You may notice some images loading slow across the Open Collections website. Thank you for your patience as we rebuild the cache to make images load faster.

Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Multilingual undergraduate writers' discourse socialization in a sheltered academic English program Haggerty, John 2019

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2020_may_haggerty_john.pdf [ 1.67MB ]
Metadata
JSON: 24-1.0385528.json
JSON-LD: 24-1.0385528-ld.json
RDF/XML (Pretty): 24-1.0385528-rdf.xml
RDF/JSON: 24-1.0385528-rdf.json
Turtle: 24-1.0385528-turtle.txt
N-Triples: 24-1.0385528-rdf-ntriples.txt
Original Record: 24-1.0385528-source.json
Full Text
24-1.0385528-fulltext.txt
Citation
24-1.0385528.ris

Full Text

    MULTILINGUAL UNDERGRADUATE WRITERS’ DISCOURSE SOCIALIZATION IN A SHELTERED ACADEMIC ENGLISH PROGRAM   by  JOHN HAGGERTY  B.A., The University of Toronto, 1992 M.A., Carleton University, 2011   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES  (Language and Literacy Education)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   November 2019  © John Haggerty, 2019   ii  The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled:  Multilingual undergraduate writers’ discourse socialization in a sheltered academic English          program                                                                                                                                                              Submitted by             John Haggerty                          in partial fulfillment of the requirements for  for the degree of        Doctor of Philosophy                                                                                        in                               Language and Literacy Education                                                                        Examining committee: Dr. Ling Shi, Language and Literacy Education                                                                                 Supervisor  Dr. Patricia Duff, Language and Literacy Education                                                                          Supervisory committee member  Dr. Steven Talmy, Language and Literacy Education                                                                        Supervisory committee member  Dr. Tom Sork, Educational Studies                                                                                                     University examiner  Dr. Guofang Li, Language and Literacy Education                                                                            University examiner       iii  Abstract Most universities in English-dominant countries have been competing to attract multilingual learners for some time, inspired by the dual need for brain power and income generation (Lee, Maldonado-Maldonado, & Rhodes, 2006). In the Canadian context, this has resulted in rising international student populations (Anderson, 2015) and the expansion of increasingly sophisticated academic language programs (Fox, Cheng, & Zumbo, 2014). Despite this, external research into the effectiveness and appropriateness of these programs from the perspectives of the students enrolled remains scarce (Keefe & Shi, 2017).     This multiple case study involves six multilingual learners enrolled in a newly-designed academic language program in a Canadian university. This first-year program provided content and academic language courses in two disciplinary areas (Arts and Sciences), which upon successful completion, qualified students for their second year in the university mainstream. In this study, I investigate how students responded to program design features and academic writing instruction. I incorporate multiple interviews with students, collection of their written assignments and feedback, observations of classrooms and other educational events, interviews with other program stakeholders, and collection of program documents. Of the six student participants in this study, four were successful and two were less successful. For the four successful students, participation in the sheltered program was perceived as an overall beneficial experience that helped them make a positive transition to mainstream studies. However, responses to academic writing instruction and practice were highly variable and influenced by students’ backgrounds and their educational or disciplinary beliefs. For two less successful students, notions of agency, identity, and appropriation became influential in their transitions as they increasingly reported confusion, frustration, and conflict in meeting academic  iv  expectations. Results suggest there are several opportunities and challenges involved in the integration of sophisticated theoretical and pedagogical approaches, some of which may not be realized for some time after instruction has ceased. The study highlights an ongoing need to: 1) (re)consider the time needed and the degree of complexity involved in academic writing instruction and, 2) maximize alignment of pedagogical objectives with multilingual learners’ backgrounds as well as their perceived academic and disciplinary writing needs.   v  Lay Summary In this study, I investigate the transitional experiences of six multilingual learners who enrolled in a specialized first-year undergraduate program in a Canadian university that provided for-credit content and language courses in two disciplinary areas (Arts and Sciences). The study focuses on students’ changing perceptions and performances of written academic discourse. I conducted several interviews with students and other stakeholders, collected written assignments and feedback, observed classrooms and other educational events, and collected program and curriculum documents. Results indicate that there were several opportunities and challenges involved for the participating multilingual learners and that their transitions were greatly influenced by their educational backgrounds and their responses to new pedagogical approaches and instructional practices. The study suggests a need to (re)consider the time needed and level of complexity involved in the teaching, learning, and assessment of academic writing in relation to language learners’ perceived linguistic, academic, and disciplinary needs.      vi  Preface All chapters included in this dissertation were researched, analyzed, and written by the author, John Haggerty. This study was approved by the University of British Columbia’s Behavioural Research Ethics Board (UBC BREB Number: H14-01669) under the original project title: Sheltered to mainstream? Language socialization, academic identities, and disciplinary writing practices in a Canadian university.      vii  Table of Contents  Abstract .......................................................................................................................................... iii Lay Summary .................................................................................................................................. v Preface............................................................................................................................................ vi Table of Contents .......................................................................................................................... vii List of Tables ................................................................................................................................. xi List of Figures ............................................................................................................................... xii List of abbreviations .................................................................................................................... xiii Acknowledgements ...................................................................................................................... xiv Chapter 1: Introduction ................................................................................................................... 1 1.1 Multilingual writers in global education contact zones ........................................................ 1 1.2 Description of study .............................................................................................................. 5 1.3 Structure of dissertation ...................................................................................................... 10 1.3.1 Chapter 2: Investigating multilingual writers in academic contexts ............................ 14 1.3.2 Chapter 3: Methods ...................................................................................................... 14 1.3.3 Chapter 4: Sheltered to mainstream? Multilingual students’ first-year experience             in a specialized academic English program ........................................................................... 14 1.3.4 Chapter 5: Multilingual writers’ disciplinary socialization in a sheltered        undergraduate language program .......................................................................................... 15 1.3.5 Chapter 6: Agency, appropriation, and identities in first-year undergraduate             writing in the Arts and Sciences. ........................................................................................... 15 1.3.6 Chapter 7: Conclusion .................................................................................................. 16 Chapter 2: Investigating multilingual writers in academic contexts ............................................. 17 2.1 Labels and distinctions in academic language programs .................................................... 17 2.1.1 An “alphabet soup” of acronyms .................................................................................. 17 2.1.2 Sheltered instructional approaches ............................................................................... 20 2.1.3 Credit-bearing language programs and courses ............................................................ 22 2.2 Pedagogical approaches to academic writing...................................................................... 23 2.2.1 From product to process to post-process? .................................................................... 24 2.2.2 Genre-based pedagogies ............................................................................................... 25 2.2.3 Explicit academic writing instruction ........................................................................... 28 2.3 Ethnographically-oriented academic writing research ........................................................ 31  viii  2.3.1 Three orientations to disciplinary socialization ............................................................ 32 2.3.2 Second language writing research ................................................................................ 34 2.3.3 Academic discourse socialization research .................................................................. 40 Chapter 3: Methods ....................................................................................................................... 49 3.1 Multiple case study.............................................................................................................. 49 3.2 Research context ................................................................................................................. 51 3.3 Participants .......................................................................................................................... 54 3.3.1 Student participants ...................................................................................................... 54 3.3.2 Non-student participants ............................................................................................... 58 3.4 Ethical considerations ......................................................................................................... 59 3.5 Data collection and analysis ................................................................................................ 61 3.5.1 Interviewing participants .............................................................................................. 61 3.5.2 Collecting assignments and documents ........................................................................ 65 3.5.3 Observing classrooms and other events ........................................................................ 68 3.5.4 Thematic analysis ......................................................................................................... 69 Chapter 4: Sheltered to mainstream? Multilingual students’ first-year experience in a        specialized academic English program ......................................................................................... 73 4.1 Introduction ......................................................................................................................... 73 4.2 Challenges in researching the student experience in academic programs .......................... 75 4.3 Description of study ............................................................................................................ 80 4.3.1 Research context ........................................................................................................... 81 4.3.2 Participants ................................................................................................................... 82 4.3.3 Data collection and analysis ......................................................................................... 82 4.4 Results and discussion ......................................................................................................... 83 4.4.1 The program perspective .............................................................................................. 83 4.4.2 The shared perspective ................................................................................................. 90 4.4.3 The student perspective ................................................................................................ 99 4.5 Conclusion ......................................................................................................................... 107 Chapter 5: Multilingual writers’ disciplinary socialization in a sheltered undergraduate         language program........................................................................................................................ 112 5.1 Introduction ....................................................................................................................... 112 5.2 Disciplinary socialization and multilingual writers .......................................................... 115  ix  5.3 Description of study .......................................................................................................... 119 5.3.1 Research context ......................................................................................................... 120 5.3.2 Participants ................................................................................................................. 125 5.3.3 Data collection and analysis ....................................................................................... 127 5.4 Results ............................................................................................................................... 128 5.4.1 Yvonne (Arts) ............................................................................................................. 130 5.4.2 Chloe (Arts) ................................................................................................................ 134 5.4.3 Dawn (Sciences) ......................................................................................................... 140 5.4.4 Theo (Sciences) .......................................................................................................... 145 5.5 Discussion and conclusion ................................................................................................ 149 Chapter 6: Agency, appropriation, and identities in first-year undergraduate writing in the           Arts and Sciences ........................................................................................................................ 158 6.1 Introduction ....................................................................................................................... 158 6.2 The socialization of written academic discourse ............................................................... 160 6.3 Description of study .......................................................................................................... 164 6.3.1 Context and participants ............................................................................................. 165 6.3.2 Theoretical and methodological approach .................................................................. 165 6.3.3 Data collection and analysis ....................................................................................... 168 6.4 Results and discussion ....................................................................................................... 169 6.4.1 Jasmine’s delayed transition to the mainstream ......................................................... 169 6.4.2 Ashlee’s disciplinary conflict ..................................................................................... 177 6.5 Appropriating written academic discourse ........................................................................ 186 Chapter 7: Conclusion................................................................................................................. 193 7.1 Teaching and learning in global education contact zones ................................................. 193 7.2 Manuscript chapter summaries.......................................................................................... 196 7.2.1 Chapter 4..................................................................................................................... 196 7.2.2 Chapter 5..................................................................................................................... 198 7.2.3 Chapter 6..................................................................................................................... 199 7.3 Limitations and additional challenges ............................................................................... 200 7.4 Questions, suggestions, and closing thoughts ................................................................... 203 References ................................................................................................................................... 206 Appendix A: Recruitment Letter ................................................................................................ 226  x  Appendix B: Letter of Initial Contact ......................................................................................... 227 Appendix C: Informed Consent Form (student participants) ..................................................... 229 Appendix D: Sample Interview Questions (student participants)............................................... 232 Appendix E: Sample Interview Questions (non-student participants) ........................................ 235 Appendix F: Classroom Observation Sheet ................................................................................ 236     xi  12 50 53 56 62 138 148 181 183 188  List of Tables   Table 1.1: Manuscript chapters and overarching research questions…………………………..… Table 3.1: Research questions, focal participants, and data collection…………………………... Table 3.2: Minimum English language admission standards for MSU and OC ...……………..... Table 3.3: Student participants’ background information ....…………………………………….. Table 3.4: Interviews conducted with student participants…………………………………...….. Table 5.1: Chloe's first-year grades ...…………………………………………………………... Table 5.2: Theo's first-year grades ..…………………………………………………………..... Table 6.1: Ashlee's introduction (first draft) ………………………………......…………...…... Table 6.2: Ashlee's introduction (final submission) .…….…………...……….………………... Table 6.3: Salient themes in participants’ written academic socialization ……......……….........    xii  List of Figures   Figure 3.1: Example of descriptive nodes identified for Chapter 4 .............................................. 70 Figure 3.2: Example of interpretive nodes identified for Chapter 4 ............................................. 71    xiii  List of abbreviations  ADS Academic discourse socialization CBI Content-based instruction CLAS Collaborative learning annotation system CLIL Content and language integrated learning EAP English for academic purposes ESL English as a second language ESP English for specific purposes IELTS International English Language Testing System L1 First language L2 Second language LS Language socialization MSU Main Stream University OC Oasis College RGS Rhetorical Genre Studies SFL Systemic functional linguistics TOEFL iBT The internet-based version of the Test of English as a Foreign Language        xiv  Acknowledgements Throughout this professional and personal journey, there have been several individuals who have provided timely advice and crucial support along the way. First and foremost, this research would not have been possible without the student participants who contributed their beliefs, opinions, and writing with me over the course of the study, some remaining in contact with me for several years. I am forever indebted to you all for sharing your experiences and personal thoughts. I would also like to express my deep appreciation to the administrators, instructors, and other program and institutional stakeholders who participated and provided support in vital ways. Despite their busy schedules and diverse responsibilities, these individuals provided me with program documents or classroom materials, participated in interviews and discussions, or enabled observations of classrooms and other events. I remain deeply inspired and humbled by their steadfast dedication to delivering the highest quality of education possible.  Without kind support and excellent mentorship from my dissertation advisor, Dr. Ling Shi, and my committee members, Dr. Patsy Duff and Dr. Steven Talmy, I am quite confident I would not have completed this research. There were several moments when I had contemplated giving up; however, someone from my committee was always there to provide sage advice or just a sympathetic ear when I needed it most. I will always appreciate their skillful assistance in guiding me through the ups and downs of the doctoral process, and how to respond appropriately when things don’t turn out as imagined. I also had the privilege of taking courses or working with several outstanding individuals at UBC. I am very thankful to Dr. Reginald D’Silva, Christopher Fernandez, Dr. Dianne Fouladi, Dr. Ryuko Kubota, Dr. Bonny Norton, Dr. Anthony Paré, Dr. Victoria Purcell-Gates, Sheri Wenman, and especially Dr. Sandra Zappa-Hollman who provided kind support and wise advice throughout my years at UBC. I’d also like to thank the examination Chair, Dr. Shauna Butterworth, and examiners, Dr. Guillaume Gentil, Dr. Guofang Li, Dr. Steve Marshall, and Dr. Thomas Sork for their careful consideration of my research and their helpful comments at the defence.  