Reading in a Foreign Language April 2013, Volume 25, No. 1 ISSN 1539-0578 pp. 72–93 A model of motivation for extensive reading in Japanese as a foreign language Ryoko de Burgh-Hirabe & Anne Feryok University of Otago New Zealand Abstract Numerous studies have reported that extensive reading (ER) has a positive influence on affect. Recent studies suggest that motivation for ER changes. This is in line with recent developments in second language (L2) motivation research that have highlighted the complex and dynamic nature of L2 motivation. This study presents a model of complex and dynamic motivation for ER. This qualitative study examined 9 Japanese as a foreign language learners’ motivation for ER. The participants were encouraged to read as many Japanese books as possible outside class for 5 to 7 months. Data from interviews and journal entries were analyzed for factors influencing their motivation. The participants’ motivation changed as different factors interacted, leading to different patterns of engagement with ER, which fit within the model. This suggests the value of using a complex and dynamic approach to L2 extensive reading motivation. Implications concern the importance of varied materials and of making ER obligatory. Keywords: Japanese as a foreign language, extensive reading, motivation, self-regulation, high school learners Since the Fijian book flood study (Elley & Mangubhai, 1981), numerous quantitative studies have offered compelling evidence of the benefits of extensive erading (ER). However, despite studies reporting that ER promotes positive attitudes to second language (L2) reading (Waring, 2001), there has been a “lack of systematic and principled attention to the affective dimenisons of second language reading” (Day & Bamford, 1998, p. 21). Only a handful of recent studies (e.g., Judge, 2011; Nishino, 2007; Takase, 2007) have addressed this. Few have been longitudinal or qualitative, although arguably this may be the best approach to ER (Arnold, 2009; Mohd Asraf & Ahmad, 2003). The role of motivation in L2 learning has also been considered through quantitative studies, from the socio-educational model (Gardner, 1985) to the process model (Dörnyei & Ottó, 1998), including the most recent modelling using a complex systems approach (Dörnyei & Tseng, 2009). Although ER studies have mentioned the complex and dynamic nature of motivation for ER (e.g., Judge, 2011; Nishino, 2007), none have explicitly used this approach . This study aims to address these lacunae in a qualitative case study of adolescent Japanese as a Reading in a Foreign Language 25(1) de Burgh-Hirabe & Feryok: A model of motivation for extensive reading in Japanese 73 foreign language (JFL) learners’ motivation for ER over two to three terms in New Zealand high schools. Literature Review Recent Conceptualizations of L2 Motivation The process model of L2 motivation (Dörnyei, 2000, 2001; Dörnyei & Ottó, 1998 a)ttempted to reconceptualize motivation as “dynamically changing cumulative arousal in a person” (Dörnyei, 2009b, p. 196). The process model draws upon action control theory (c.f., Heckhausen, 1991), which divides the motivational process into discrete temporal segments. In the preactional stage, choice motivation describes how initial wishes and desires are first transformed into goals and then into intentions. In the actional stage, actualizing executive motivation leads to accomplishing the goal. In the postactional stage, action is retrospectively analyzed. The other important dimension of this process model is motivational influences. Dörnyei and Ottó (1998) claimed that the motivational influences differ in each phase, so in principle motivational influences at one phase do not have an effect on other phases, which is consistent with action control theory which stresses the differences between choice motivation and executive motivation (Dörnyei, 2001). Dörnyei (2005) identified two issues with thep rocess model: identifying the boundaries of actional processes and recognizing potential interference from other actional processes. As he points out, school learners are typically engaged in embedded and parallel learning activities involving both social and cognitive aspects of motivation. Motivation for learning through a particular activity may be influenced by other activities; motivation for learning a particular topic may be influenced by other topics; motivation for both activities and topics may eb influenced by schooling in general; and in any case school students are doing much more than just learning in school, as they are also developing in other ways. In other words, motivation for language learning is neither simple nor linear. This perspective aligns with considering motivation to be complex and dynamic, like other aspects of second language learning (Larsen-Freeman, 1997; Larsen-Freeman & Cameron, 2008). N. C. Ellis and Larsen-Freeman (2006) note, “Motivation is less a trait than a fluid lpay, an ever- changing one that emerges from the processes of interaction of many agents, internal and external, in the ever-changing complex world of the learner” (p 563). Language learners need to be understood as agents acting in particular environments( Dörnyei, 2009a; Ushioda, 2009, 2011). Yet Dörnyei (2005) points out “hardly any research has been done to examine how people deal with multiple actions and goals, how they prioritize between them, and how the hierarchies of superordinate and subordinate goals are structured” (p. 87). Although this has begun to change (e.g., Dörnyei & Tseng, 2009), it has not yet been applied to motivation for ER. Such a study would also need to include factors specific to motivation for ER . Reading in a Foreign Language 25(1) de Burgh-Hirabe & Feryok: A model of motivation for extensive reading in Japanese 74 Components of L2 Reading Motivation by Day and Bamford (1998) Day and Bamford (1998) propose four major components of L2 reading motivation, drawing on expectancy-value theories (e.g., Eccles et al., 1983; Feather, 1982; Wigfield & Eccles, 1992). According to expectancy-value theories, motivation to perform various tasks is determined by the individual’s expectancy of success in a given task and the value the individual attaches to success in that task (Dörnyei, 2001). Day and Bamford suggest that the more L2 learners perceive the likelihood of understanding a book and the value of reading, the more likely it is that they will read. Two components, materials and reading ability, are associated with expectancy of success. When L2 readers perceive that materials are interesting anda ppropriate and are readily available, they are likely to engage in reading. However, low level reading ability may mean low expectations of success—and consequently low motivation to read. Attitudes toward L2 reading and sociocultural environment are associated with value. The more positive students are toward L2 reading, the more they perceive its value, and thus the more motivated they are to read. In the sociocultural environment, family, friends, teachers, and other learners influence the decision to read in the L2. Day and Bamford suggest that materials and attitudes toward L2 reading exert greater influences on L2 reading motivation than the other two components and that therefore they can compensate for low reading ability or an unsupportive socoi cultural environment. ER Motivation Studies Two qualitative case studies conducted in university JFL settings have explored the affective aspect of ER, both offering empirical support for the claim that materials are critical to ER motivation (Day & Bamford, 1998). In a 20-week diary self-study, Leung (2002) concluded that success in ER depended on sufficient quantities of suitable books. In a thre-emonth case study, Tabata-Sandom and Macalister (2009) found that suitable materials compensated for the participant’s difficulties withk anji. There have been three studies on motivation for ER in EFL with Japanese high school students. Two highlighted the role of the sociocultural environment in ER. Takase’s (2007) mixed methods study of 219 EFL learners in a Japanese high school showed parents and university entrance exams shaped motivation, which supports other research (e.g., Tachibana, Matsukawa, & Zhong, 1996) on their powerful influence. Nishino (2007) investigated motivational change over 2.5 years of ER. While generally supporting Day and Bamford’s four components, this study also highlighted the role of university entrance exams. Nishino (2007) also suggested that L2 reading motivation was a dynamic process influenced by various factors. This poinits also made by Judge (2011), whose study focused on “enthusiastic readers” (p. 165). Although two factors (love of L1 reading transferring to the L2 and the appeal of ER autonomy) emerged as important, the findings showed there were no clear patterns to ydnamically changing motivations for ER. Reading in a Foreign Language 25(1) de Burgh-Hirabe & Feryok: A model of motivation for extensive reading in Japanese 75 A Model of Motivation for ER Reading We present a model of motivation for ER (see Figure 1). The model appears linear because at its largest level it was generated from a project with a beginning and end and with idscrete phases, which is represented by the main phase cycle. Motivation for participating in an ER project could be repeated if the project were repeated, which is represented by the arrow that goes from the project postactional phase to the project preactional phase. Motivation for ER in itself (at the level of book-reading experiences as opposed to the project experience) is represented as a subphase cycle within the project actional phase. This level appears linear in the model insofar as selecting ab ook at school, taking it home to read, and evaluating it after it was read was linear. However, because some factors may influence participants across the subphase cycle, broken lines delineate the subphases to represent permeability. The multiple iterations of book reading and the influence of each experience on the next are represented by the arrow going from the ER postactional subphase to the ER preactional subphase. The explanations of the phases and subphases, discussed below, will account for how motivation is complex and dynamic . Project preactional phase. This phase concerns decisions to participate in the ER project. It includes influences from the preactional phase of Dörnyei and Ottó's (1998) model. It also includes three influences from Day and Bamford’s model for ER (1998),a ttitudes toward Japanese and Japanese learning environment and contextual influences. The latter includes environmental effects (e.g., expectations of the teachers and parents), which are often related to external demands (e.g., examinations). Project actional phase. As shown in Figure 1, the actional phase of our model includes a smaller cycle with its own preactional, actional, and postactional subphases. It includes influences from both Dörnyei and Ottó's (1998) model and Day and Bamford’s (1998) model. The preactional subphase, in which ER decision-making occurs, is where the intention to read is actualized. It includes incentive values, intrinsic pleasure (such as that derived from the interesting topics of the materials, and autonomy, which is related to making choices about when, where, and what to read), and perceived progress and success. Contextual influences and self- regulatory strategies are also influences, as each