussion, everyone was as quiet as a. y. Nat. mouse. They only tried to comprehend the text on their own rather than discussed. er. io. sit. with their peers. The teacher reminded them to discuss with their peers. Some students heard that and gave a smile back, but no one started to talk with others. The. al. n. v i n C hdid not discuss withUpeers. One student suggested that teacher further asked why they engchi they did not know what to question and how to discuss with their peers, while others. nodded in agreement. Therefore, the teacher interfered in the discussion. The teacher questioned the students’ first interpretation of the text and guide them to discuss what may be misinterpreted and which word meanings may be more suitable. For example, according to the students’ prior knowledge, ‘poor’ means ‘bad’ as in ‘I am poor at math.’ Therefore, they comprehended the sentences, “A very poor man comes in and says…” as ‘a very bad man’. To correct their misinterpretation, the teacher raised a question.. 31. DOI:10.6814/NCCU201900597.

(38) T: A bad man? Are you willing to buy a cup of coffee for a bad guy? S1: Of course not. T: Then, why do people buy suspended meals for a bad man? S2: So the word ‘poor’ does not mean ‘bad’. Let’s check the dictionary. S3: In the dictionary, ‘poor’ can also means ‘having little money’.. Week 5 The fifth and sixth class were scheduled to play the visual novel. One visual. 政治大 together. Different from reading English passages in the previous classes, the students 立 novel is introduced to the students, and they were asked to read the visual novel. showed some interests in playing the visual novel game; they leaned forward and paid. ‧ 國. 學. full attention to the teacher’s instruction on the visual novel. One of the students even. y. Nat. other peers’ curiosity toward the visual novel game.. ‧. shared his experiences of reading visual novels in Chinese, which further aroused. er. io. sit. After a brief introduction to the gameplay, the students were given full control of reading and playing the visual novel. They soon found there was only one heroine. al. n. v i n on the screen and they would act asC the hero to have a dateUwith that girl. They further hengchi guessed that the heroine may be simply a software from her name Ai, which was. believed to be the abbreviation of ‘artificial intelligence’ by the students. Though the students were exploring the multiple media of the visual novel excitedly, they had a huge difficulty in reading the English script. They spent much time on looking up every single words in dictionary. They could only read three sentences in 15 minutes. All the students expressed that they prefer playing visual novels to reading articles. For example, one student said in the first stimulated recalled interview, “playing the visual novel had much more fun than reading the passage from the magazine” (IN1-1101018). However, all the students also identified unknown words 32. DOI:10.6814/NCCU201900597.

(39) and English grammar as the reading difficulties they had in reading the visual novel. One student said, “vocabulary words and grammar are the major problems which keep us away from comprehending the script” (IN1-1101018), while others nodded in agreement enthusiastically. Compared with reading short passages, they did look up more words while playing visual novel. Too many unknown words may be one of the reasons. There may be another reason that the students felt insecure to move on to the next sentence without fully understanding the current sentence. One student specifically pointed out that “I want to make sure I understand every single word in. 政治大. case I could not comprehend the following plots” (IN1-1101018).. 立. Week 6. ‧ 國. 學. During this week, the students were limited to looking up three words for a. ‧. sentence. By so doing, it was expected that the students could read more sentences in. y. Nat. the class and learned to tolerate ambiguity in reading. However, it still took much. er. io. sit. time for them to figure out every word’s meanings. Instead of spending time looking up the words, the students tried hard to recall where they may have learned about the. al. n. v i n C hmeanings with eachUother. For example, one student words and then discussed word engchi. said that she had seen the word ‘hour’ before, and another student added that the word was taught in one unit of their textbook. Then, the three students tried hard to recall what the word means. Hence, they still spent much time on comprehending every word in the text by triggering their memory and so they read slowly for only four sentences in 20 minutes. At the end of the class, the students showed lower motivation to read the script of the visual novel. The students found it tiresome looking up unknown words and recalling from their memory. They complained about the large number of unknown words and felt too tiring to getting the meaning of every word. Stopping 33. DOI:10.6814/NCCU201900597.