Although it seems like an eon ago, I feel privileged to have entered the program in the company of superb colleagues. To my good friend and fellow PhD trekker, Melanie Wong, I can’t thank you enough for all our frank discussions and regular rant sessions – it was always therapeutic! I also cherish the close friendships I’ve developed with my cohorts, Tomoyo Okuda, Ernesto Peña, Espen Stranger-Johansson, Anar Rajabali, and Harini Rajagopal. Despite moving on in our lives and working across different disciplines, I feel as close to you now as when we all first met as naïve PhD students in our first year. I am also indebted to several senior PhD colleagues, Tim Anderson, Ryan Deschambault, Ismaeil Fazel, Joel Heng Hartse, and Won Kim, who provided very timely assistance and mentorship through their scholarship as well as open discussions of their doctoral research experiences.   I must also acknowledge the support of LLED and UBC who awarded me a 4-year Doctoral Fellowship and several opportunities to develop my teaching experience. I am also very grateful to the Social Sciences and Humanities Research Council of Canada who awarded me the Joseph Bombardier Canada Graduate Scholarship. These funding sources were instrumental in enabling me to successfully complete my program of study.  In closing, I thank my wife, Seong Hye, who has experienced all the highs and lows with me throughout my graduate studies. You’ve been the person I’ve relied on the most in times of struggle, and your encouragement always inspired me to continue. I’m so lucky you were by my side all these years. This is as much your accomplishment as it is mine. 1  Chapter 1: Introduction 1.1 Multilingual writers in global education contact zones The international student population in many universities has risen dramatically in recent decades (Anderson, 2015, 2016), bringing with it highly valued cultural, linguistic, and economic capital (Foreign Affairs, Trade and Development Canada, 2012, 2014; Knight, 2013). Responding to and driving these trends, a wide variety of academic language programs have been designed explicitly for multilingual learners.1 There are now several so-called pathway or bridging programs that offer university-entry options not previously available (Agosti & Bernat, 2018; Elsom, Greenaway, & Marshman, 2017). Some universities choose to enter into agreements with privately-run companies such as Navitas (https://www.navitas.com/partnerships) or Kaplan (https://kaplan.com/individuals/ prepare-for-university) to have these external organizations design and deliver programs intended to prepare multilingual learners for mainstream university study. In the case of Navitas, pathway programs are offered that, when completed successfully, enable students to proceed into the second year at the partner university. Fox, Cheng, and Zumbo (2014) observed that Canadian educational institutions have been increasingly turning to such external services to design and deliver their academic English programs and courses (p. 78). However, other universities opt to develop academic language programs in-house, relying primarily on internally-invested stakeholders most familiar with the culture and practices of the faculties and departments within the institutional structure. Despite the proliferation of these programs and the diverse theoretical and pedagogical approaches on which they are based, research conducted by those not directly involved in the design or delivery of the program remains scarce, as is longitudinal research that investigates the perspectives of the students who enrol in these programs (Fox et al., 2014;                                                  1 I use the term multilingual learners/writers throughout this dissertation to characterize participants who draw on linguistic repertories consisting of two or more languages in their communications.   2  Keefe & Shi, 2017; Tweedie & Kim, 2015).2 Students’ experiences within these programs should be a significant concern given the considerable costs, financial and otherwise, incurred by culturally and linguistically diverse learners around the world.  For a myriad of reasons, students continue to value the educational opportunities made available by English-medium universities in “inner circle” countries (Kachru, 1985). Often counselled by family members, agents, and others, many high school students compete each year to gain access to these resources and reap the benefits they believe an international education can bring. These students are likely to experience the same challenges that face many domestic high school students who make this transition (e.g., living away from family, establishing social connections, communicating with administrators and instructors, adjusting to the pace of instruction, understanding abstract concepts in lectures and academic readings, completing writing assignments, etc.). However, for students who have received all or most of their previous education in settings where English is not the primary medium of instruction, transitioning to an English-dominant context can be uniquely challenging (Duff & Talmy, 2011; Kobayashi, Zappa-Hollman, & Duff, 2017; Leki, 2007; Morita & Kobayashi, 2008; Zamel & Spack, 2004). As Duff and Anderson (2015) note, the “social, cultural, linguistic, and educational differences between students’ prior and current learning contexts, including the classroom discourse and interaction they engage in, can result in challenges and struggles for students and instructors” (p. 337). In addition to perceptions influenced by sociocultural norms and previous educational approaches, multilingual students bring with them highly individualized discourse practices, some of which may not be well understood or highly valued in certain English-medium post-secondary institutions or disciplines (Atkinson & Ramanathan, 1995; Cheng & Fox, 2008; Horner, Lu, & Matsuda, 2010; Shi, 2010,                                                  2 This does not include in-house research conducted to assess student needs, evaluate program objectives, or improve curriculum design and delivery.  3  2011). This can create several social, cultural, and linguistic challenges for multilingual learners in negotiating what is valued (i.e., expected and rewarded) as they transition to new sociocultural contexts (Duff, 2007a, 2010).  For the above reasons and others, concerned researchers have been investigating the perceptions and practices of teachers and learners who are making transitions to new cultural and linguistic educational environments, or what some scholars have conceptualised as “global education contact zones” (Canagarajah & Matsumoto, 2017; Scotland, 2014; Shi, 2009; Singh & Doherty, 2004). Pratt (1991) introduced the notion of educational “contact zones” in part “to contrast with ideas of community that underlie much of the thinking about language, communication, and culture that gets done in the academy,” wherein languages are mis-conceptualized as “discrete, self-defined, coherent entities, held together by a homogeneous competence or grammar shared identically and equally among all the members” (p. 37). Rather, within sociohistorical and sociocultural contact zones, “cultures meet, clash, and grapple with each other, often in contexts of highly asymmetrical relations of power” (p. 34).  The notion of contact zones has been applied to educational contexts in which English language instruction is provided for multilingual learners, including English for academic purposes (EAP) programs, English as second language (ESL) programs, English writing courses, and foundational preparatory programs (e.g., Canagarajah, 1997; Canagarajah & Matsumota, 2017; Singh & Doherty, 2004). Within these programs, as Canagarajah and Matsumota (2017) point out, “negotiations are marked by the stresses and strains of inequality” (p. 391) and the explicit and implicit understandings of the individual actors involved directly impact how languages are conceptualized and the kinds of communication that can and cannot occur. As a result, newcomers to these zones may face challenges understanding the practices and expectations of more  4  experienced and powerful others. While all incoming students can struggle in their transition to undergraduate studies, having experience with and more implicit understanding of the sociocultural and socioeducational practices surrounding and informing institutional policies and instructional methods reduces some of the complexities involved.  While there are numerous language and literacy practices in need of further investigation within various contact zones around the world, the primary focus of this study is the socialization of written academic discourse. This focus is not intended to minimize other vital aspects of academic socialization in undergraduate contexts such as classroom discourse, peer interaction, academic presentations, academic counselling and tutoring, networks of practice, and numerous others. Rather, it is intended to enable a more focused perspective on an academic activity, commonly termed “academic writing,” that has long been associated with success in higher education (Flowerdew, 2016; Hyland, 2013). Written academic discourse has also been identified as a significant source of concern, confusion, and apprehension for multilingual learners (Lillis, 2001; Ridley, 2004), is the focus of numerous required “foundational” programs in English-medium universities (Braine, 1996; Matsuda, Fruit, & Burton Lamm, 2006), and is featured prominently in pre-university and undergraduate-level academic language support programs around the world, most notably academic writing centers (Okuda, 2019; Williams, 2002).  The perceived importance of written academic discourse in university contexts is reinforced by its inclusion on internationally-recognized English language proficiency tests such as the internet-based version of the Test of English as a Foreign Language (TOEFL iBT) and the academic version of the International English Language Testing System (IELTS). These large-scale standardized language tests (and others) have had a considerable impact on education systems around the world, often encouraging decontextualized and highly-structured approaches to English  5  language teaching intended to maximize test performance (Cheng, Watanabe, & Curtis, 2004). In some test-intensive educational contexts, this impact extends into early adolescence as students prepare for and take such tests to qualify for highly-competitive international high schools, generally considered a stepping stone to prestigious English-medium universities abroad (Choi, 2008; Haggerty & Fox, 2015). Such transnational testing practices, and the washback they engender in public and private schools around the world, can greatly influence multilingual learners’ understanding and practice of written academic discourse, sometimes in ways that are incongruent with the disciplinary practices of the English-medium universities they subsequently enter.  1.2 Description of study In this multiple case study, I investigate six multilingual learners’ perceptions and experiences as they transitioned into and out of a new academic language program situated in a Canadian university. This 11-month program was intended for multilingual students who had not fully satisfied the direct-entry English language proficiency requirements required by the university (often based on TOEFL iBT or IELTS Academic results). The program offered first-year credit-bearing undergraduate content and academic language courses in either an Arts or Sciences Stream.3 Program curriculum drew upon a range of theoretical and pedagogical approaches, including but not limited to content and language integration (CLIL), systemic functional linguistics (SFL), genre-based pedagogies, task-based learning, flipped classrooms, two-stage collaborative group exams, apprenticeship scholarship, experiential learning, multiliteracies, disciplinary-specific materials, and a range of educational technologies such as the collaborative learning annotation system or CLAS. This sophisticated integration was also designed internally through close collaboration among stakeholders working within the institution in which it was                                                  3 The Sciences Stream had two specializations (physical and computational sciences), but all Sciences Stream students took the same academic language courses.  6  housed. After successfully completing this program, students qualified for entry into their second year of mainstream study within the wider institution. All six focal participants in this study were enrolled in the inaugural year of this program. This provided an opportunity to investigate the mediating influence of this program, including its innovative curriculum and integrated academic language courses, on students’ first-year experience. Since the program included explicit instruction of academic and disciplinary writing, it provided an opportunity to investigate the socialization of written academic discourse while students were completing writing assignments for their respective language and content courses. It also afforded an opportunity to track the perceptions and performances of these students as they transitioned into the university mainstream. Throughout this study, I refer to this program with a pseudonym, Oasis College (OC), and characterize it as “sheltered” because participants took courses exclusively with other multilingual students who, like them, had been unable to satisfy direct-entry English language admission requirements.4 According to Short, Fidelman, and Louguit (2012), sheltered instruction involves the “practice of integrating language development with techniques to make curricular topics more comprehensible to ELLs [English language learners]” wherein the “main focus is on subject-specific curriculum and the instructor is a content specialist” (p. 335). Instructional strategies associated with sheltered instruction include adjusting the cognitive load by more carefully selecting content to be focused on but not altering complexity or grade-level appropriateness. This approach affirms and enables the abilities of multilingual learners to participate in higher-order thinking and rejects a deficit model of instruction (Echevarría & Graves, 2010). Sheltered instructional approaches have been incorporated into K-12 and higher education contexts for decades (Knoblock & Youngquist, 2016; Short et al., 2012), and although applied in various ways,                                                  4 It is important to note that the term sheltered was not used in program documentation or by program administrators.  7  the term captures a salient aspect of the academic language program involved in this study: namely, that its primary purpose was to assist a specific subset of the student population to perform in the mainstream by providing focused and targeted content and language instruction. OC was also “sheltered” in that it included several credit-bearing content courses taught by content experts that were designed with the needs of multilingual learners in mind. However, this program also included credit-bearing academic language courses taught by language specialists, some explicitly linked to content courses and others that were not. Unlike some sheltered instructional approaches, this program did not incorporate the use of students’ first language (L1) in school-sanctioned curriculum. The use of L1 in sheltered content and academic language courses remains controversial because “monolingual immersion ideologies are still dominant in many contexts in the world” (Lin, 2015, p. 74). As such, there has been a tendency to depreciate learners’ first language as unnecessary, inferior, or interfering with English language instruction (McMillan & Turnbull, 2009) despite a growing body of research that points to several cognitive and affective benefits that can come about with moderate inclusion of multilingual learners’ full linguistic repertoires (Canagarajah, 2011, 2013; Cummins, 2007; Littlewood & Yu, 2009). Researchers who have explicitly investigated the use of students’ L1 in combination with English in higher education contexts have discovered that students often benefit when opportunities are made available to communicate with others in multiple languages (e.g., Galante et al., 2019; Laupenmühlen, 2012; Storch & Wigglesworth, 2003).         In undertaking this study, I did not intend, nor was I able, to conduct program evaluation, student needs analysis, or assess the efficacy or appropriateness of pedagogical methods. Nevertheless, I draw on scholars and researchers publishing in these areas of inquiry to better understand the wider context surrounding this research site and the challenges and opportunities  8  language programs have experienced in English-dominant academic contexts. However, my primary objective in conducting this study was to investigate the perceptions and experiences of six multilingual learners as they transitioned through this program and progressed into the university mainstream. More specifically, I was focused on how these multilingual learners responded to program design features and their academic writing instruction while making their transitions. I also draw on the perspectives and experiences of other stakeholders (a consultant, two administrators, and two instructors), my observations of classrooms and other educational events, program and curriculum discourse, and my research notes. Given that academic language programs experience many challenges (Brown, 2009), particularly in their first year, it is important to keep in mind the pressures program designers, administrators, instructors, and students are typically under and their need to address any “messes” that occur as they happen (Morris, 2006, p. 587). The need to hire suitable and effective instructors and acclimatize them to program objectives and approaches creates challenges for most administrators. Instructors, for their part, require time to learn and adjust to valued pedagogical objectives and incorporate them into their classroom practices. Also, the students who enrol in these programs do so for a wide variety of reasons and may have expectations based on previous learning experiences that are incongruent with program practices. Some students may not feel they need additional language support or disagree with the pedagogical approaches taken regardless of their presentation. For these reasons and others, there is a need to maintain realistic expectations and make cautious interpretations given the challenges involved in preparing multilingual learners for the academic and disciplinary expectations of English-medium higher education institutions (Bazerman et. al, 2017; Duff, 2010; Spack, 1997).  9  In conducting this research, I have been greatly inspired by my personal, professional, and academic experiences over the last two decades. I first became aware of the many challenges and accomplishments of multilingual writers when I went to South Korea in 2001 to teach English. Although not planned, I ended up staying for over a decade assisting young, adolescent, and university-level learners for various academic purposes. In the process, I gained deep appreciation for the difficulties many students experienced learning to write academic English in a country where English was not the dominant medium of instruction. In my role as an academic writing teacher, I assisted many learners who were preparing to meet the expectations of numerous standardized language tests (e.g., TOEFL, IELTS, etc.). The pressure to perform well on these tests was extremely high as the results regulated access to highly-valued educational opportunities. Approaches to academic writing instruction were generally geared towards performing well on standardized language tests (sometimes referred to as ‘teaching to the test’) and tasks were highly-controlled and largely decontextualized. I struggled to find ways to engage students in academic writing and develop deeper understandings of meaning-making practices that enabled them to question, negotiate, and take more control of their writing. I also struggled because I worried about their ability to transfer what they were learning to more organic academic writing environments.  In my various teaching roles in South Korea and Canada (language instructor, content instructor, teacher educator, language test writer, etc.), the influence of the social context has never been far from my mind. As a “native” white male English speaker in South Korea, it quickly became apparent to me that I enjoyed a privilege based on a widespread perception of the value and necessity of English. Over the years, I have become increasingly concerned about the effects of “native speaker” conceptions in relation to the teaching, learning, and assessment of writing, and the impact this can have on multilingual learners around the world (Atkinson & Ramanathan, 1995;  10  Holliday, 2005; Leung, Harris, & Rampton, 1997; Shohamy, Or, & May, 2017). These experiences have impressed on me the need to investigate and engage with multilingual perceptions and experiences to gain a deeper understanding of what pedagogical approaches can work, when, and why. It is only by questioning commonly-accepted practices and opening safe spaces for meaningful dialogue that I believe we can take full advantage of the opportunities available, while limiting the negative consequences that can result when assumptions are left unexamined. These experiences have motivated me to explore the longer-term perceptions and experiences of multilingual learners who are making transitions to new socioeducational environments, and who, unlike my transition to South Korea, must do this in their second or additional language and without the same level of social, cultural, and linguistic capital that I possessed (Bourdieu, 1977).  In this study, I investigate students’ perceptions and performances of academic writing, which includes consideration of the texts being co-constructed as well as aspects of the (micro to macro) contexts involved in their co-construction. In exploring these changing perceptions and performances, I am interested in how multilingual writers were socialized or apprenticed into written academic discourse, how they participated in the process, and the various ways they responded over time and across contexts.          1.3 Structure of dissertation In this dissertation, I incorporate a manuscript-based format (Paltridge, 2002; Paltridge & Starfield, 2007), which is composed of a series of independently-publishable articles that explore separate (albeit related) aspects of a larger research project.5 The structure of a manuscript-based dissertation is variable and depends on the field of research, the nature of the research undertaken, and departmental and institutional norms. However, unlike a “traditional” dissertation, manuscript-                                                 5 The three research articles presented (the manuscript chapters) have not been submitted for publication. UBC does not require these to be published prior to submission of a manuscript-based dissertation. The purpose of this format is to expedite timely dissemination of the work either during or immediately after the PhD program.   11  based formats tend to differ in terms of the intent and location of the literature review, often found within and restricted to the specific focus of each manuscript chapter. As such, there are often notable differences in terms of the “intended audience, level of detail, and display of knowledge” (Paltridge & Starfield, 2007, p. 74). A manuscript-based approach creates advantages and disadvantages for both the writer and the reader. While this can assist the writer in developing stand-alone articles to be published from their dissertation (something that can be challenging in a traditional format), it also presents difficulties in terms of structuring the chapters in a readable and logically-sequenced manner while also meeting the expectations of an audience who may be unacquainted with variation in the dissertation genre. For the reader, comprehending separate but related manuscript chapters can feel disjointed in that there is not one grand narrative running through all the chapters. It may also seem repetitive (e.g., the theoretical frame, research context, participants, etc.). For these reasons, steps have been taken throughout to try to maximize the advantages and minimize the disadvantages of a manuscript-based dissertation format as much as possible.  