decision to read can be influenced by etxernal changes. The actional subphase occurs once the individual starts reading. It includes materia,l sas shown in Day and Bamford's model, and self-regulatory strategies, from the actional phase in Dörnyei and Ottó's (1998) model. The postactional subphase involves the learner's evaluation of ER and possible inferences for future ER. It includes attributional style and biases (variation in how people explain the event); self-concept beliefs (self-confidence, self-efficacy, self-competence, and self-worth); and evaluational clues (the students’ own perceived progress). The arrow linking the postactional subphase to the preactional subphase shows how evaluations of specific ER actions can affect other specific ER actions in a repeated cycle . Reading in a Foreign Language 25(1) de Burgh-Hirabe & Feryok: A model of motivation for extensive reading in Japanese 76 During the project actional phase in the larger cycle (i.e., the ER project), the smaller cycle was repeated multiple times. When the project came to its end, participants moved into the final phase of the larger cycle, namely the project postactional phase . Project postactional phase. The postactional phase of the larger cycle concerns the evaluation that the learner makes about ERa fter the project ended. Motivational influences on postactional evaluation in the larger cycle are the same as those in the smaller yccle, but concern the project as a whole. Reading in a Foreign Language 25(1) de Burgh-Hirabe & Feryok: A model of motivation for extensive reading in Japanese 77 Context Context Project Actional Phase Project Project Preactional Postactional Phase ER ER ER Phase Preactional Actional Postactional Subphase Subphase - Incentive values Subphase - Attributional - Instrumental benefits - Incentive values - Attributional style & style & biases - Materials - Intrinsic pleasures - Instrumental benefits biases - Attitudes - Self-regulation - Self-concept - Intrinsic pleasure - Perceived - L2 learning beliefs beliefs - Materials progress/success - Contextual influences - Environmental effects - Autonomy - Self-concept beliefs - Evaluational - External demands - Perceived progress/success cues - Evaluational cues - Attitudes - Self-regulation - Contextual influences - Distracting influences - External demands - Environmental effects Context Context Figure 1. Model of L2 ER project phases with embedded model of L2 ER subphases . Reading in a Foreign Language 25(1) de Burgh-Hirabe & Feryok: A model of motivation for extensive reading in Japanese 78 Complex and Dynamic Motivation at Two Levels We assert that motivation for ER is dynamic and complex at both the project (main phase) and book-reading (subphase) levels. Motivation is complex and dynamic in the ER project for two reasons: Some factors interact at different times in the main phase cycle, and different patterns emerge for different participants . Motivation is complex and dynamic at the level of ER book-reading experiences for two reasons: Some factors have ongoing influences so that they interact with different factors at different times, and each book-reading experience influences the next so different interactions among different factors may occur in different iterations. Aims of the Study In this exploratory case study we investigate JFL learners' motivational change during an ER project. By focusing on individual case studies, the changes each underwent can be considered. The model will be then be discussed with respect to patterns among the case studies. The research questions are: 1. Did the participants’ motivations to read extensively in JFL change? 2. What factors influenced the participants’ motivation to read in JFL? 3. Are different patterns of motivation for ER represented by the model? Methodology Research Setting The data for this study were collected from two high schools in a New Zealand city. Japanese was offered as an optional subject with four one-hour classes of instruction a week. Data was collected between February and September 2009, comprising three of four school terms (but see below) to avoid exam preparation in the fourth term . In the New Zealand school system, the National Certificate of Educational Achievement (NCEA) uses internal and external assessments to measure student skills and knowledge in different subjects. For each passing assessment, students receive credits, and high achievement is recognized. There are three levels of certificates for the three usual seconadry school years, and an optional fourth year scholarship examination (aimed at university entrance) . Students selecting Japanese had three internal assessments and an external exam. Each level of the NCEA lists the sentence structures, vocabulary andk anji which students are expected to master. Reading in a Foreign Language 25(1) de Burgh-Hirabe & Feryok: A model of motivation for extensive reading in Japanese 79 Participants Nine students in the highest levels (years 11, 12, and 13)p articipated in this study, as they were familiar with the Japanese writing system. At the request of the participating institution, sthe teachers initially invited their students to participate in the project. The first author then met with interested students to answer questions. Table 1 summarizes background information on the participants. Table 1. Study participants and background information Name (Gender) L1 Age Year NCEA Participation Alan (M) English 16 12 Level 2 Terms 1, 2 Ben (M) English 16 12 Level 2 Terms 1, 2 Drew (M) English 16 12 Level 2 Terms 1, 2 Emma (F) English 17 13 Level 3, Scholarship Terms 1, 2 Jack (M) English 15 11 Level 1 Terms 1, 2, 3 Jane (F) English 16 13 Level 3, Scholarship Terms 1, 2, 3 Josie (F) English 