(40) comprehending the text, the students started to discuss the visual arts. One criticized the drawing techniques of the author, while another one complained that the author should have provided more vivid visual arts. Although the students showed low motivation to read the script and play the game, they were still curious about the upcoming plots. They choose to quickly skim the whole visual novel through the visual arts in order to guess what the following plots may be. The students identified that unknown words are the major problem of comprehending the visual novel in the stimulated recall interview again. “Those unknown words make me feel unwilling to. 政治大. read the visual novel,” said one of the students (IN2-1108018).. 立. Week 7. ‧ 國. 學. In the last class of the first cycle, the teacher and the students reviewed what. ‧. they have learned in the previous classes and evaluated their learning experience so. y. Nat. far. A semi-structured interview was also conducted in order to understand the. er. io. sit. students’ thoughts of the interaction with the visual novels, their feelings of the gameplay and English reading, and the feedback to the intervention.. al. n. v i n C hall the students agreed In the semi-structured interview, that they are more engchi U. willing to read the visual novel than articles from the textbook or English learning magazines. One student said, “I like to read the visual novel. It is the most interesting English reading experience I have ever had” (IN3-1115018). However, two of the students criticized the visual arts again, while the other one defended that the visual novel as an indie should not be criticized so hard. Anyway, they thought the visual arts were not attractive enough and did not give them hints about the script or the plots. The students also identified vocabulary words as the major reading difficulty for them to play the visual novel. Two of the students said, “All I did in reading the 34. DOI:10.6814/NCCU201900597.

(41) visual novel is to look up the vocabulary words” (IN3-1115018). For the students, English reading simply means knowing the word meanings. In addition, there were too many unknown words for the students, which lowered their motivation to read the visual novel. One further expressed, “There were too many words I need to look up in the dictionary. It was too tiring to continue reading the scripts” (IN3-1115018). However, the students wanted to read the visual novel for knowing the following plots. They hoped to finish a whole storyline before the end of the semester.. Cycle 2: Weeks 8~13 政治大 The second cycle of action started from the eighth week to the end of the 立. English remedial course. After collecting data from the first cycle, the teacher decided. ‧ 國. 學. to adopt a way to teach English reading with the visual novel. First, the students only. ‧. read a few lines of the visual novel and had interests in reading the visual novel, so. y. Nat. the teacher continued adopting the visual novel as the reading materials in the second. er. io. sit. cycle. Second, the students showed that vocabulary words is a critical issue for them to comprehend the text, and they did not have much successful reading experience.. al. n. v i n Cprovide Therefore, the teacher should facilitation and scaffolding in their U h e nmore i h gc. reading process. In the second cycle, the teacher first modelled how to read the visual novel, and then the students were given some controlled exercises. After the students became familiar with how to read the visual novel, they would be given a task with scaffolding removed. There are two tasks in the second cycle. That is, the students would repeat the reading process twice. In the final week of the cycle, the teacher and the students reflected on the outcomes and experiences of the planned intervention in this cycle (See Table 9).. 35. DOI:10.6814/NCCU201900597.

(42) Table 9. Timetable for the second cycle Week. Task. Teaching Focus. 8. a. Identify key words Getting the main idea b. Tolerate ambiguity. 9 10. a. Prediction 11. Unscrambling the scripts b. Reread & re-tell. 12 13. Review. Reflect on the learning experiences. 立. Weeks 8~9. 政治大. ‧ 國. 學. The eighth and ninth classes aimed to help the students read fluently, specifically increasing their reading speed. According to the results of the first cycle,. ‧. having too many unknown words in the script prevented the students from reading. sit. y. Nat. fluently. The students wanted to get full understanding of every word before they. io. er. moved on to the next line. Therefore, the teacher decided to model how to get the. al. main idea of a passage. The students were asked to identify the key words in a. n. v i n sentence, who did what. That is, theC students to identify what the subject and the h e nneed gchi U action are in a sentence. Fortunately, the students knew the basic structure of an. English sentence and parts of speech. Hence, they were able to get the key words with some exercises. Moreover, the students were only allowed to look up the key words. They could only make a guess or leave the rest of unknown words there. By so doing, the students would learn to tolerate ambiguity in reading and keep the reading flow. Because the teacher led the reading process at the very beginning, the students seemed to be more relaxed in the classroom. They had more smile on their faces and were more willing to interact with the teacher. In the modelling process, the students. 36. DOI:10.6814/NCCU201900597.