The format of this dissertation is non-traditional and perhaps best described as hybridized in that it incorporates elements of traditional and manuscript-based approaches. In keeping with more traditional formats, Chapter 2 provides a historical overview and discussion of relevant research and the theoretical approach that has guided and informed the overall study. Chapter 3 provides a discussion of the research methods I incorporated. Then, breaking with traditional formats, there are three manuscript chapters (Chapters 4, 5, and 6), each with a different thematic focus. While Chapter 2 provides more breadth, the literature reviews in each of these chapters are more selective in that they target studies of most relevance. Also, in terms of theoretical and methodological approaches, only information that is unique to each study is reported within the respective  12  manuscript chapter. Information on methods that is common across all three studies is presented in Chapter 3 in more detail. Finally, a concluding chapter (Chapter 7) brings together these results and provides a discussion of the overall findings, implications, and suggestions.   As shown in Table 1.1, each manuscript chapter was initially guided by one of three overarching research questions. Although these should not be viewed as mutually exclusive, these broad questions led to more focused research questions established at various points in the research process (described in Section 3.1). Table 1.1: Manuscript chapters and overarching research questions Chapter Overarching research question Four How do program design features shape participants’ reported beliefs about the program and their experiences within it? Five How do academic language courses mediate students’ perceptions and performances of written academic discourse? Six What are some of the struggles experienced by multilingual writers as they transition through this program?  Several writing scholars and researchers working across multiple fields of inquiry have emphasized the importance of investigating the context surrounding and informing academic text and have called for research designs and methods that can help to narrow the text-context gap (e.g., Bazerman, 1981, 1988; Dias & Paré, 2000; Lillis & Curry, 2010; Miller, 1984; Paltridge, Starfield, & Tardy, 2016). Over three decades ago, Swales (1985) called for more recognition of the social context surrounding and informing texts because it had been given “too great a place in nature” by scholars and researchers who “believed a thick description of text is the thickest description of them  13  all” (p. 219). This sentiment is echoed by several English composition and second language writing scholars, arguing for a longer-term perspective on writing development, when they state: … writing education needs to address all the elements of writing, be built on meaning making and effective communication, and recognize social, linguistic, cognitive, affective, sensorimotor, motivational, and technological dimensions of writing development. Each dimension matures and develops across many experiences, but each writing experience brings all the dimensions together in a unified communicative event. (Bazerman et al., 2017, p. 357) Inspired by such calls, these three studies (Chapters 4-6) join others that aim to narrow this gap and situate texts and writers within their social contexts. Through triangulation of contextual and textual aspects of written academic discourse, I aim to gain a deeper understanding of the multilingual writers in my study in terms of how they understand written academic discourse and the complex reasons why academic text appeared as it did in their written assignments. I draw on the notion of triangulation, not to establish a unitary truth or a fixed reality that exists apart from individual perceptions, but rather to clarify meaning from multiple perspectives, each an interpretation of multiple realities that are inextricable from human perception, interests, and stakes (Duff, 2008; Stake, 2005). According to Natow (2019), “triangulation is useful from constructivist and critical perspectives as well” because it enables researchers to incorporate “various conceptions and understandings of reality” (p. 5). While some researchers associate this notion exclusively with positivism or realism (and paradigmatically-rigid notions of validity and reliability), they “mistakenly believe in inevitable logical connections between paradigm positions and techniques” while more tolerant methodologists argue it has a “place within a variety of paradigms” (Seale,  14  1999, p. 472).  Before proceeding, I provide a summary of all subsequent chapters in this dissertation. 1.3.1 Chapter 2: Investigating multilingual writers in academic contexts   This chapter provides an overview of literature that is intended to contextualize the study, the research site, and the specific research questions guiding each manuscript chapter. First, I discuss research that has characterized academic language programs in various ways and identify some of the salient issues identified in the design and delivery of these programs in English-medium universities. Second, I discuss a range of pedagogical approaches currently informing academic writing instruction as well as some of the issues that have been identified in relation to the explicit or implicit nature of this instruction and its proposed effectiveness and appropriateness for multilingual writers. I conclude this chapter by describing the overarching theoretical and methodological approach guiding this research, language socialization (LS), and more specifically the emerging subfield of academic discourse socialization (ADS), along with discussion of the growing body of research this approach has inspired. 1.3.2 Chapter 3: Methods This chapter provides more details about the research context and introduces the six multilingual students that are the focal participants for this study. In addition to discussing some of the ethical considerations influential in carrying out this study, I describe the methods utilized to collect data and the procedure followed to analyze and interpret the results.   1.3.3 Chapter 4: Sheltered to mainstream? Multilingual students’ first-year experience in a specialized academic English program In Chapter 4, the first manuscript chapter, I focus on how aspects of program design and delivery mediated the academic socialization experiences of six multilingual learners as they  15  apprenticed into their disciplinary areas of study. I also include the perspectives of other stakeholders in the program, including a consultant, two administrators, two instructors, and discourse from program documents (brochures, curriculum documents, website, etc.).  1.3.4 Chapter 5: Multilingual writers’ disciplinary socialization in a sheltered undergraduate language program Chapter 5 focuses on four multilingual writers (two enrolled in the Arts Stream and two in the Sciences) who satisfied OC program requirements and transitioned successfully to MSU in their second year of study. After leaving the program, these students reported an overall positive perception of their first-year program and indicated it helped them make a smoother transition to their mainstream programs of study. In this way, they represent success stories. In this chapter, I focus on students’ ongoing responses to academic writing instruction in their academic language courses and the various ways they understood and applied instructional concepts in their written assignments over time. The longer-term trajectories of two participants who remained in the study into their fourth year are also investigated.  1.3.5 Chapter 6: Agency, appropriation, and identities in first-year undergraduate writing in the Arts and Sciences. The final manuscript chapter focuses on two multilingual writers, Jasmine (Arts Stream) and Ashlee (Sciences Stream), whose transitions into the university mainstream were marked by considerable confusion, frustration, and conflict. In this study, I focus on how the students responded to academic writing instruction in their academic language courses and the social, cultural, or linguistic challenges they experienced in their transitions into their respective programs of study. The longer-term trajectory of one participant who remained in the study into her fourth year is also investigated.  16  1.3.6 Chapter 7: Conclusion In the concluding chapter, I summarize and synthesize the main findings and contributions from each manuscript chapter and the ways in which they collectively add to a deeper understanding of written academic discourse socialization. I also discuss pedagogical implications and provide some suggestions for future research. While the results of this study cannot be generalized to other contexts, I provide tentative recommendations for researchers, program designers, administrators, instructors, and students interacting in global education contact zones around the world.     17  Chapter 2: Investigating multilingual writers in academic contexts In this chapter, I draw on scholars and researchers working across multiple fields of inquiry to situate this dissertation and my research focus within each manuscript chapter. This chapter is divided into three sections. First, I provide an overview of terminology used in the literature to describe academic language programs and discuss some issues that have been identified in their design and delivery. Second, I provide a brief historical overview of pedagogical approaches to teaching writing and reflect on what we currently know about multilingual writers from around the world who are transitioning to undergraduate programs in English-medium universities. Finally, drawing on Morita and Kobayashi’s (2008) taxonomy of “second language (L2) disciplinary socialization research,” I situate my study within their classification and provide a review of socially-informed research into the socialization of academic and disciplinary discourse that incorporates an “ethnographic perspective” in the research design (Green & Bloome, 2014; Lillis, 2008; Paltridge et al., 2016). 2.1 Labels and distinctions in academic language programs The diversity of tertiary EAP program designs, and their associated curriculums, has made it challenging to describe them in generally agreed-upon terms. Given this study was conducted in a newly-designed academic language program that drew on a range of theoretical and pedagogical approaches, some discussion of how these programs have been characterized in the literature is necessary. 2.1.1 An “alphabet soup” of acronyms  Academic language programs and courses created by, or designed for, English-dominant universities have been described in various ways in research reports, so much so that some have commented on the “veritable alphabet soup of acronyms” currently in use (Brinton, 2012, p. 1). In  18  perhaps an overly expansive characterization, programs have been characterized as either English as a second language (ESL) or English for academic purposes (EAP). For example, Fox et al. (2014) distinguished language programs across Canada based on curriculum objectives and classroom activities. For their study, “ESL” programs were identified as those that emphasized “speaking, social interaction, and general language development,” while “EAP” programs emphasized “academic reading, writing, and language development” (p. 57).  As a general distinction, ESL and EAP are still commonly used to characterize English language programs (e.g., Basturkmen & Lewis, 2002; Fernandez, Peyton, & Schaetzel, 2017). However, the emergence of EAP as a discipline is one amongst several disciplinary-specific and genre-aware approaches to language instruction. As English language teaching became more globalized after the Second World War, the field of English for Specific Purposes emerged to fill a need to teach vocational and professional English in a more specialized manner (Charles & Pecorari, 2015, p. 8). EAP emerged from this desire for more specificity and has since been bifurcated into English for General Academic Purposes and English for Specific Academic Purposes based on whether a more generalized or disciplinary-specific approach to language and literacy instruction is conceptualized, with specific approaches associated with a wide array of subject areas (Hutchinson & Waters, 1987). However, the boundaries between specific and general approaches have become increasingly blurred over time (Brinton, 2012).  The academic language program that is the primary research context for this study could be characterized as EAP (rather than ESL) in that it was focused on developing academic language awareness and written performance rather than developing general English language proficiency. However, there was also an emphasis placed on social engagement and the development of students’ overall academic and life skills, and therefore included “ESL” and other social and  19  cultural aspects as well. Because the program also incorporated disciplinary-specific instruction in academic writing (separated into Arts or Sciences “streams”), the program could be characterized as English for Specific Academic Purposes (rather than general). However, there were aspects of both approaches in the academic language curriculum, and the degree of specificity was limited given this was a first-year program composed of multilingual learners deemed in need of additional academic language instruction and practice. The above challenges in characterizing the program at a general level highlight the difficulties researchers have in describing or comparing the efficacy and appropriateness of various academic language programs, whether through quasi-experimental designs, quantitative survey studies, or investigating the perspectives and experiences of administrators, instructors, or the diverse multilingual learners who enrol in them.  Academic language programs have also been influenced by approaches intended to integrate English language teaching with content instruction such as content and language integrated learning (CLIL), which originated in Europe, and content-based instruction (CBI), which originated in Canada (Cenoz, 2015). Both approaches are often found in sheltered environments that explicitly connect the teaching of language with the teaching of content in various ways. According to Brinton (2012), a CBI classroom is intended for multilingual learners who are “separated or ‘sheltered’ from their first language (L1) (i.e., mainstream) peers” based on the theoretical premise that “exposure to the rich academic language and complex concepts presented in the sheltered class provides the necessary conditions for L2 acquisition to occur” (p. 4). However, the precise nature of the curriculum designed and delivered for this purpose is variable and can be informed by several interconnected theoretical and pedagogical approaches (Brinton, 2012; Cenoz, 2015). The program in this study incorporated CLIL (by name) into its curriculum, but the way in which this impacted  20  the curriculum within specific academic language courses was variable (see Section 3.2 for more details).  2.1.2 Sheltered instructional approaches At a general level, academic language programs and their curriculum can be distinguished based on whether groups of learners are instructed separately from others based on shared background characteristics. Although typically applied to primary or secondary school contexts, the notion of sheltered instruction has been incorporated into higher education contexts as well, some arguing that such approaches can be “beneficial to culturally and linguistically diverse students with different language abilities” (Knoblock & Youngquist, 2016, p. 51). In the context of K-12 education in the U.S., Short et al. (2012) state that sheltered instruction integrates “language development with techniques to make curricular topics more comprehensible” and typically involves “second language learners (usually minority language speakers) who are studying content in the new language (usually the majority language), which is the medium of instruction” (pp. 335-336). One of the primary objectives of sheltered instruction is to prepare learners to enter the mainstream, whether that be in an elementary, secondary, or higher education setting.  In the context of mandatory first-year undergraduate writing courses common in many U.S. universities, some have argued for the benefits of sheltered instruction over mainstreaming. For example, Silva (1994) argued that mandatory placement of multilingual learners in mainstream courses constituted an unprincipled and unpredictable “sink or swim” approach (p. 38). This can lead to unfortunate outcomes for some multilingual learners who may not have the linguistic and cultural background needed to comprehend instructional content and academic expectations at the speed it is delivered. Several studies have also discovered a positive overall perception among students who participated in sheltered ESL or EAP language programs (e.g., Edwards et al., 1984;  21  Fox et al., 2014; Hauptman, Wesche, & Ready, 1988). Braine (1996), in a study of first-year writing courses in a U.S. university, found that ‘‘an overwhelming majority’’ of L2 students preferred to enroll in “ESL composition” classes over those offered in the “mainstream” (p. 99). In his study, the rate of ESL student withdrawal from mainstream writing courses was five times higher than that for sheltered ESL courses. Some of the reasons students provided for withdrawing from mainstream courses included fear or embarrassment of having to speak in these classes.  There have also been criticisms of sheltered approaches, some charging that this can lead to the “specter of segregation” (Silva, 1994, p. 40). Some have argued that the instruction provided in sheltered language programs might be overly simplified, resulting in challenges for students in coping with the speed and complexity of instruction when they enter the mainstream (e.g., Atkinson & Ramanathan, 1995; Bunch et al., 2001). Additionally, scholars and researchers have warned of the negative consequences that can result from grouping (or labeling) learners based on pre-established aspects of their English-language backgrounds. Such negative consequences include the potential to reinforce “deficit” perceptions and beliefs about multilingual learners (Canagarajah, 1999, 2013; Gallagher & Haan, 2018; Heng Hartse & Kubota, 2014; Horner et al., 2010; Marshall & Walsh-Marr, 2018; Rose, 1998). The labels that are co-constructed, and that subsequently gain currency in school contexts, have the potential to impact learning environments and individual learners in ways not intended or foreseen (e.g., Harklau, 2000; McCarthy, 1987; Talmy, 2009; Zappa-Hollman, 2007a).  Personally, I would argue that by separating learners from mainstream populations there is an inherent risk in creating or reinforcing labels (and their underlying stereotypes) to the detriment of preferred educational objectives and to the multilingual learners involved. While these decisions need to be carefully deliberated and the potential for negative consequences minimized, the benefits  22  of sheltered instruction in some contexts justify its choice, particularly when a sink or swim approach would likely lead to more negative consequences (Silva, 1994). However, these decisions should not be made without a sound rationale for the decision and a method for monitoring and addressing any negative consequences.  The program in this study separated multilingual students from mainstream content classrooms and incorporated pedagogical approaches consistent with a sheltered approach. The learners in this program did not qualify for direct-entry to a Canadian university based on their standardized English language proficiency test scores (e.g., TOEFL iBT or IELTS Academic). The curriculum was designed specifically for these learners and instruction was intended to help prepare them for future mainstream studies in their respective disciplinary areas (Arts or Sciences). The program included both content and language courses, which were linked in various ways. The program was not characterized by upper-level stakeholders as sheltered and this term was not used in program documentation I collected. Nevertheless, this characterization captures salient aspects of the program, and therefore, I refer to it as sheltered throughout this dissertation.   2.1.3 Credit-bearing language programs and courses In some cases, the completion of academic language programs or courses will count as credits that can be applied towards a larger program of study within the university, either by replacing the need to complete other required courses or serving as program credits on their own. In this way, specific courses (or even entire programs) can be characterized as for-credit (or credit-bearing). Typically, courses offered within English language programs (particularly those characterized as ESL) tend to be non-credit in this sense, although some do provide the credentials necessary to access valued educational resources (e.g., pre-admission qualifying programs). The issue of whether English language courses should be credit-bearing in the contexts in which they  23  are offered can be the subject of contentious debate. At the high school level in the Canadian context, there have been calls to make English language courses count as legitimate credits (e.g., B.C. migrant students want credit for English courses, 2013). Those who advocate for change argue that since other language courses (e.g., French) are credit-bearing, there is no reason for English language courses not to count in the same way. For many researchers and instructors working with multilingual learners, including myself, there is a pressing need to properly acknowledge the work that multilingual learners do in furtherance of their academic studies. The program in this study included credit-bearing content courses as well as academic language courses that were required to complete the program, but not necessarily credit-bearing in relation to students’ future programs of study. This is discussed in more detail in Section 3.2. The above discussion highlights how challenging it can be to characterize academic language programs or assess their relative effectiveness or appropriateness for the multilingual learners who enrol in them. The diversity of theoretical and pedagogical approaches often woven together to achieve program objectives, as well as differing beliefs about teaching, learning, and assessment, underscore these challenges. In the next section, I discuss a range of pedagogical approaches currently informing academic writing instruction to provide some context for the instructional approach taken in the program. 