17 13 Level 3, Scholarship Terms 1, 2, 3 Nick (M) English 14 11 Level 1 Terms 1, 2, 3 Tracey (F) English 17 13 Level 3 Terms 1, 2 Note. Students’ names are pseudonyms. The ER Project This study differs from most ER studies because ER was done outside of class and was completely voluntary. Nine students participated in the project during term 1 (eight weeks) and term 2 (ten weeks), with four (Jack, Jane, Josie, Nick) continuing in term 3 (nine weeks). At the beginning, thfei rst author met the participants in their schools to discuss the materials (see below) and provide journals and reading records. Students were instructed to choose na easy interesting book, and to read as much as they could, preferably at least one book a week. Keeping a journal and completing the reading record were demonstrated. Students then chose their first book . During terms 1 and 2, the first author made on average 17 near-weekly visits to each school to see the participants when they returned and borrowed the ER books. The purpose of the visits was to provide ongoing support and to remind participants to read as much as possible. In term 3, she made four visits to each school. Materials Eighty graded readers (Japanese Tadoku Research Group, various dates) in four levels, which previous studies in a JFL context (e.g., Hitosugi & Day, 2004; Leung, 2002) did not use, were available to the participants in thiss tudy. In term 3, six newly published books at the lowest level were provided. From levels 0 to 4 respectively, these had approximately 100, 40-01500, 1500- 2500, 2500-5000, to 5000-10,000 characters (See Appendix A). The first author also collected 67 children's books, which were graded into four levels using the criteria in Hitosugi and Day Reading in a Foreign Language 25(1) de Burgh-Hirabe & Feryok: A model of motivation for extensive reading in Japanese 80 (2004). Procedures This study is based on a larger case study investigating JFL learners’ perceptions and motivation for ER (de Burgh-Hirabe, 2011) using multiple data collection methods. This study used interviews and journal entries. The first author conducted three individual 30 to 60-minute interviews with each participant in terms 1 and 2, and an additional fourth interview with each participant who continuedin term 3 (see Appendix B). She audio recorded and transcribed the interviews. She also collected a journal entry from each participant between the first and second interviews and another between the second and third interviews. The first author iteratively analyzed the data. In the initial coding of interview transcripts and journal entries, themes related to the research questions were underlined and assigned a code that clearly responded to the theme (Ellis & Barkhuizen, 2005). For example, “theyj ust help so much” was underlined and assigned the codeE R is helpful, and “I had a lot of speeches” was assigned the code busy with schoolwork. An experienced coder also coded one transcript using the code list. Discrepancies were discussed and resolved,a nd modifications made to some codes . After transcripts and journal entries were coded, the first author assigned more abstract codes (categories) that subsumed similar themes from the initial coding (Dörnyei, 2007). At this level of coding, the first author borrowed relevant terms from Dörnyei and Ottó’s (1998) process model, which listed a range of motivational influences. To illustratei,n strumental benefits was assigned to the code ER is helpful, and external demands was assigned to the code busy with schoolwork. These categories were also discussed with the experienced coder and discrepancies resolved, with some categories rejected while others were rearranged, and labels refined. Thus both inductive and deductive approaches were used to analyze the adta (Ellis & Barkhuizen, 2005). The definition of motivation for ER has been operationalized as the observable commitment to the extensive reading project displayed through the number of books read and amount of time spent. This operationalization is not intended as reductive, but to offer an observable and quantifiable element to anchor the inferences required to extract information about mental states from participant statements. Findings We present case studies of the 9 participants’ ER experineces, focusing on how their motivation changed and what influenced those changes. The case studies begin with the participants’ reasons for entering the project prior to its actual start (thep roject preactional phase). They then present what happened during terms 1 and 2, and for those who continued, term 3 (the project actional phase, which subsumes the subphase cycle). They conclude with the participants’ postproject evaluations (the project postactional phase) . Reading in a Foreign Language 25(1) de Burgh-Hirabe & Feryok: A model of motivation for extensive reading in Japanese 81 An overview by school terms is provided ni Table 2, which shows the number of ER books the participants read, and Figure 2, which shows the time they spent on ER . Table 2. The number of books the participants read Term1 (8 weeks) Term 2 (10 weeks) Term 3 (9 weeks) Total Alan 6 10 ---- 16 Ben 3 4 --- 7 Drew 5 3 --- 8 Emma 15 8 --- 23 Jack 6 4 2 12 Jane 13 2 0 15 Josie 17 11 6 34 Nick 2 4 8 14 Tracey 16 10 --- 26 Note. The students who did not participate in term three have- -- in that cell. Figure 2. Time the participants spent on reading. Alan Alan decided to participate in ER because “it’ll only make me better… not worse.” In term 1 Alan noted that ER “can take time to get used to,” so he used rme inders to establish a weekly routine. By the end of term 1, Alan found the readers intrinsically satisfying and interesting. Reading in a Foreign Language 25(1)