(43) felt safe to skip unknown words and tolerate ambiguity. As one of the students said, “I did not feel anxious about those unknown words. [The teacher] would tell us the meaning, if it is necessary, anyway” (IN4-1206018). In the second stage of doing controlled exercises, the students practiced what the key words may be in a sentence and tolerated the ambiguity of unknown words. With the teacher’s coaching, the students were not bothered by the unknown words and had some successful reading experiences. They became more confident in their reading comprehension and gradually speed up the reading process. In the final task, the students were given 12. 政治大 whether their comprehension is correct with the teacher from time to time, the 立. lines of reading and asked to get the main idea. Although they wanted to check. students were able to get the main idea and finished 12 lines in 10 minutes.. ‧ 國. 學 ‧. Weeks 10~12. y. Nat. The following three classes aimed to cultivate the students’ ability to predict the. er. io. sit. development of the script. Predicting not only prepared the students to read the following lines but allowed the students to become actively involved in the reading. al. n. v i n C h the students would process. While making predictions, activate their prior knowledge engchi U and connect what they have read or seen in the visual novel to their own personal. experiences. Furthermore, they would need to reread some of the script in order to check whether their predictions were correct. The teacher first modeled how to make a prediction based on prior knowledge and then guided the students to share how they made a prediction. Then, the whole class read the next line to check which prediction was the closest to the original script. The students viewed making the closest prediction as a game and competed with their peers. In the stage of doing controlled exercises, the students first read a few lines and then chose the most possible next line from three or four choices. It seems that the 37. DOI:10.6814/NCCU201900597.

(44) exercises were quite easy for them to get most of the exercises correct. For the final task, the students were asked to unscramble 24 lines into the correct order. Although it took them 40 minutes to complete the task, the students did not feel the task too difficult. They spent much time on reading and rereading the lines and trying to match the lines.. Week 13 As the last class of the second cycle, the teacher and the students reviewed what. 政治大 The second semi-structured interview was also conducted in order to understand the 立. they had learned in the previous five classes and evaluated their learning experience.. students’ thoughts of the interaction with the visual novels, their feelings of the. ‧ 國. 學. gameplay and English reading, and the feedback to the intervention.. ‧. After the five classes in the second cycle of the course, the students still. y. Nat. expressed that they would like to read visual novel rather than articles from English. er. io. sit. learning magazines. As one student said, “It is fun reading the visual novel. I would like to read another visual novel in English class next semester” (IN4-0110019).. al. n. v i n C hand tiring to read the Another one added, “I felt less stressed visual novel than to read engchi U the articles” (IN4-0110019). The students also pointed out that sharing their thoughts with each other while reading visual novel is interesting. They may not have that much opportunity to interact and get to know each other if they simply read and comprehend articles. Compared with having full control to read the visual novel in the first cycle, the students would like to be guided to read the visual novel. As one student said, “I like the way [the teacher] guided us to read the visual novel so that I could comprehend the visual novel more” (IN4-0110019). Though these low-achieving students still needed much help or support from the teacher to comprehend the visual novel, they 38. DOI:10.6814/NCCU201900597.

(45) would like to read some parts on their own. “I still want to read the visual novel by ourselves if we were able to read since it is a game for me to play,” said one of the student (IN4-0110019). Having too many unknown words was still the main reading difficulty for the student to comprehend the visual novel, though they could use dictionary to solve part of the problem. The students specified that they did not know which meaning of the unknown words fit in the context. Moreover, they could not read fluently because of the acts of looking up dictionary.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. 39. DOI:10.6814/NCCU201900597.

(46) 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. DOI:10.6814/NCCU201900597.

(47) CHAPTER 5 DISCUSSION. The previous chapter presents the students’ experience of reading the visual novel in the English remedial course. Since the current study aims to gain insights into how visual novels as digital game-based reading instruction could enhance low-. 政治大 this chapter discusses the three research questions. The first research question focuses 立 achieving EFL learners’ reading ability and raise their motivation to read in English,. on how the learners interact with the reading task of the visual novel, while the second. ‧ 國. 學. research question reveals the reading difficulties the learners encountered while. ‧. playing the visual novel and how they coped with the difficulties. Finally, the third. y. sit. io. er. visual novel.. Nat. research question explore the learners’ motivation to read in English after playing the. al. n. v i n RQ1: How do the learnersC in the remedial courseUin a junior high school in Taiwan hengchi. interact with the reading tasks requiring them to play visual novel in the two cycles? There are four types of interaction the students had during the reading task of the visual novel: interaction with visual arts, with the texts, with peers, and with the teacher.. Interaction with visual arts After given the visual novel in the first cycle, the students relied much on the visual arts to comprehend the visual novel rather than reading the words directly. The students expected that the visual arts provide hints of the context of the visual novel 41. DOI:10.6814/NCCU201900597.