2.2 Pedagogical approaches to academic writing Over the last half century, novice writers entering English-medium universities in Canada and the U.S. have been taught academic writing in diverse ways, reflecting prevailing views about what effective writing looks like and how this knowledge, skill, or ability might best be developed in others. Up until the 1970s, according to Dias et al. (1999), learning to write in academically appropriate ways was largely a decontextualized and inauthentic experience for novice writers. The  24  focus of instruction, to the extent it was provided, was largely restricted to explicit instruction of grammatical features and rhetorical structures based on disciplinary exemplars (i.e., products). Most research up to that time was largely restricted to analyzing students’ final written assignments in relation to the instruction being provided (Tobin & Newkirk, 1994; Zamel, 1982). The pedagogical aim of this product-centered approach was focused on replicating the writing of supposed experts and assessing the degree to which this was accomplished in classroom contexts. There was little consideration for the writer or the reader, how texts came to be, or how writing practices change over time. However, in the 1970s, the rise of process approaches (e.g., Zamel, 1976) helped pave the way for teachers to adopt new roles as “empathetic and supportive readers rather than expert correctors and assessors” (Dias et al., 1999, p. 7).  2.2.1 From product to process to post-process?  According to the historical narrative (Matsuda, 2003), scholars and researchers in L1 composition and L2 writing studies became dissatisfied with the product-centered approaches of the past. As a result, writing scholars began to develop process-oriented and student-centered approaches that aimed to integrate an understanding of how various written texts were produced in their specific contexts (Hyland, 2003). Process writing research has been characterized as “expressivist” or “cognitivist,” the former emphasizing self-discovery, personal voice, and empowerment (e.g., Berlin, 1987; Elbow, 1988a, 1988b) and the latter emphasizing stages in the writing process, higher-order thinking, and the problem-solving strategies of successful writers (e.g., Bereiter & Scardamalia, 1987; Flower & Hayes, 1981). However, both approaches share several pedagogical aspects, including recognition of the “need to understand and cultivate novice writers’ composing processes as generative, recursive, individuated, and ‘inner-directed’” (Ferris & Hedgcock, 2014, p. 66). As such, process approaches have been predominantly cognitive,  25  incorporating methods such as think aloud protocols and examinations of writers’ texts over time, designed to better understand what was happening in the minds of L1 and L2 writers (Ferris & Hedgcock, 2014). While many tenets existed long before its apparent rise in popularity in the 1970s, the process “movement” has been characterized as the “most successful in the history of pedagogical reform in the teaching of writing” (Matsuda, 2003, p. 69).  However, dissatisfaction with aspects of the process movement, such as its predominantly cognitive theoretical orientation and formulaic set of procedures, inspired the rise of “post-process” approaches (e.g., Trimbur, 1994). Among other limitations, process approaches’ lack of explicit instruction privileged those “immersed in the values of the cultural mainstream” (Hyland, 2003, p. 19). Post-process approaches challenged writing researchers and instructors to more meaningfully consider the social and cultural contexts that shape (and have shaped) writing practices. These developments reflected a “social turn” (Block, 2003; Gee, 2000) occurring across the academy (e.g., linguistics, composition studies, second language acquisition, etc.), a turn motivated by a “lingering sense of frustration” with theories and concepts that did not adequately address “language learning as social practice and language as a social phenomenon” (Firth & Wagner, 2007, p. 801). This social turn has had an impact on writing scholars and researchers working across diverse fields of inquiry and has contributed new theoretical and methodological insights that continue to re-shape how academic texts are conceptualized for research and instructional purposes.  2.2.2 Genre-based pedagogies  Increasing recognition of the social context of writing and the implications this had for teaching, learning, and assessment resulted in significant changes in how academic writing has been conceptualized and taught in language courses. In composition studies, investigations into the co- 26  constructed nature of genre and rhetoric provided deeper insights into not only ‘what writing does’ but also ‘how it does it’ (Bazerman & Prior, 2004). This encouraged scholars and researchers to investigate aspects of the social context surrounding and informing the co-production of academic text (e.g., Artemeva & Freedman, 2008; Bazerman, 1981, 1988; Dias et al., 1999). In the field of linguistics, functional approaches to the study of language such as systemic functional linguistics or SFL (e.g., Halliday, 2007; Martin & Rose, 2003) developed sophisticated notions and frameworks for better understanding the underlying functions (i.e., social purposes) guiding language use in various disciplines and genres of written communication, developments which continue to have an impact on pedagogical approaches to academic writing instruction (e.g., de Silva Joyce & Feez, 2012; Rose & Martin, 2012). In addition, advances in computer technology have enabled increasingly sophisticated analyses of large collections (or corpuses) of academic texts produced by diverse writers across a range of contexts, disciplines, and genres (Hyland, 2009, 2012).  The rise of genre in research on written communication has led to an explosion of genre-based pedagogies whose influence is still being realized across the world. Hyland (2003) identifies three “broad, overlapping schools of genre theory,” which includes English for specific purposes (ESP), New Rhetoric, and the “Sydney School” of systemic functional linguistics or SFL (see Johns et al., 2006, and Artemeva & Freedman, 2015, for reviews). These three areas of genre pedagogy are unique in their theoretical, methodological, and pedagogical approaches; however, all share an interest in the social and functional nature of language and the need to make this knowledge more apparent to learners (Artemeva & Freedman, 2015; Hyland, 2003). While once described as three separate “camps” (Hyon, 1996), some argue these distinctions have become far more commensurate than in the past (e.g., Artemeva & Freedman, 2015; Swales, 2009). Some genre scholars have emphasized the importance of crossing “boundaries” to recognize the shared desire to  27  “encompass in theory and practice the complexities of texts, contexts, writers and their purposes, and all that is beyond a text that influences writers and audiences” (Johns et al., 2006, p. 247).     Nevertheless, there are important differences, many of which revolve around the relative attention given to text (i.e., a linguistic orientation) or the social context (i.e., a sociological orientation) (Bawarshi & Reiff, 2010). New Rhetoric (also referred to as Rhetorical Genre Studies or RGS) has been characterized as a “contextual approach” (Flowerdew & Wan, 2010) that focuses on the “functional relationship between text type and rhetorical situation” (Coe, 2002, p. 197). Similarly, Johns et al., (2006) note that the New Rhetoricians “begin with context or ideology,” while others “derive their theories and data first from the texts themselves” (p. 247). This leads researchers to focus on the “rhetorical situation and its broader contexts” in conjunction with the analysis of texts (Coe, 2002, p. 198), recognizing that all communication is ultimately social action (Miller, 1984). ESP has been characterized as primarily a “linguistic approach” (Flowerdew & Wan, 2010) and its practitioners as conceptualizing “genre as a class of structured communicative events employed by specific discourse communities” (Hyland, 2003, p. 21). ESP researchers are largely concerned with identifying regularities in the structure of texts, how these differ depending on the genre in which they are produced, and how to best disseminate this disciplinary and linguistic-specific knowledge to learners (Hyon, 2018). Finally, SFL-informed researchers from the Sydney School share aspects of the linguistic approach taken in ESP (Bawarshi & Reiff, 2010) as well as their “strong belief in the value of teaching students about genres” (Hyon, 2018, p. 21). In contrast, RGS theorists question the degree to which explicit teaching of genres in classrooms can be effective, and at times become counter-productive (e.g., Freedman, 1993). Undaunted, researchers from the Sydney School argue that “systemic functional metalanguage (related to the tools of genre and register) can be used in planning for, reflecting on, and assessing student literacy  28  across the curriculum” (Macken-Horarik, 2002, p. 18). These researchers draw on SFL to understand the “systematic relationship between the social environment on the one hand, and the functional organization of language on the other” (Halliday & Hasan, 1985, p. 11-12) and how to explicate this knowledge through teachers to students (Martin, 2000; Martin & Rose, 2003).  Among the three “schools” of genre identified by Hyland (2003) and others, RGS researchers are more skeptical about the ability to explicitly instruct learners on the wide range of genres in operation at any given time and place, most notably because these rules are nebulous and, at best, “stabilized for now” (Miller, 1984). This is not the case for proponents of the other two approaches. As Hyon (2018) notes, “while ESP and Sydney School scholars also recognize the changeability of genres and their contexts, they tend to assume a greater degree of stability (and thus teachability) of genre characteristics” (p. 22). This awareness of genre and the accompanying metalanguage required to promote discussion, analysis, imitation, and transformation, it is argued, enables more informed choices to be made by learners by helping them make necessary connections between textual features and their social meanings within disciplinary communities (Martin, 2000; Ramanathan & Kaplan, 2000). Program curriculum informed by these approaches tends to take a more direct and disciplinary-specific approach to writing instruction, often with the belief that increasing awareness of content-specific genre features (linguistic and otherwise) provides teachers and learners with the knowledge necessary to better understand how meaning is typically expressed within various speech communities, and how learners might go about accommodating, negotiating, adapting, resisting, or seeking to change communicative practices.  2.2.3 Explicit academic writing instruction  For novice researchers and teachers approaching academic writing instruction today, the complexity and breadth of theories and approaches currently informing various educational  29  contexts can seem overwhelming. The growth of corpus-based research and sophisticated rhetorical and textual-analytical techniques continue to advance our understanding of academic and disciplinary writing. This has provided language program designers, instructors, and learners with a wealth of readily-accessible information (e.g., de Silva Joyce & Feez, 2012; Flowerdew & Costley, 2017). However, as Freedman (1993) noted more than two decades ago, teachers (many of whom are also researchers) have an irresistible “urge to explicate” what they know for the benefit of their learners; and given this proclivity, there is a continual need to interrogate the grounds “for believing such explication will in fact enhance their learning” (p. 223). Freedman argues that there is an inherent danger involved in explicit instructional approaches.6 For this reason, she offers two models (or hypotheses) regarding explicit genre instruction: 1) a “strong hypothesis” which views explicit instruction as unnecessary and ineffective, and 2) a “restricted hypothesis” in which, under certain conditions, it “may enhance learning,” but that it always introduces a “risk of overlearning or misapplication” (p. 226). However, as Kalantzis and Cope (2016) have noted, explicit and didactic pedagogical “habits” persist in many classrooms around the world (para. 7). Therefore, given the role explicit instruction may play in various educational contexts, there is a need to continually interrogate our assumptions regarding the potential benefits and disadvantages that can arise. There is still little consensus on the effectiveness and appropriateness of explicit writing instruction, when it should be utilized, and to what extent. Some scholars and researchers have questioned the degree to which explicit and targeted pedagogical approaches serve to inculcate and/or perpetuate “Western” standards, thus ignoring how the English language is being used (i.e., adapted and appropriated) by English speakers around the world (e.g., Heng Hartse & Kubota,                                                  6 For additional perspective, see Freedman’s (1993) discussion of a “concerted campaign” to teach genre features to elementary students and her description of an SFL approach to Little Red Riding Hood for second graders (p. 223).  30  2014; Horner et al., 2010). Kalantzis and Cope (2016) point out that “some members of the genre school, while agreeing in principle that it is important to be explicit about text structure and its relation to social activity,” also argue that “formalistic tendencies and their tendency to present textual recipes, take us dangerously close to a revival of a didactic pedagogy” (para. 14). However, Kalantzis and Cope argue that a pragmatic approach to functional literacy instruction can empower students rather than seek to control them if tendencies towards didactic and culturally-insensitive approaches can be minimized. I would argue that both explicit instruction and inquiry-based approaches are necessary (and not in conflict) in most English-dominant educational contexts in which multilingual learners are involved, especially when students have little exposure in similar sociocultural settings. The precise balance that should (or can) be maintained is dependent on the socioeducational context in which instruction in taking place, including the ideological, cultural, and institutional constraints surrounding and informing instruction. However, I generally prefer approaches that minimize the level of explicit instruction to that necessary to scaffold inquiry and discovery-based learning approaches. I also appreciate approaches that attempt to include use of learners’ full linguistic and cultural repertoires as much as possible.  The multifarious ways in which academic language programs have been characterized in the literature, and the wide variety of pedagogical approaches that have been adopted or adapted within them, make it challenging to develop common understandings about what might work best, when, where, with who, and why. Given the current diversity in theories, approaches, and students enrolling in these programs, the need for ethnographically-oriented research into the perceptions, experiences, and performances of multilingual writers over time is perhaps more important than ever. As Kobayashi et al. (2017) have argued, “the proliferation of academic English preparatory or  31  bridge-like programs specifically designed to orient students to the local academic discourses and practices” has created a pressing need for ethnographically-oriented research that investigates socialization practices at the “institutional level (i.e., beyond the course level, which has been the focus of much research” (p. 250).  2.3 Ethnographically-oriented academic writing research  Given the perceived importance of academic writing in making a successful transition to English-dominant universities (Flowerdew, 2016; Hyland, 2013), a considerable body of research has accumulated on this topic. The theoretical and methodological approaches taken across various fields of inquiry over the last half century have been diverse. However, several writing scholars and researchers have incorporated an ethnographic orientation into their research design (Paltridge et al., 2016). Lillis (2008) describes various “levels” of ethnography that have been undertaken in academic writing research. First, “ethnography as deep theorizing” represents the highest level since it “fundamentally challenges the ways in which text and context in writing research are often conceptualized” and emphasizes the need to develop analytic tools to overcome this limitation (p. 355). At a minimal level is “ethnography as method” or “talk around text” since the researcher moves “beyond the text, but the text remains the primary object and the analytic lens” (p. 359). However, between these two extremes, Lillis identifies “ethnography as methodology” because it enables the researcher to “explore and track the dynamic and complex situated meanings and practices that are constituted in and by academic writing” (p. 335). While I do not characterize my research as ethnography, I do incorporate an ethnographic perspective (Paltridge et al., 2016), one which could be characterized as “ethnography as methodology” as described by Lillis (2008) with a sustained presence in the research site over a prolonged period, and with attention given to both emic and etic perspectives and cultures of academic socialization.  32  For the remainder of this chapter, I focus on ethnographically-oriented studies of most relevance to the present study. First, I situate my review within three proposed “orientations” to L2 disciplinary socialization (Morita & Kobayashi, 2008). Given its importance in relation to the focus of this study, I describe the gradual development of socially-focused research studies in the field of L2 writing. I also highlight relevant L2 writing studies that have taken an ethnographic orientation (Paltridge et al., 2016). Finally, I conclude this chapter with a review of several case studies that explicitly incorporate (as mine does) an academic discourse socialization perspective (Kobayashi et al., 2017). Given the paucity of case studies that have investigated multilingual writers’ longer-term transitions into English-medium universities, I include research conducted in diverse contexts and at various educational levels. However, given the context of this study, I spotlight research conducted with undergraduate multilingual writers transitioning to English-medium universities whenever possible. 2.3.1 Three orientations to disciplinary socialization Within the wider field of applied linguistics, researchers have been investigating sociocultural aspects surrounding the teaching, learning, and performance of academic and disciplinary discourse for some time. These researchers have utilized a wide variety of theoretical and methodological approaches and have used various kinds of nomenclature to characterize their more specific areas of inquiry and favoured concepts and notions. Morita and Kobayashi (2008) have described three general orientations to “disciplinary socialization” involving multilingual learners.7 Research from the first orientation they identify, comprised largely of EAP needs analyses and genre-based research, is primarily focused on what students need to know and tends to investigate the textual features and patterns characteristic of academic discourse (p. 244). One of                                                  7 It should be noted that their review does not include studies conducted after 2004, but it provides an informative taxonomy nonetheless.  33  the primary assumptions behind much of this research is the need to develop instructional content and pedagogical methods that can better prepare multilingual students to meet the English academic and disciplinary writing demands of universities, their specific disciplines, and their associated professions. This orientation continues to produce trade books and guidebooks purporting to prepare learners for general academic and disciplinary-specific English writing expectations (e.g., de Silva Joyce & Feez, 2012; Swales & Feak, 2012).  The second orientation includes studies that focus on how students are socialized and aims to “document the socially and temporally situated process of socialization,” often from the perspectives of the students and instructors involved (Morita & Kobayashi, 2008, p. 244). Researchers with this orientation explicitly set out to investigate the mediating influence of the social and interactional context surrounding the co-construction of discourse, rather than relying primarily (or exclusively) on textual products to extrapolate the complex social and cultural aspects involved in text production. These researchers focus on numerous socioculturally-mediated aspects surrounding and informing the teaching and learning of oral, written, and multimodal academic discourse, as well as the various ways more macro sociocultural and sociohistorical contextual aspects are reflected in (and reflective of) more micro socialization practices (Duff & Talmy, 2011).  The third orientation includes studies that explicitly foreground notions of power and approach academic discourse not as “a set of neutral linguistic conventions but a value-laden, social practice that constructs and is constructed by unequal relations of power” (Morita & Kobayashi, 2008, p. 245). This includes, but is not limited to, how cultural, historical, and institutional forces combine to fundamentally shape the nature of the interactions that take place, as well as the various ways those with less power might accommodate, resist, or possibly transform social dynamics. For these researchers, there is a pressing need to acknowledge and confront issues of power surrounding  34  language in use and investigate how notions of race, gender, sexuality, and other socially-constructed notions shape discourse and activity, often in ways that marginalize or otherwise disadvantage some groups or individuals over others.   My doctoral research project fits most readily into the second orientation described above since I am primarily interested in how students are socialized (or apprenticed) as well as the shared and individuated ways they respond to this apprenticeship over time and across contexts. I incorporate multiple data sources throughout my study to investigate learners’ perspectives and experiences over time, thus further placing me squarely within this orientation. However, I am also inspired and informed by researchers approaching L2 disciplinary socialization from other angles, whether focused on understanding the social nature of academic or disciplinary texts (as co-constructed artifacts) or focused on the omnipresent institutional structures and power relations impacting academic discourse and the real-world impact this has on individuals and groups.  In the next section, I discuss the growth of ethnographically-oriented theoretical and methodological approaches in the field of second language writing. Given the sparsity of longitudinal case studies that have investigated multilingual writers transitioning to English-dominant universities, I include discussion of studies conducted at both the graduate and undergraduate level as they provide additional perspective. In the final section of this chapter, I discuss a small but growing number of case studies that have explicitly incorporated an academic discourse socialization perspective. 2.3.2 Second language writing research  In the field of second language (L2) writing, topics that have been investigated at the graduate and undergraduate student level include text-linguistic features, the composing processes of L2 writers, and the influence of learner background variables to name a few (see Leki,  35  Cumming, & Silva, 2008; Matsuda & Silva, 2005, Pelaez-Morales, 2017; Riazi, Shi, & Haggerty, 2018, and Silva & Matsuda, 2001 for reviews). Much of this research has focused on academic texts and writers apart from the social contexts surrounding them, or what Lillis (2008) describes as a “textualist-analytic lens” (p. 354). For these researchers, written text is generally conceptualized as an “inert object, complete in itself as a bearer of abstract meanings” (Bazerman & Prior, 2004, p. 1) and language acquisition is conceptualized as largely a cognitive process that occurs “in the heads” of learners (Atkinson, 2002). This body of research continues to provide valuable insights into academic and disciplinary-specific text. However, these approaches have been limited in their ability to “give insight into development as an individual and contextualized process” (Paltridge et al., 2016, p. 133).  To widen the lens, several L2 writing researchers have incorporated sociocognitive approaches into their research designs to help bridge the text-context divide (e.g., Berkenkotter & Huckin, 1995; Katznelson, Perpignan, & Rubin, 2001; Riazi, 1997). These researchers have provided valuable insight into what occurs not only in the minds of multilingual writers, but also how this is inextricably connected to the perceptions and experiences of learners within the social and cultural contexts in which they write. In addition, ethnographically-oriented research conducted by English composition and L2 writing scholars has discovered highly intertextual writing practices in professional and academic environments which are mediated by historical, social, and cultural aspects as well as the ideological predilections and power relationships of those involved (e.g., Atkinson & Ramanathan, 1995; Dias et al., 1999; Herrington, 1985; Ivanič, 1998; Leki, 2007; Leki & Carson, 1994, 1997; McCarthy, 1987; McKay, 1993; Prior, 1998; Spack, 1997; Shi & Cumming, 1995; Zamel, 1983). These researchers aim to provide more nuanced accounts of what gets written in academic and professional contexts and, equally if not more important, why. For these  36  researchers, texts are conceptualized as an artifact of interaction with others that can only be properly understood within the historical, social, and cultural contexts that motivated and guided their construction. Data collection methods might include extensive interviewing over time, participant or non-participant observation, and the collection of written artifacts co-constructed over time, amongst others.  In an early study that aimed to broaden the text-analytical lens dominating research at the time, Zamel (1983) incorporated direct observation and interviews to investigate the composing processes of six multilingual writers who voluntarily took her undergraduate composition course in an American university. She noted that, up to that time, ESL writing was largely taught as if “form proceeded content, as if composing were a matter of adopting preconceived rhetorical frameworks” (p. 167). In her study, Zamel discovered that the composing processes for both the skilled and unskilled writers were largely non-linear, idiosyncratic, and far more chaotic than process approaches had often assumed. Her participants did not have a “clear sense of direction” or an “explicit plan” throughout the composing process, but instead were “creative and generative” in their use of strategies, often aimed at “further discovery and exploration” (p. 180). Based on these results, Zamel cautioned against “assigning essays that are supposed to represent ideal rhetorical models,” and instead, emphasized the need for pedagogical approaches that can assist multilingual writers in appreciating that “decisions about form and organization only make sense with reference to the particular ideas expressed” (p. 181). Leki and Carson (1994, 1997) investigated the perspectives of undergraduate and graduate students in an American university who were studying in a sheltered EAP writing course while taking mainstream content courses. They reported that in undergraduate EAP courses, students were often not required to meaningfully engage with content but instead focus on discrete language  37  forms. The researchers lamented that, “hearing ESL students repeatedly describe writing classes as friendly but not intellectually challenging … is alarming and disheartening and calls for, we believe, deeper reflection on how we as teachers ask our students to spend their time” (Leki & Carson, 1997, p. 64). They characterized their research as “an initial step in the direction of eventually learning more about how ESL writers make the transition between ESL writing classes and other academic courses.” (p. 43). Since the ESL writing courses in their study were not explicitly connected to relevant content courses, the researchers suggested that linked (or adjunct) courses are a necessary step towards making these necessary connections. However, they also call for research into writing courses that “bill themselves as more discipline related” and the degree to which they focus on content or emphasize “strictly linguistic and rhetorical issues” as they discovered for the sheltered EAP courses in their study (p. 60). Some of the EAP courses taken by multilingual writers in the present study were linked to content courses in various ways. As a result, it provides an opportunity to investigate how disciplinary-specific and relevant students perceived them to be, both while they were taking sheltered content courses within the program, and later, while they were taking content courses in the university mainstream. Those who have investigated undergraduate multilingual learners studying in academic programs in English-dominant countries have discovered their learning trajectories to be non-linear, impacted by previous educational experiences, as well as their ability to understand and negotiate the expectations of others in a new socioacademic context (Casanave, 2002; Leki, 2007; Zamel & Spack, 2004). In a three-year study of the transitional experiences of an undergraduate student from Japan (Yuko) transitioning to an American university, Spack (1997) discovered that Yuko’s struggles in her first year of study included a perceived inability to adjust to an “American style” of writing that was incongruent with her educational experiences in Japan. However, over time Yuko  38  adjusted her reading strategies and began to focus more on general meaning rather than specific details, which helped her overcome her fear of not having the requisite linguistic and cultural background knowledge to make a successful transition. Leki (2007), in a study involving four multilingual learners transitioning to various disciplinary programs in an American university, also discovered that her participants’ long-term trajectories were non-linear and unpredictable, influenced by a lack of intellectual and disciplinary engagement in their language courses as well as the impact of language ideologies and power relationships in the socioacademic context.  The importance of engagement was also highlighted in Tardy’s (2005) two-year case study involving two multilingual graduate students studying in an American university. The author reports on the struggles and successes these students experienced while making their transition, from focusing largely on the transmission of content in their written assignments (i.e., demonstrating content knowledge), to persuading others of the value, significance, and credibility of their writing (i.e., demonstrating rhetorical knowledge). Despite their advanced linguistic abilities, academic literacy development was slow and highly dependent on: 1) positive mentoring and collaboration experiences, 2) conceptions of identity that affirmed and acknowledged their linguistic and cultural backgrounds, and, 3) multilingual learners’ willingness and ability to meaningfully participate in the academic discourse community.  L2 writing researchers have continued to investigate the social and cultural contexts surrounding and informing multilingual writers, those who instruct them, and the texts they co-produce (e.g., Morton, Storch, & Thompson, 2015; Pomerantz & Kearney, 2012; Shi, 2010, 2012). These researchers have recognized, as Atkinson (2003) notes, that there is a need for conceptualizations of L2 writing that consider the “full range of social and cultural contexts impacting L2 writing, rather than focusing narrowly on skills and processes of writing (in the  39  classroom) in themselves” (p. 29). This growing body of research has highlighted many of the complexities involved in teaching, learning, and assessing academic writing within situated educational contexts. These researchers have discovered, amongst other findings, that what multilingual learners write in English-dominant educational contexts is fundamentally shaped by course-specific factors, the educational backgrounds and ideological beliefs of those involved, as well as historical, social, cultural, and linguistic aspects shaping the activities that take place (Zamel & Spack, 2004). These L2 writing researchers have joined several others in pursuit of closer observation of the social contexts surrounding the teaching and learning of academic and disciplinary discourse. By incorporating ethnographic perspectives and methods, these researchers aim for more sustained engagement with multilingual learners in the various contexts in which they write, as well as how socially-constructed notions of language, culture, power, race, gender, and identity (among others) shape the interactions that (can) take place. The theoretical and methodological choices made by these researchers also recognize the vital importance of situating academic texts within the social, cultural, and historical context. Defined in this way, this body of research has provided valuable insights into what is believed, thought, and done by teachers and learners in the co-construction of disciplinary discourse, or what some researchers have referred to as “L2 disciplinary socialization” (Morita & Kobayashi, 2008).  In the next section, I discuss several L2 composition studies that have taken a socialization perspective in their research of graduate and undergraduate-level multilingual learners. This overview is intended to provide a sense of the diversity and unpredictability of socialization practices and processes in situated contexts and how various socialization agents mediate multilingual learners’ perceptions and experiences over time. Given my research focus is the  40  socialization of written academic discourse, I spotlight studies that have investigated the teaching and learning of academic writing in English-medium higher education contexts, although unfortunately these studies remain scarce (Duff & Anderson, 2015). For this reason, I discuss studies conducted in graduate and undergraduate-level contexts. 2.3.3 Academic discourse socialization research Theoretically and methodologically, I have been informed by academic discourse socialization scholars and researchers, an emerging area of inquiry that is based on language socialization (LS) theory. The disciplinary lineage of LS theory traces back to linguistic anthropology and interactional sociolinguistics and has been characterized as both a theoretical and methodological approach (Bayley & Schecter, 2003; Duff & May, 2017; Duff & Talmy, 2011; Duranti, Ochs, & Schieffelin, 2012; Zuengler & Cole, 2005). According to Ochs (1986), LS involves socialization through language and socialization to use language wherein "children and other novices in society acquire tacit knowledge of principles of social order and systems of belief” (p. 2). LS researchers investigate socialization to better understand how broader forms of knowledge and associated social practices instantiate and reflect notions of culture, ideology, power, and identity, amongst others. Studies of language socialization are often longitudinal and utilize a variety of ethnographic methods to investigate the complex cognitive, contextual, and sociocultural aspects involved in the socialization (or apprenticeship) practices of relative newcomers and more experienced others.  LS researchers envision language and literacy development as “culturally situated, as mediated, and as replete with social, cultural, and political meanings in addition to propositional or ideational meanings carried or indexed by various linguistic, textual, and paralinguistic forms” (Duff, 2010, p. 172). Much of the early focus in LS research was on how young children acquire  41  the sociolinguistic competence to perform in ways deemed appropriate by caregivers and other adults, with much focus placed on interactional routines or speech acts (Duff, 2007b, p. 310). However, this research has expanded over time to include a wide range of language socialization contexts that learners experience throughout their lifespan (Duff & May, 2017).  An academic discourse socialization (ADS) perspective conceptualizes language and literacies development as inherently and inextricably a social, cognitive, and cultural activity. The socialization of academic discourse is conceptualized as multidirectional, continually negotiated, and a potential site of struggle, particularly for those with less experience in the sociocultural and sociolinguistic context (i.e., newcomers or novices). ADS researchers investigate the capability of multilingual learners (influenced by their own and others’ beliefs and practices) to participate in disciplinary-appropriate ways, including the ability to develop “one’s voice, identity, and agency in a new language/culture” (Duff, 2007a, p. 4). ADS is compatible with other sociocultural approaches to language and literacy development such as Activity Theory and academic literacies, among others. ADS researchers have reported on several challenges that multilingual learners experience in making transitions to English-medium higher education contexts, including those related to “intertextuality, unfamiliar or evolving academic genres, and social stratification and marginalization, which may be exacerbated by students’ proficiency in the language of education.” (Kobayashi et al., 2017, p. 239).  However, some New Literacy scholars have claimed that academic literacies “encapsulates the academic socialisation model, building on the insights developed there as well as the study skills view” (Lea & Street, 1998, p. 158). These scholars argue that the socialization model “presumes that the disciplinary discourses and genres are relatively stable” and that academic socialization simply involves learning the “ground rules” of discourse so that multilingual learners  42  will be able to “reproduce it unproblematically” (Lea & Street, 2006, p. 227). However, as Kulick and Schieffelin (2004) have noted in relation to LS, studies that restrict investigation to the “acquisition of normatively sanctioned practices” will be susceptible to the accusation that they are “merely behaviorism in new clothes” (p. 355). To avoid this charge, LS/ADS research should include analysis of practices where socialization does not occur or occurs in ways not anticipated, especially when there may be potentially negative consequences for multilingual learners. In my view, ADS and academic literacies approaches are commensurate since they both focus on “macro and micro contexts for language development, forms of knowledge and practice valued, material products or tools involved in literacy, and outcomes” (Duff, 2010, p. 171). Power, authority, identity, institutional structures, and meaning-making practices are of central concern in an academic literacies model (Lea & Street, 2006, p. 227). Yet, these and other sociocultural aspects are also integral to an LS/ADS perspective that conceptualizes language not as prescriptive or deterministic, but as an “innovative, transformative, and sometimes contested process” (Kobayashi et al., 2017, p. 239).   ADS researchers have focused on a range of activities common in undergraduate content and academic language courses, whether in sheltered or mainstream environments. These include, but are not limited to, delivering academic presentations (e.g., Duff & Kobayashi, 2010; Kobayashi, 2004; Morita, 2000; Yang, 2010; Zappa-Hollman, 2007b), interacting with instructors and peers (e.g., Morita, 2009; Poole, 1992; Seloni, 2008, 2012), developing social or individual networks of practice (e.g. Zappa-Hollman, 2007a; Zappa-Hollman & Duff, 2015), interacting on discussion boards and in other computer-mediated environments (e.g., Liew & Ball, 2010; Yim, 2011), and the perceptions and experiences of multilingual writers learning to perform written academic discourse (e.g., Bronson, 2004; Godfrey, 2015; Séror, 2008). While several ADS researchers have explicitly  43  drawn on language socialization (LS) theory (e.g., Bayley & Schecter, 2003; Duff & May, 2017; Zeungler & Cole, 2005), others draw on commensurate theories, notions, and approaches (e.g., Bakhtin, 1981, 1986; Bourdieu, 1977, 1991; Engeström, 2014; Lave & Wenger, 1991; Vygotsky, 1978) and incorporate similar ethnographic approaches and methods to study socialization practices and processes over time (Kobayashi et al., 2017).   At the graduate level, ADS researchers have reported on several challenges and accomplishments of multilingual writers (e.g., Anderson, 2016; Bronson, 2004; Fujioka, 2014; Seloni, 2008, 2012; Waterstone, 2008). These researchers have discovered that the pedagogical beliefs and practices of mentors, advisors, and instructors are highly influential, but also that multilingual learners draw on a repertoire of literacies as they negotiate and often overcome the challenges that confront them. In addition, how feedback and other types of academic support is conceptualized and delivered creates positive and negative socialization experiences which serve to enable and constrain learners’ sense of agency and the development of preferred identities. For example, Anderson (2016) discovered how “resilient, grounded, and exceedingly talented” several of his graduate-level participants were in their successful transition to a Canadian university “in the face of considerable adversity” (p. ii). Five of six participants, described as accomplished young scholars, were appreciative of the written feedback they received, even feedback that questioned their writing abilities. In some cases, the academic support led to negative socialization experiences (e.g., confusing TA roles) or resulted in missed opportunities. Nevertheless, most participants were able to successfully negotiate these challenges and claim or assert their preferred academic and disciplinary identities.  Studies have also discovered that multilingual learners experienced positive benefits when they were able to communicate meaningfully about their social, cultural, and linguistic challenges,  44  whether with peers or more experienced others. For example, Bronson (2004) reported that graduate student participants’ academic and disciplinary progress was enhanced when their written tasks “challenged” and “invited” them to express themselves while also providing the necessary “social and moral support” (p. ix). However, progress was hindered when there was a perceived misalignment between students’ expectations and the instruction or feedback being provided. Seloni (2008, 2012) reported that graduate student participants collaboratively constructed meaning in and outside of classrooms, which included diverse interactants such as family members, friends, peers, mentors, advisors, and others. For the graduate students in Seloni’s studies, socialization of written academic discourse was described as a multilayered process, one that was enhanced when students could discuss their challenges with others in “safe houses” (Pratt, 1991; Canagarajah, 1997), which encouraged them to “collectively resist and question the academic literacy practices that they are exposed to within institutional academic spaces” (Seloni, 2012, p. 54). In the studies involving graduate-level learners described above, most of the participants had lived in English-dominant countries and/or attended English-medium universities for one year or more prior to commencing graduate-level studies. This may have enhanced their abilities and willingness to negotiate, resist, or satisfy expectations and, in turn, assume or assert their preferred academic identities. However, undergraduate multilingual students transitioning into English-medium universities, particularly those who attended schools in which English was not the dominant language of instruction, are more likely to have limited background knowledge or experience to draw on and, as a result, more difficulty in negotiating expectations, exercising agency, and developing positive conceptions of themselves as academic writers. The remainder of this chapter will focus on some of the main undergraduate-level challenges identified in ADS  45  studies and how socialization practices within programs and courses can enable or constrain multilingual learners’ transitional experiences. Some ADS studies have reported on the difficulties undergraduate-level multilingual writers have in understanding assignment instructions, the expectations of their instructors, and how they are being assessed (e.g., Godfrey, 2015; Séror, 2008; Zappa-Hollman, 2007a). These misunderstandings can lead to significant confusion, frustration, and in extreme cases, the perception of writing as “torture” (Zappa-Hollman, 2007a, p. 117). Often, these challenges are perceived by learners to be influenced by their lack of familiarity with new socioculturally-influenced expectations and practices, some of which may be at odds with their previous educational experiences, preferences, or beliefs. In addition, researchers have discovered that the degree of engagement with written tasks assigned in academic language courses was often connected to how authentic or relevant students regarded them in relation to their disciplinary area of study. While not drawing on ADS explicitly, Lui and Tannacito (2013) incorporated Bourdieu’s (1991) notion of social capital and Norton’s (2001) discussion of imagined communities to investigate the perceptions of two Taiwanese exchange students and their responses to academic writing instruction in their English language courses. Based on pre-conceived beliefs about “Western” styles of writing, instructional practices in these courses were characterized by the students as too formulaic and not “American ways of writing” (Lui & Tannacito, 2013, p. 262). Instead, the style of instruction was perceived as more consistent with a “cram school” style of instruction experienced while in high school in Taiwan.  Several studies also report that multilingual learners struggled to keep up with and comprehend content because of the amount of required reading as well as the cultural and linguistic complexity of the text (e.g., Séror, 2008; Zappa-Hollman, 2007a). These challenges often interfered  46  with students’ abilities to incorporate content into their written assignments in ways that demonstrated comprehension and critical engagement. However, Godfrey (2015) reports on how five multilingual writers valued, and responded positively, to the efforts of their instructor in a first-year writing course to scaffold reading their strategies, develop their critical awareness, and provide guidance in incorporating reading content into written assignments. These results align with several L2 writing studies that have reported how multilingual writers struggle to borrow text effectively from what they read and incorporate this information appropriately in their written assignments (e.g., Keck, 2006; Shi, 2011, 2012; Yi & Casanave, 2012).  