(48) so that they can comprehend the visual novel. The expectations of the students were consistent with O’Neil’s (2011) idea that the visual arts should help to establish or enhance the text of visual novels. However, the visual arts in this indie were not abundant to attract the learners and guide the comprehension. There were often few pictures with minor changes in an episode. The chosen visual novel did not meet Cavallaro’s (2009) argument that visual arts are attractive to players; hence, the participants were not motivated to read by the visual arts. In the second cycle, the learners no longer paid much attention to the visual. 政治大 attention to the visual arts when there was a specific purpose. For example, they 立. arts. In fact, the learners ignored the visual arts most of the time. They only paid. interpreted the facial expressions as hints for the unscrambling task. The learners. ‧ 國. 學. adopted a role of interpreter during the act of reading, which echoes with Serafini’s. ‧. (2012) four roles of reader-viewer. Although the visual arts in this case did not play an. y. Nat. important role in the learners’ reading comprehension, they still served some acts of. n. er. io. al. sit. reading during the reading process.. Interaction with the texts. Ch. engchi. i n U. v. When the learners had full control of reading the visual novel in the first cycle, the learners adopted bottom-up model to read the visual novel. Bottom-up model is known as getting pieces of translation and information in the texts (Grabe & Stoller, 2011). While the learners focused on understanding the linguistic units but neglected other types of schema, which may weaken the significance of reading comprehension as Li, Wu, and Wang (2007) argued. Besides, the learners’ linguistic knowledge was not sufficient enough to adopt bottom-up model for reading comprehension. In the second cycle, the learners were guided to adopt top-down model, which focus on text understanding. They activated their background knowledge to set some expectations 42. DOI:10.6814/NCCU201900597.

(49) or predictions about text information and seek for enough information from the text to check their expectations and predictions. The findings show that the learners relied on both bottom-up model to process the words and top-down model to activate background knowledge so that they can comprehend the visual novel. Such results correspond schema theory that reading comprehension is an interactive process (Adams & Collins, 1979; Rumelhart, 1980).. Interaction with peers. 政治大 their peers during the reading task of visual novel. While the learners tended to read 立 In addition to interaction with the visual novel, the students also interact with. the given articles from English learning magazine individually, they read the visual. ‧ 國. 學. novel together in the first cycle. It may be the gaming factor of visual novel that. ‧. differentiate reading visual novel from reading articles. Playing visual novel creates. y. Nat. an informal learning atmosphere in the classroom, which makes the students ‘play’. er. io. sit. together, while reading articles is more like taking a reading comprehension test that the students should read individually. The finding that visual novels created an. al. n. v i n Cishconsistent with Richard-Amato’s informal learning atmosphere (1988) expectation engchi U of using digital games in language learning.. Although the students read the visual novel together, they in fact simply shared the workload of looking up unknown words in dictionary. It was not until the teacher guided the reading process in the second cycle that the students actively interacted with each other. They gradually formed an “affinity group” as Gee (2007) defined. The act of sharing understandings of the texts, past experiences, background knowledge, and reading strategies with each other is similar to Hoda, Henderson, Lee, Beh, and Greenwood’s (2014) and Yang’s (2010) findings that an affinity group would do. By so doing, the learners can comprehend the visual novel better. 43. DOI:10.6814/NCCU201900597.

(50) Moreover, through interacting with their peers, the students had the purpose of reading to discuss the contents and the plots with others.. With the teacher In the first cycle, the students were given full control of reading the visual novel, and the teacher played the role of an input provider and a monitor in the reading process. The teacher’s role met Larsen-Freeman and Anderson’s (2011) description of teachers’ role in task-based language learning. However, the results of. 政治大 own. To facilitate the learners reading visual novels in the second cycle, the teacher 立. the first cycle showed that the learners had difficulties in completing the task on their. played a more active role in the reading process. The teacher acted as a supporter in. ‧ 國. 學. the task and engaged in the team. By so doing, the learners felt supported and secured. Nat. y. ‧. to try various methods to comprehend the texts.. er. io. sit. RQ2: What kinds of reading difficulties, if any, do the learners encounter while playing the visual novels in each cycle? How do they cope with the difficulties?. al. n. v i n C hEnglish learning magazines, When they read the articles from the lowengchi U. achieving students had some of the reading difficulties, including many unknown words, lacking reading strategies, and low motivation to read, which have been discussed and investigated extensively by many scholars in literature (e.g., Guthrie, McRae, Coddington, Lutz Klauda, Wigfield, & Barbosa, 2009; Yang, 2010). The visual novel provided a gaming setting for the learners which naturally motivated the learner to read as Agusalim (2015) found visual novel could attract learners to read. Also, the learners knew more reading strategies as they were taught and used the strategies more often than before. However, the small vocabulary size was still the major difficulty for these low-achieving students to read the visual novel. While it 44. DOI:10.6814/NCCU201900597.