Several ADS studies have also discovered that students’ perceptions and reactions to the feedback they receive on their written assignments are highly variable and dependent on several factors, including the degree to which commentary focuses on form over content, institutional constraints impacting how feedback is delivered, and how intelligible and actionable feedback is perceived by students (Godfrey, 2015; Séror, 2008). For example, Séror (2008) reported on some of the negative reactions students had to feedback that “regularly focused on [their] language difficulties” because it frustrated their “desire for a fuller response from professors to their ideas and the content of their papers” (p. 125). Similarly, Zappa-Hollman (2007a) reported that when her participants received feedback that identified their language skills as deficient, it left a “strong imprint” on them, which highlighted the inherent power it held to “position learners as deficient due to limited mastery of English” on the one hand, or more productively motivate and boost their confidence on the other (p. 166). These results align with research conducted with graduate-level multilingual learners as well (e.g., Anderson, 2016; Bronson, 2004; Waterstone, 2008).     Many ADS studies have been conducted for comparatively short durations: one course, often for no longer than one or two semesters. ADS researchers may experience challenges gaining  47  the necessary access to institutions, classrooms, and the various socialization agents multilingual learners interact with over time (Kobayashi et al., 2017). However, across fields of study, some researchers have investigated the longer-term transitions of multilingual learners over multiple years of study in one program (e.g., Leki, 2007); across different programs or levels of study (e.g., Harklau, 2000), or across university contexts in different countries (e.g., Zappa-Hollman, 2007a). These studies have reinforced the importance and potential benefits of taking a longer view of writing development. As Bazerman et al. (2017) point out, “high-stakes decisions about curriculum, instruction, and assessment are often made in unsystematic ways that may fail to support the development they are intended to facilitate” (p. 353). To discourage short-term misconceptions, the authors established eight working principles that aim to provide a fuller account of the lifespan of writing development. These principles are reproduced below: 1. Writing can develop across the lifespan as part of changing contexts 2. Writing development is complex because writing is complex 3. Writing development is variable; there is no single path and no single endpoint 4. Writers develop in relation to the changing social needs, opportunities, resources, and technologies of their time and place 5. The development of writing depends on the development, redirection, and specialized reconfiguring of general functions, processes, and tools 6. Writing and other forms of development have reciprocal and mutually supporting relationships 7. To understand how writing develops across the lifespan, educators need to recognize the different ways language resources can be used to present meaning in written text 8. Curriculum plays a significant formative role in writing development (pp. 354-357)  48   The present study joins these researchers and others in seeking a longer-term perspective on multilingual learners’ written academic discourse socialization, which is needed to deepen our knowledge of how (or if) the challenges identified in several studies are eventually overcome, how perceptions about written academic discourse might change, how agency is enabled or constrained and for what reasons, and how academic and disciplinary identities are expressed, ascribed, assumed, resisted, or possibly eschewed altogether. This study also makes a contribution to a growing body of ADS literature by focusing on multilingual learners’ written academic discourse socialization. Given the importance of academic writing in language programs, and its perceived importance for gaining any level of success in undergraduate studies (Flowerdew, 2016; Hyland, 2013), a deeper understanding of these socialization practices and processes is needed. Yet, as Duff and Anderson (2015) have noted, there is “little research [that] has examined classroom socialization into such academic writing practices” (pp. 345-346). This study also contributes by investigating multilingual learners’ performances during and after completion of academic language and content courses and across classroom, program, and institutional contexts. As Kobayashi et al. (2017) have observed, such “longitudinal studies of ADS across learners’ academic programs (i.e., within and across courses) over an extended period are needed” (p. 2). The next chapter provides additional background and rationale for the theoretical and methodological approach taken in this study.     49  Chapter 3: Methods In this section, I describe in more detail the multiple case-study approach undertaken, the primary research context, the participants who contributed to this study, and the data collection and analytical techniques incorporated.  3.1 Multiple case study This is a multiple case study (Duff, 2008; Yin, 2009) utilizing qualitative ethnographic methods. Qualitative case study methodology, amongst numerous other affordances, enables researchers to “provide an understanding of individuals’ experiences, issues, insights, developmental pathways, or performance within a particular linguistic, social, or educational context” (Duff, 2014, p. 233). As discussed in Chapter 2, researchers incorporating an “ethnographic orientation” into their design aim to “find out what actions and events mean to the people being observed” (McKay, 2006, p. 79). This approach enabled me to investigate an academic practice of perceived importance (English academic writing) while remaining as close as possible to “real-life situations and its multiple wealth of details” (Flyvbjerg, 2006, p. 223). It is not the aim of this study to generalize findings to other contexts, but instead maximize the trustworthiness and reliability of the results by describing data collection and my interpretive analysis in detail. In collecting interview and written discourse over time, supplemented with observations of contexts and discussions with multiple stakeholders, I aim to enhance the depth of description and triangulation of my interpretations (Duff, 2008, p. 44). In collecting and analyzing data in a qualitative manner, I aim to maximize its transferability (Lincoln & Guba, 1985), which assigns responsibility to the reader to determine “congruence, fit, or connection between one study context, in all its richness, and their own context, rather than have the original researchers make that assumption for them” (Duff, 2008, p. 51).   50  As noted above, this dissertation is written in manuscript format (Paltridge, 2002; Paltridge & Starfield, 2007) and is composed of three separate but interrelated studies (the manuscript chapters). These three interconnected studies are presented in Chapters 4, 5, and 6. Table 3.1 lists the overarching research question, focused research questions, focal participants, and data collected for each manuscript chapter.  Table 3.1: Research questions, focal participants, and data collection Overarching research questions  Focused research questions Focal Participants Data  collection  Chapter 4: How do program design features shape participants’ reported beliefs about the program and their experiences within it?  a) How are program objectives and design features characterized in documents and reported on by non-student participants (i.e., consultant, administrators, and instructors)? b) How do student participants characterize the program and describe its influence on their transitions to mainstream university studies?   6 students • Chloe, Jasmine, and Yvonne (Arts) • Ashlee, Dawn, and Theo (Sciences)   2 administrators 2 instructors 1 consultant   Primary: • Interviews (39) • Program documents   Secondary: • Observations of        22 classrooms and   1 two-day event • Research notes  Chapter 5:  How do academic language courses mediate students’ perceptions and performances of written academic discourse?   a) How do four successful students characterize their experiences learning and performing academic writing in their language courses?  b) In what ways do participants report understanding and applying written academic discourse concepts from these courses in their written assignments?   4 students: • Chloe (Arts) • Yvonne (Arts) • Dawn (Sciences) • Theo (Sciences)     Primary: • Interviews (22) • Assignments (83)  Secondary: • Observations of      16 classrooms and     1 two-day event • Curriculum documents  • Research notes   51   Chapter 6: What are some of the struggles experienced by multilingual writers as they transition through this program?  a) How do two less successful students characterize their experiences learning and performing academic writing in their language courses? b) How do social, cultural, or linguistic challenges mediate their transitions to their respective programs of study?  2 students:  • Ashlee (Sciences) • Jasmine (Arts)  Primary: • Interviews (12) • Assignments (64)  Secondary: • Observations of      10 classrooms and    1 two-day event • Curriculum documents • Research notes  In Chapter 4, all six student participants were included in the study and the additional perspectives of other program stakeholders were included in the analysis. In Chapter 5, four “successful” student participants from the original six were selected for a focused analysis. These students were able to satisfy all OC program requirements, entered the second year in their chosen program of study within MSU, and maintained an overall positive perception of the OC program. In Chapter 6, two “less successful” student participants from the original six were selected for a focused analysis. These students reported considerable confusion, frustration, or conflict (in comparison to other participants) in satisfying program requirements and transitioning to the university mainstream. One student (Jasmine) failed to meet program requirements, attended another university-affiliated college for another year, and did not enter the mainstream until the subsequent year. The other student (Ashlee) satisfied program requirements but reported significant disagreement, confusion, and frustration in adjusting to instructional practices and disciplinary writing expectations, especially in one of her academic language courses. 3.2 Research context The primary context for this study, as described in Chapter 1, was a newly-designed academic language program, given the fictional name Oasis College (OC) for the purposes of this  52  study. It was physically located in a large publicly-funded university in Canada, which has been given the fictional name Main Stream University (MSU). OC was designed for international students unable to meet direct-entry English language proficiency requirements for MSU. Typically, Canadian universities rely on international large-scale standardized language tests such as the TOEFL iBT (https://www.ets.org/toefl/ibt/about) and IELTS Academic (https://www.ielts.org/what-is-ielts/ielts-for-study) amongst others. Such English language proficiency measures are commonly required by English-medium universities as they provide indicators of students’ English academic language abilities. Despite their limitations, universities and faculties establish yearly cut-off scores for these tests that apply to all prospective international students who have not studied in a recognized English-medium educational context for a specified amount of time. To enter MSU directly, undergraduate students must have completed at least three years of study at a recognized institution. For those unable to meet these requirements, there are often alternative programs available (e.g., bridging or pathway programs).  When student participants in this study applied for admission to the OC program, they were required to submit their scores on a recognized English language proficiency test. All participants had completed either the TOEFL iBT or IELTS Academic at least once in the year before applying for admission to this program. Table 3.2 displays the language proficiency scores required for direct entry into MSU and those required for OC. It should be noted that these are the minimum scores required. Students would need to score higher on some of the subsections to achieve the overall minimum score.  Most of the student participants in this study could not confidently remember all their language proficiency test scores when asked in the initial interview. Some students were more confident about their overall result but could not remember scores for each section. Some  53  Table 3.2: Minimum English language admission standards for MSU and OC Test Institution Speaking Listening Reading Writing Overall TOEFL   iBT MSU 21 22 22 21 90 OC 16 16 16 16 70 IELTS Academic MSU 6.0 6.0 6.0 6.0 6.5 OC 5.0 5.0 5.5 5.5 5.5  participants appeared to be uncomfortable discussing their test scores. As this was not a major focus of the study, I did not ask them to submit their official scores. Also, to respect their privacy, I did not request access to their official files without their permission. For these reasons, I decided to exclude any self-reported scores as they were incomplete and there was no way to confirm them. However, it is safe to conclude that all six students had achieved the minimum scores required by OC but were not able to achieve all the minimum scores required by MSU. OC offered credit-bearing first-year undergraduate courses in two main disciplinary areas, called “Streams.” In the Arts Stream, content courses were offered in psychology, geography, and political science. In the Sciences Stream, content courses were offered in math, physics, and chemistry. Arts students took one elective course outside OC (i.e., in MSU) in the third term of their first year. Science students took an additional course administered within OC in either physical or computational sciences, depending on their desired specializations.  In both disciplinary Streams, students took concurrent academic language courses (and interdisciplinary seminars) that focused on a wide range of academic skills and abilities, including designing and carrying out research, delivering academic presentations, and writing research reports, amongst several others. There were three academic language courses in the program that  54  included a substantial writing component in their design and delivery. While not their only educational objective, each course included disciplinary-specific instruction and practice in academic writing informed by, amongst others, genre theories, content and language integrated approaches (e.g., CLIL), and/or functional approaches to language (e.g., SFL). However, these courses were stand-alone and varied in their purpose and the extent to which they drew on these theories and approaches. For example, one of these courses was connected to disciplinary content areas in various ways (e.g., incorporation of content curriculum for language analysis purposes). This course was also divided into separate sections, each incorporating course material from a different content area. These smaller tutorials provided an opportunity for more targeted discussion of the academic language being used in their course readings. This course also included direct instruction and controlled practice using language analysis tools informed by SFL. In this way, the course drew more heavily on CLIL and SFL than the other two academic language courses, which were not directly linked to specific content courses and drew on a range of theoretical and pedagogical approaches (genre-based pedagogies, SFL, task-based learning, etc.). All students in this study took most or all of their first-year courses with other OC students within their respective disciplinary Streams. In the third and final term of the program (in the summer), students participated in a two-day interdisciplinary conference in which they shared the results of their independent research projects. If all program requirements were successfully completed, students qualified for entry into their second year of study in MSU.  3.3 Participants 3.3.1 Student participants The choice of how and when to recruit participants for this study was greatly impacted by access to the research site. Given this was a newly designed program in its inaugural year, I was  55  advised by multiple stakeholders to be sensitive to the challenges being faced by program designers, administrators, and instructors at the start of the program. It was also important to provide adequate time for newly-arrived students to acclimatize to their new surroundings and become familiar with program and course requirements. As a result, I did not recruit students at the start of the program and instead met with administrators to discuss aspects of the program and possibilities for recruitment. I also studied any program and curriculum documents made available to me and investigated publicly-available documents regarding the program (e.g., program brochures, institutional websites, media articles, etc.). During this time, I interviewed an MSU faculty member who was an early consultant for the program (but who was no longer consulting) and two lead administrators for the program who were also MSU faculty members.  After the first month of the program, I was given permission to attend faculty meetings in both Streams to briefly explain my study and request permission to attend classes to introduce the study and distribute a recruitment letter (see Appendix A). In the Sciences Stream, I was permitted to attend one class in which all students were enrolled and was given some time at the start of the lecture to discuss my study with students. In the Arts Stream, I was permitted to recruit students through email only. When students expressed an interest in participating in the study, an information letter was sent to them that included additional details (see Appendix B). After any questions or concerns were addressed, students were asked to sign an informed consent form (see Appendix C).  Based on recruitment efforts over the first term, a total of eight students agreed to participate in this longitudinal study. However, three students who initially signed informed consent letters dropped out of the study early in their second term because they became overwhelmed with their workload. This left a total of five students remaining in the study. Chloe and Yvonne were  56  studying in the Arts Stream and Ashlee, Dawn, and Theo in the Sciences (all names are pseudonyms). However, during the second term, an Arts student (Jasmine) approached me after a classroom observation and asked me if she could join the study. She said she was having considerable difficulty in her courses, particularly with writing, and would appreciate the opportunity to talk more about it. With Jasmine, a total of six students maintained their participation in this study into their mainstream programs of study. Because of their willingness to do so, I was able to stay in touch with Chloe, Jasmine, and Theo in their third and fourth year of study at MSU. Ethics approval was extended on a yearly basis for this purpose. We met informally on several occasions, most often in coffee shops on campus. These discussions were not recorded but I took research notes. Table 3.3 lists some of the background information for the student participants in this study.  Table 3.3: Student participants’ background information Pseudonym Sex/Age  Stream Born Previous education  Chloe F, 18 Arts South Korea Completed elementary and 4 ½ years secondary in South Korea; attended two public high schools that prepared for the national university-entrance test (Suneung); relocated to Canada and completed 1 ½ years of high school in the same province as MSU.   Jasmine F, 19 Arts China Completed elementary and secondary school in China; attended an international high school and took classes from Chinese and Canadian teachers; took extensive language test preparation courses outside of high school.  Yvonne F, 19 Arts Russia Completed elementary and secondary school as a “foreigner” in China; attended an international school; prepared for a foreign version of the university-entrance test (Gaokao) described as “much easier” than the domestic test.   Dawn F, 18 Sciences China Completed elementary and secondary school in a state-funded school in China; attended what she described as a “normal” and “strict” public high school with “so much homework, much more than [OC].”  57  Ashlee F, 19 Sciences China Completed elementary and secondary school in China; attended an international high school that incorporated an “international curriculum” based on the “British system.”  Theo M,19 Sciences China Completed elementary and secondary school in China; attended an international high school with many Canadian teachers and curriculum from the same province as MSU; courses were the “kind of courses people here study.”  It should be noted that most of the student participants in this study had attended an international high school in their home country before arriving at OC. In addition, two participants (Chloe and Theo) had direct experience with Canadian-designed curriculum. Ashlee, Dawn, Jasmine, and Theo were born in China and had completed their elementary and secondary education entirely in their country of birth. Yvonne was born in Russia but moved to China when she was five years old and Chloe was born in South Korea. Only Chloe and Dawn had attended public (or state-funded) high schools that followed a standardized curriculum designed (amongst other objectives) to prepare learners for domestic university-entrance tests (e.g., the Gaokao in China and the Suneung in South Korea). Chloe and Dawn described their experience in these environments as highly competitive and stressful. They described most of their classes in these schools as focused on preparing learners to perform well on these national tests, including their highly-structured English components.  Ashlee, Theo, Jasmine, and Yvonne attended international high schools in China and were taught by both Chinese and non-Chinese instructors. They reported that the curriculum in their schools was not designed to prepare them for the Gaokao. Ashlee described her high school as following an “international curriculum” which was based on the “British system” (Ashlee, 1st interview, term 1). Theo reported several of his teachers were Canadian and that the curriculum was modelled on one used in the same Canadian province as MSU (Theo, 1st interview, term 1).  58  Jasmine describe the curriculum in her high school as “specialized” and “innovative” and designed to prepare learners for university study abroad. Jasmine also reported taking intensive language test preparation courses outside of her high school. Yvonne, as a “foreigner” in China, did not need to prepare for the same college matriculation examination as those born in the country. Instead, she prepared for a foreign version of this test, one she described as “much easier than the one Chinese students do” (Yvonne, 1st interview, term 1). She also reported that the curriculum in her high school included less explicit English language instruction and more collaborative activities than public schools in China. Chloe was the only participant who attended multiple high schools (three) and did so across two countries (South Korea and Canada). After having difficulty adjusting to two different high schools in South Korea that focused on preparation for the Suneung, her parents permitted her to continue her studies abroad, something she had been wanting to do for some time. Accompanied by her mother, she completed a year and a half as an international (fee-paying) student in a Canadian high school in the same province as MSU.  3.3.2 Non-student participants Non-student participants in this study include an MSU faculty member who was an early consultant in the design of the program, two administrators of the program who were also MSU faculty members, and two OC instructors who agreed to both an interview and multiple classroom observations. To protect the identities of these participants, further details about them are omitted. They are also not given pseudonyms nor are any gender pronouns used when describing their contributions to this study. Nevertheless, these stakeholders were essential in providing access to observable spaces and in helping me better understand various aspects of the program. While I formally interviewed these participants only once, I had several informal conversations which were described and reflected on in my research notes. There were several other non-focal participants  59  who contributed to the study in various ways, including six instructors and one TA delivering OC courses and four instructors and one TA delivering MSU courses. These non-focal participants, due to various constraints, were unable to conduct a formal (recorded) interview but did agree to unrecorded classrooms observations.  3.4 Ethical considerations During the design and recruitment phase of the study, there were several ethical issues to consider. One important decision was what to offer students in exchange for their participation. In other studies, researchers have offered academic writing support (e.g., Anderson, 2016; Bronson, 2004) and I did consider this option since learners would likely be highly receptive to it. However, in the end I decided against this for three reasons: 1) students would be getting extensive academic support in the OC program and could take advantage of MSU services as well, 2) my advice might be at odds with curriculum objectives and/or instructors’ expectations, and 3) I hoped to develop engaging and reciprocal conversations with participants and felt that providing academic writing support might set up an instructor-student dynamic that could stifle this open dialogue. Students were instead offered a $20 MSU bookstore gift card for each term of their participation.     While interacting with participants, I needed to by mindful of my presence in the classroom and other educational events. I regularly reflected on my role in shaping interviews and discussions with students. As a white Anglo male in my mid-40s, I was communicating with multilingual learners more than half my age who were living away from home for the first time. Therefore, I needed to remain mindful of several potential issues and take pre-emptive steps to minimize any conflicts or misinterpretations. First, all recorded interviews with students were conducted during the day on university grounds in study rooms located in the learning commons. Informal discussions were conducted in public locations (e.g., campus coffee shops). Throughout the study, I  60  made a conscious effort not to express myself in an authoritative or absolute manner and to maintain respect for the thoughts, experiences, and opinions of my participants at all times. However, I also expressed my opinions on various topics and issues to encourage more engaged discussion. In interviews, I sought as meaningful and honest an exchange of information as possible, but I did not want to cause any additional stress or anxiety for students. This was challenging at times when students discussed their difficulties, frustrations, and conflicts. I regularly ensured that students knew who they could talk to about these issues by providing them with a list of services available to them with contact information (for OC and MSU). However, most participants willingly discussed many of their personal issues and reported that they enjoyed having a chance to talk about them. At no point were any participants pressed to discuss issues that were not comfortable for them to discuss.  While observing classrooms and other educational events, it was very important for me to guard against any negative impact my presence might have on participants or others. First and foremost, I needed to ensure the identity of student participants be kept confidential. This was stipulated in the ethics application and assured to participants on their consent form to minimize risk to them (e.g., being singled out by instructors or peers and treated differently) as well as to promote trust that what they shared with me would not be discussed with administrators or their instructors. I therefore avoided interaction with students immediately before, during, and after classroom observations. When interacting with the course instructor, I did not reveal the identity of student participants. I also needed to be aware of how I might be perceived by others while observing any class or event. For example, I could be viewed as a teacher, an administrator, or an evaluator working with the program. While this perception could not be completely controlled, it was important for me to be aware of this potential and clearly explain my position within the  61  university structure and my purpose for being in the class. Depending on some instructors’ invitations to do so, I did interact in some classes more than others. However, I tried to keep these interactions as brief and as non-intrusive as possible. Early in the data collection phase of my research, I was asked to assist with internal research being conducted by program stakeholders, which would involve the analysis of survey and interview data. Since I would not become directly visible to the student population in this role, and it would enable me to better understand various aspects of the program, I felt this would not unduly interfere with my dissertation study. In consultation with my committee, I took on this role which I believe enriched my perspective and helped me to better understand and discuss some of the contextual issues with student participants. In this role, which lasted one year, I transcribed interviews for participants (not involved in my study), coded and analyzed data, assisted in the preparation of internal reports, attended meetings, and participated in a presentation at a local conference. None of the work I performed in this role for the program is included in the present study. 3.5 Data collection and analysis In the following subsections, I discuss in more detail the data collection and analytical methods chosen for this study which include: a) interviewing participants, b) collecting assignments and documents, c) observing classrooms and other educational events, and d) thematic analysis.  3.5.1 Interviewing participants Interviews were scheduled with student participants as close as possible to the start of their program of study. Subsequent interviews were scheduled after the end of each term, and preferably after all assignments were completed, submitted, and graded. Due to scheduling conflicts, this was not always possible. In ongoing semi-structured interviews with students, we explored various  62  topics guided by a list of probing questions (see Appendix D). In total, I conducted 34 interviews with student participants (representing 40+ hrs of discourse) over the course of five terms. Table 3.4 lists the total number of interviews conducted with each participant as well as the term and the month they took place over the course of the study (terms 1-5). There were three comparable semesters (or terms) at MSU and OC: Term 1 (Sept. to Dec.), Term 2 (Jan.-Apr.), and Term 3 (May-August). Table 3.4: Interviews conducted with student participants     First Year Second Year Name Stream Total Term 1 Sept-Dec Term 2 Jan-Apr Term 3 May-Aug Term 4 Sept-Dec Term 5 Jan-Apr Chloe Arts 5 Nov Feb May, July n/a Feb Jasmine Arts 6 n/a Mar May, July Nov Mar, May Yvonne Arts 6 Nov Feb May, July Nov Apr Ashlee Sci 6 Nov Jan, Mar May, July Nov n/a Dawn Sci 6 Nov Feb, Mar May, July n/a Feb Theo Sci 5 n/a Jan May, July Nov Feb Total 34 4 8 12 4 6  Three of the six student participants (Chloe, Yvonne, and Theo) expressed a desire to stay in touch beyond the original time frame established for this study (16 months) and ethics clearance was extended on a yearly basis for this purpose. We continued to communicate by email and occasionally arranged to meet on campus for informal discussions of their experiences in the university mainstream. While only one participant expressed a preference not to have these additional discussions recorded, I decided not to record these discussions and instead take notes  63  during and after these meetings. This was based on my desire to encourage more open, organic, and collegial conversations that were not guided by a pre-established list of interview questions.  I also interviewed several non-student participants at various points throughout the study including a consultant before the program commenced, two administrators early in the first term of the program, and two instructors at the end of the second term. These interviews were semi-structured in that they were guided by a list of pre-determined questions (see Appendix E), but interviewees were encouraged to elaborate on any topics that came up in the discussion. All interviews were recorded and transcribed for qualitative analysis, a process which focused on reproducing the words spoken by participants and by me as accurately as possible while maximizing readability. Therefore, I omitted some discourse features (e.g., false starts, hesitations, reformulations, etc.) and corrected minor grammatical issues to improve readability (e.g., subject-verb agreement, tense, articles, etc.). I used square brackets around text to indicate that the words spoken had been replaced by me to maintain confidentiality (e.g., removing identifying information).  Participant interviews are one of the most commonly utilized qualitative research techniques (Roulston, 2010), so much so that some have warned of the hidden assumptions inherent in our “interview society” (Atkinson & Silverman, 1997). For example, there has been a tendency for researchers incorporating interviews to (explicitly or implicitly) assume that the narrative produced represents an authentic or truthful representation of reality, rather than being contingent upon interaction and always co-constructed. In many research reports that incorporate interview data, there has been a tendency to decontextualize the process of data construction and instead, “focus on the respondents’ turns as if they were discrete speech events isolated from the stream of social interaction in which – and for which – they were produced” (Wooffitt & Widdicombe, 2006, p. 39).  64  The under-theorized nature of interviews is troubling given how commonly they are used in applied linguistics research (Potter & Hepburn, 2005; Prior, 2018; Talmy & Richards, 2011).  Talmy (2010) has offered a useful heuristic to help identify the theoretical underpinning of interview approaches. First, those utilizing interviews as a “research instrument” tend to conceptualize interviews as an “unproblematic technology for investigating objective facts, subjective experience, and authentic feelings” (p. 26). For these researchers, the influence of the interviewer in shaping interviewee’s responses is minimized. This approach underscores the (often implicit) assumptions regarding the nature of reality and the interviewer’s ability to uncover (or reveal) the interviewee’s “true” perception or experience. However, researchers conceiving of interviews as a “social practice” problematize these assumptions and approach interviews “not as sites for the excavation of information held by respondents, but as participation in social practices” in which the co-construction of discourse is fully acknowledged and investigated as part of the inquiry (p. 28).  In interviews with participants, I regularly encouraged participants to initiate discussion whenever they felt the need to question terminology or ideas or to more deeply discuss a meaningful perception or experience they reported. I also initiated discussion to gain a deeper level of understanding. This enabled me to clarify participants’ intended meaning as well as acknowledge and clarify my own ideological and pedagogical beliefs. Howe (2004) describes this approach to interviewing as “critical dialogue” that involves “bringing expert knowledge to bear and subjecting the views and self-understandings of research participants to rational scrutiny” (p. 55). In conducting interviews in this way, and by acknowledging the active role I played in the co-construction of the discourse generated, I am approaching interviews as a social practice (Talmy, 2010). My changing beliefs and perceptions were inevitably influenced by having a background in  65  teaching English to multilingual learners and the development of collegial relationships with some participants. However, as it is not a focus of this study, I do not conduct an extended analysis of co-constructed interview discourse and my role in its creation. Nevertheless, I fully acknowledge the situated and co-constructed nature of the discourse generated in interviews and recognize that, as such, it cannot represent internal “truths” that simply emerge through the technology of the interview.  3.5.2 Collecting assignments and documents I asked students to voluntarily submit their academic writing assignments over the course of their participation in the study. Whenever possible, I also collected course syllabi, assignment instructions, outlines, drafts, and any feedback received. My primary objective in collecting written assignments was to better inform subsequent interviews and deepen discussions with students. Reading students’ written assignments provided additional context for interviews and ongoing discussions of various aspects of written academic discourse, including the concepts and notions they were being explicitly taught in their classrooms and how this influenced what was written and why. Depending on the assignment being discussed, it sometimes provided a springboard for a wider discussion about personal issues, interactions with peers or the instructor, opinions about specific courses or the program, or wider sociocultural issues.  While it was important to collect, read, and discuss the academic writing being performed by student participants, data collection and analytical efforts were not intended to be all-inclusive in this regard. To provide a more nuanced perspective, I decided to focus on assignments that participants wanted to discuss because they were meaningful or impactful to them in some way. I encouraged participants to submit as many assignments as they felt comfortable with, whether we would have the time to discuss them in detail or not. Thankfully, unlike what happened in Leki’s  66  (2007) study in which her participants did not seem to have much to discuss in terms of academic writing (a result that led her to abandon this initial focus), all but one of my participants displayed their willingness (and sometimes desire) to discuss their understanding and performance of academic writing in interviews, the exception being Ashlee. Theo and Chloe began identifying specific areas of their writing assignments they wished to discuss and sometimes created specific questions they wanted to explore further. Perhaps this enhanced engagement was influenced by giving students a choice of what to discuss and having something concrete to discuss in the interview. Also, participants might have been influenced by being explicitly instructed in academic and disciplinary writing, which may have provided them with some additional motivation and background knowledge to engage in the discussion.  A total of 146 written assignments were submitted by student participants during the study. 85/146 (58%) of these assignments were from students in the Arts Stream and 61/146 (42%) were submitted from students in the Sciences. Some participants submitted more assignments than others, Ashlee submitting 33, Jasmine, 31, Chloe, 30, Yvonne, 24, Dawn, 15, and Theo, 14. The length and type of assignments varied considerably, ranging from one or two paragraphs of largely descriptive writing to argumentative essays longer than five pages in length with citations and references included. Many of these were typed in a word processing program and submitted in electronic form (e.g., Word), but others were handwritten and submitted on paper (e.g., in-class writing, timed exams, etc.). All handwritten documents were scanned and converted to electronic form (.pdf) or in some cases a picture (.jpg) was taken.  Types of assignments submitted included journals, lab reports, short responses, summaries, extended definitions, mathematical proofs, data commentaries, article reviews, position pieces, comparative papers, mid-term and final exams, research reports, argumentative or analytical papers,  67  and others. Some participants were more willing to share their feedback and grades than others. Chloe and Theo provided all their grades for all courses completed at OC and in their second and third-year courses at MSU. However, Yvonne and Dawn did not provide any final grades for their courses, only some of their assignment grades, and seemed reluctant to include instructor feedback. Given the voluntary nature of assignment submission, this created considerable variation across participants and Streams. While a cross-case linguistic or textual analysis would have provided a valuable additional perspective on learners’ language and literacy development, it was beyond the scope of this study. Written assignments were very helpful in interviews to contextualize and deepen our discussion. Often specific sections, sentences, or phrases in these assignments were highlighted for more targeted discussion, either by me or by the student. This process continued throughout the study.   Additional writing was collected from some of the students based on their desire to contribute more to the study. While I encouraged students to contribute as much as they were comfortable with, I did not explicitly request this additional writing. For example, Jasmine provided writing she did while in the final grade of high school. Also, since she did not successfully complete the OC program, she could not enter MSU in her second year. However, she enrolled in another MSU-affiliated college on campus. She submitted several of the writing assignments she completed in that program. The following year (her third), Jasmine was accepted into the MSU mainstream, and she submitted additional assignments from some of her MSU courses. Chloe and Theo also provided writing assignments they completed in their third and fourth year in the MSU mainstream (note: these additional assignments are not included in the totals listed above).  To address the second overarching research questions (see Table 1.1), I had originally asked participants to write reflections on any memorable, impactful, or critical incidents they had while  68  learning about academic writing or completing their academic writing assignments and to submit these after the end of each term. However, only Yvonne had submitted a reflection by the middle of the second term. I asked remaining participants if this was a challenge for them and most indicated they would rather talk about it than write about it. Also, all participants indicated they felt overwhelmed with their coursework. For these reasons, I did not ask for any additional end-of-term reflections. However, in their fourth year, Theo and Chloe agreed to write a written reflection of their first-year experience in OC and their transition to MSU. I also collected classroom-level and program-level documents, whether publicly available (e.g., in program brochures or on websites) or provided to me by participants, as well as media reports related to the academic language program. In addition, I took research notes during and after observations and in discussions with various stakeholders throughout this study.  3.5.3 Observing classrooms and other events According to Marshall and Rossman (2016), observation “entails the systematic noting and recording of events, behaviors, interactions, and artifacts (objects) in the social setting” (p. 143). It is rarely the only data collection method and is normally combined with others such as interviews or document analysis (Duff, 2008, p. 141). There are several potential challenges faced while conducting observations, including but not limited to gaining access, ethical dilemmas, managing one’s role in the observation site, or being able to collect the data in a systematic, focused, and intelligible fashion (Duff, 2008; Marshall & Rossman, 2016). These are some of the challenges I faced while conducting observations and warrant additional discussion.  Throughout this study, I conducted non-recorded observations of lectures and tutorials for which the written assignments submitted by students had been assigned. I conducted a total of 22 classroom observations (representing over 30 hours of instructional time). Fourteen of these  69  observations were in OC classrooms (i.e., sheltered) and eight were in MSU classrooms (i.e., mainstream). Of the eight mainstream classrooms observed, two were conducted in elective classes in the third term of OC and six were conducted in the second year after students transferred to MSU. I also observed an end-of-term student-led interdisciplinary conference hosted in another city over two days in which all student participants presented the results of their research projects. My primary objective in conducting these observations was to better contextualize and triangulate data gathered through interviews, discussions, and documents. For all observations, I took notes about the physical orientation of the class, notable discourse generated, and what participants were doing (see Appendix F). I also wrote additional notes as soon as possible after the event to express my personal thoughts or insights.  3.5.4 Thematic analysis Thematic analysis, although highly diverse in its application, has been described as a foundational method in qualitative research (Braun & Clarke, 2006) and has been characterized as a “process that can be used with most, if not all, qualitative methods” (Boyatzis, 1998, p. 4). Thematic analysis is generally utilized for “identifying, analyzing and reporting patterns (themes)” in a consistent manner that “minimally organizes and describes your data set in (rich) detail” (Braun & Clarke, 2006 p. 79). Braun and Clarke (2016) identify two broad approaches to thematic analysis that represent divergent ontological and epistemological assumptions. First, those who take a “coding reliability” approach tend to conceive the process as one of capturing, recognizing, or noticing themes, implying they “conceptually pre-exist the analytic and interpretive efforts of the researcher” (p. 741). However, for those taking a bottom-up inductive approach: Coding and theme development processes are organic, exploratory and inherently subjective, involving active, creative and reflexive researcher  70  engagement. The process of analysis – rigorous coding followed by a recursive process of theme development – involves the researcher ‘tussling with’ the data to develop an analysis that best fits their research question. (Braun & Clarke, 2016, p. 741)  My approach to thematic analysis is best characterized as exploratory, inductive, iterative, and subjective. While I was informed by my theoretical and methodological framework, reading of relevant literature, and personal and professional experiences, I did not develop any pre-established categories or codes nor were any themes developed early in the analytical process. I regularly revised the coding of collected data and actively sought further triangulation in subsequent data collection efforts. All data was analyzed with the aid of NVivo qualitative analysis software (version 12). Figure 3.1 is a screenshot of some of the descriptive themes developed for Chapter 4.                                       Figure 3.1: Example of descriptive nodes identified for Chapter 4  71  Data was organized according to its source (e.g., interviews, written assignments, observations, curriculum documents, research notes, etc.) and according to the student and non-student participants involved in this study. Figure 3.2 is a screenshot of some of the interpretive themes identified for Chapter 4. Note that the file structure in the left menu (under Nodes) has changed from “Descriptive analysis” to “Thematic analysis.”                                              Figure 3.2: Example of interpretive nodes identified for Chapter 4 Initial analysis continued throughout the data collection period and consisted of creating tentative codes (termed “nodes” in NVivo). This was intended to capture my initial thoughts and impressions of the data, better inform subsequent interviews and data collection efforts, and to help identify and develop themes influential in student participants’ written academic discourse socialization. I also read participants’ submitted assignments and discussed these with participants  72  in ongoing interviews and discussions. This enabled connections to be made between what participants said about these academic discourse features and the various ways they were textualized in writing, not always in ways they intended. A bottom-up, inductive approach was followed during the descriptive phase with the aim of creating a list of nodes that richly describe the data. This process consisted of identifying discourse that exemplified participants’ perceptions and experiences in relation to the research questions posed. Nodes were continually created, expanded, and collapsed. This initial stage of descriptive analysis led to the identification of “lower order themes” (O’Neill, 2013, p. iii) that were continually adjusted. The second stage of analysis was more interpretive and led to “higher order themes concerned with analysis and drawing conclusions” (p. iii).  I now present the three manuscript chapters as described in Chapter 1. To reduce redundancy in each of these chapters, I have omitted some information about the theoretical and methodological approach taken, the research context, the participants’ background information, and the data collection and analytic procedures.     73  Chapter 4: Sheltered to mainstream? Multilingual students’ first-year experience in a specialized academic English program   4.1 Introduction The need to attract world-class talent, mounting budgetary concerns, university ranking practices, and a host of other factors have contributed to intense competition for educational resources around the globe (Knight, 2013). As a result, most universities have resolved to internationalize or globalize their campuses inspired by the dual need for brain power and income generation (Lee, Maldonado-Maldonado, & Rhoades, 2006). In the Canadian context, reduction in federal funding for public universities across the country has intensified the need for universities to seek funding from other sources, and international education has become an increasingly attractive option since domestic student tuition increases are subject to yearly government restrictions (Xiao, 2018). In line with these developments, the international student populations in many Canadian universities have increased substantially, a development that is expected and encouraged to continue into the foreseeable future (Anderson, 2015; Foreign Affairs, Trade and Development Canada, 2012, 2014).  Driven by and driving these trends, universities continue to invest in academic language programs designed to attract international students to their campuses and support them in their transition to their respective programs of study after they arrive. The design of these programs varies considerably and is informed by an eclectic range of theoretical and pedagogical approaches, often woven together to achieve diverse program objectives (Cheng & Fox, 2008; Fox et al., 2014). In English-dominant countries, these programs have often been referred to as English for academic purposes (EAP) or English as a second language (ESL), but there are a wide range of pre- and post-admission language programs that have been developed (e.g., foundational writing courses,  74  pathway or bridging programs, etc.). The challenges involved in understanding the rapidly changing nature of these programs has made investigations into their usefulness and appropriateness arduous, particularly from the perspectives of the learners who enrol in them (Keefe & Shi, 2017; Tweedie & Kim, 2015).   The design and delivery of any language program is shaped by historical, social, cultural, and economic forces, as well as the professional histories and personal preferences of the individual actors involved. As such, there is an inextricable link that connects the design of any program to the teaching and learning practices that ultimately transpire, even if these are not intended (Barton & Donahue, 2009; Brown, 2009; Fox, 2009). However, for international students who must choose from among these programs, there is often limited information available to make informed choices. Nevertheless, the ability of multilingual learners to understand and appropriately respond to program expectations will shape how they are assessed and what level of success (however defined) they can ultimately achieve in the program and beyond. Given the growth, range, and pedagogical complexity of academic language programs operating in university contexts around the world, the need to investigate their role in shaping the perceptions and experiences of multilingual learners is more important than ever. In this study, I investigate the ways in which a newly-designed academic language program mediated the transitional experiences of six international students who had enrolled in its inaugural year. This foundational-year program was designed internally by stakeholders already working in the university in which the program would operate. Program designers drew upon a wide range of theoretical and pedagogical approaches to achieve their educational objectives including but not limited to genre-based pedagogies, systemic functional linguistics (SFL), content and language integration (CLIL), task-based learning, experiential learning, multiliteracies, and disciplinary- 75  specific materials, amongst others. Students enrolling in this program were unable to satisfy direct-entry English language admission requirements established by the host university but met the admissions criteria for this specialized foundational-year program (see Section 3.2 for additional details). Since all program courses were taken exclusively with other international students enrolled in the program, this shared first-year experience offered an opportunity to investigate the role of program design and delivery in mediating the transitional experiences of these students. This study contributes to a deeper discussion of how program features mediate academic apprenticeship practices involving multilingual learners. Such a focus is especially relevant in the Canadian context given the rapid and sustained growth of the international student population and the continual deployment of programs designed for them (Fox et al., 2014).  Given the complexity involved in the design and delivery of all academic language programs, it is important to bear in mind the difficulties faced by administrators and instructors in the first year of any program and the need to identify and address problems as they occur (Brown, 2009; Morris, 2006). It is also important to note that this study was not intended nor able to evaluate the mix of theories and pedagogical approaches utilized in the program or conduct a student needs analysis. Instead, the primary objective was to better understand student participants’ changing perceptions and experiences as they transitioned through this program and into the mainstream. Before describing the study in more detail, I discuss some of the ways academic language programs involving multilingual learners have been researched to contextualize the approach taken in this study and highlight its main contributions. 4.2 Challenges in researching the student experience in academic programs  Academic language programs within higher education contexts have been studied from various perspectives, for diverse purposes, and for the benefit of both internal and external program  76  stakeholders. Many academic language programs have conducted internally-designed and administered needs analyses (or assessments) in which various aspects of the program are investigated, including the perceptions and academic performance of enrolled students. Unfortunately, the results of such assessments tend to be used for program-internal purposes and are not generally made publicly available. These initiatives tend to favor quantitative over qualitative methods and often focus on expected program outcomes (Brown, 2016; Kiely, 2006) as opposed to unforeseen developments that may be overlooked or missed because they do not directly inform the pre-determined variables of interest. Some scholars have provided enlightening narratives that describe some of the challenges experienced in conducting such assessments (e.g., Brown, 2009; Morris, 2006). These narratives suggest that learner needs analysis and program evaluation initiatives are a highly dynamic, complex, and contested endeavor. What specific learner needs get identified as important and how they are measured are influenced by numerous factors, including but not limited to: institutional policies and preferred ideologies, the technological and human resources made available, and the theoretical and pedagogical predilections of individual decision-makers. As Brown (2016) notes, “different groups of stakeholders in a program are likely to hold different views of what the students need to learn” (p. 12). While needs assessments and program evaluation initiatives incorporate the perspectives of multilingual learners studying in these programs, they are most often designed for curriculum-renewal purposes or to provide evidence of program effectiveness to upper-level administrators, thus restricting their ability to investigate the wider sociocultural context or the individual experiences of multilingual learners.  To investigate the social context surrounding language programs and the decision-making practices occurring within them, some researchers have investigated the role of institutional policies and individual actors in the decision-making process. For example, in the context of an Australian  77  university, Fenton-Smith et al. (2017) reported on a unique institution-wide “mandatory programme of credit-bearing discipline-specific English language enhancement courses” (p. 465). The authors reported that within the wider university community, there was potential to view such a program as a ‘silver bullet’ that could ‘solve’ perceived language issues occurring in the university. Additionally, there was risk of setting unrealistic expectations and timeframes regarding the speed of language development possible in such a program as well as potential reinforcement of the misguided belief that language instruction is solely the responsibility of language specialists. They also cautioned that “institutional stakeholders need to be made aware of the complex and protracted nature of language learning and the confounding variables, many of them external to formal instruction, which complicate attempts to quantify linguistic development” (p. 475).  In the above study, multilingual learners’ perceptions of program design features and their responses to curriculum delivery were investigated over time. Student surveys were administered after each semester and the results informed curriculum changes. However, Fenton-Smith et al. (2017) acknowledged that there were limitations in the student survey data collected and warned of the risk that such results become “measures of student satisfaction” as opposed to “measures reflecting aspects relevant to student learning” (p. 475). The authors called for “richer, more qualitative data” that can shed much needed light on the numerous student, teacher, and course-related variables involved (p. 475). Also, researchers reported challenges in assessing the efficacy and appropriateness of the program because of the need to “validate its implementation” to upper-level administrators who “expect results to justify the magnitude and expense of the operation” (p. 474). In addition, faculty members expected results to support the usefulness of the program and students needed “reassurance” that they were not “wasting time and money by undertaking more language study” (p. 475).  78  In the North American context, researchers have conducted large-scale survey studies within and across higher education contexts to obtain a broader picture of how program design features influence the perceptions and experiences of students enrolled in them (e.g., Braine, 1996; Fox et al., 2014). While somewhat limited in their ability to address the social, cultural, and linguistic aspects impacting individual students, these studies provide valuable birds-eye views of the types of language programs in operation and the various ways multilingual learners respond and perform within them. For example, in a recent large-scale survey study of 641 multilingual learners studying in 36 English language programs situated within 26 Canadian universities, Fox et al. (2014) tested a complex quantitative model and found that language program characteristics influenced students’ reported levels of academic and social engagement. This relationship was “mediated” by the language support activities occurring in ESL or EAP programs and further “moderated” (i.e., lessened or strengthened) by student background characteristics (e.g., stress, anxiety, motivation, etc.). The researchers found that program design features have a “direct, positive, and significant effect on the academic and social engagement” of students and posit that the quantitative model they developed helps to better understand the factors that influence student engagement in positive and negative ways (p. 77). They concluded that their results “begin to clarify” complex and dynamic socialization processes taking place but pressed other researchers to “systematically investigate and further clarify the relationship between language support program emphasis, students’ personal characteristics, and successful transition to university study” (p. 78).  To overcome some of the limitations of traditional needs analyses and large-scale surveys in understanding the perceptions and experiences of multilingual learners in language programs, some have taken a more critical approach that, instead of focusing on description of target situations, envisions them as sites for potential reform (e.g., Benesch, 1996; Helmer, 2013). As Benesch  79  (1996) points out, “taxonomies of needs not only hide their ideological basis but also disregard the unequal social positions of the different parties involved and the possible effects of such inequality on curriculum development” (p. 724). These researchers emphasize that multilingual learners are studying under unequal language and power structures, and that efforts need to be made to assure that those who are in subordinate positions can appropriate the dominant discourses that surround them in ways that also benefit them (e.g., legitimate and draw upon their full linguistic repertoire). Other scholars and researchers warn of the inherent dangers of unexamined assumptions about language conventions that ignore, de-legitimize, censor, or discriminate against the Englishes currently being used around the world (Canagarajah, 1999, 2013; Heng Hartse & Kubota, 2014; Horner et al., 2010). These researchers warn of the limits of strictly descriptive and language-deficit approaches that fail to consider the larger sociohistorical, sociopolitical, and sociocultural dimensions that surround and inform decision-making practices. These critical approaches to assessing the effectiveness and appropriateness of program and curriculum design recognize the need to deeply consider learners’ perceptions and responses to the situated practices occurring in academic language programs.  While not explicitly focused on program design or curriculum delivery, an increasing number of researchers have incorporated qualitative approaches and ethnographic methods to investigate the socialization practices surrounding and informing multilingual learners studying in English language programs, including consideration of social, cultural, and linguistic power differentials and their influence on socialization agents. These ethnographically-oriented researchers (Paltridge et al., 2016) have investigated a range of educational contexts including graduate programs (e.g., Anderson, 2016; Bronson, 2004; Casanave & Li, 2008; Zappa-Hollman, 2007b), undergraduate programs (e.g., Godfrey, 2015; Morton et al., 2015; Séror, 2008; Spack,  80  1997), language exchange programs (e.g., Shi & Beckett, 2002; Zappa-Hollman, 2007a), as well as language programs operating in K-12 educational settings (e.g., Duff, 1995; Harklau, 2000; Talmy, 2008, 2009). This growing body of research draws on sociocultural approaches to language and literacy to investigate the complex, contested, and sometimes conflicted socialization practices occurring in various language programs. Collectively, these studies have discovered highly individualized and dynamic socialization practices occurring over time that both enable and constrain multilingual learners (and other stakeholders) in various ways. The results of these studies provide valuable insight into learners’ perspectives and experiences in a wide range of instructional contexts, including academic language programs. The present study is inspired and informed by this body of research.  While the available research continues to expand in this area, the role of academic language programs in mediating multilingual learners’ perceptions and experiences over time has received far less attention (Fox et al., 2014). There have also been few studies that have tracked learner perceptions and experiences after leaving a program or course and continuing their academic studies in another socioeducational context (although see Harklau, 2000, and Zappa-Hollman, 2007a, for two notable exceptions). This is especially important given the pace of change in the international student population in Canada, along with the continual design and deployment of new academic language programs intended to both capitalize on and respond to shifting university student demographics around the world.  4.3 Description of study  In this study, I investigate the perceptions and experiences of six international students as they transition into (and out of) a foundational-year academic language program designed exclusively for them. Through an analysis and comparison of student and non-student discourse in  81  characterizing aspects of the program, I seek a deeper understanding of the role of program design in shaping the reported experiences of administrators and instructors as they work to deliver the program effectively, as well as the changing perspectives of students as they work through the program to gain entry to the university mainstream. The following research questions guided this study: 1. How are program objectives and design features characterized in documents and reported on by non-student participants (i.e., consultant, administrators, and instructors)? 2. How do student participants characterize the program and describe its influence on their transition to mainstream university studies? Theoretically and methodologically, I draw on an academic discourse socialization (ADS) perspective (Duff, 2010; Kobayashi et al., 2017; Morita & Kobayashi, 2008), an area of inquiry that explores the complex and dynamic ways that language, language users, and social contexts are implicated in the socialization practices that transpire (see Section 2.3.3 for more on this approach). 4.3.1 Research context All six student participants in this study were enrolled in the inaugural year of this program. I refer to this program with the pseudonym Oasis College (OC), and the host university, the pseudonym Main Stream University (MSU). The university is a large research-intensive institution with a highly-diverse student population. OC offered programs of study in two main disciplinary Streams: 1) the “Arts” which included subject-area courses in psychology, political science, and geography, and 2) the “Sciences” which included courses in math, physics, and chemistry. Course instructors for these courses were content specialists who had experience teaching (or were currently teaching) courses in the university mainstream (MSU). In both the Arts and Sciences  82  disciplinary Streams, students were also required to complete academic language courses. Instructors for these courses were language specialists, some having backgrounds in specific content areas (see Section 3.2 for additional details).  4.3.2 Participants All six student participants had enrolled in the same academic language program in the same year. Four students had turned 19 before the start of the program and two were 18. Three were enrolled in the Arts Stream and three in the Sciences. One student was born in South Korea, one in Russia and four in China. All student participants remained in the study throughout their first-year in the OC program and at least one term in MSU. For all the students in this study, the OC program offered an alternate pathway to enter MSU because they were unable to meet direct-entry English language requirements. Non-student participants in this study include a consultant involved in the early design of the program, two administrators with specializations in each disciplinary Stream (Arts or Sciences), and two instructors teaching within each disciplinary area. There were also non-focal participants in this study who contributed in various ways. Although not formally interviewed, ten additional instructors and two teaching assistants agreed to classroom observations (see Section 3.3 for additional details about participants and their contributions to this study). 4.3.3 Data collection and analysis Primary data sources for this study include interviews conducted with student and non-student participants and the collection of program and curriculum documents. I conducted 34 interviews with six student participants (5-6 interviews each), representing 40.5 hours of audio-recorded data. I also conducted interviews with other program stakeholders at various points in the study. The consultant was interviewed before OC initially opened its doors and both administrators within two months after the program began (and within weeks of each other). Two OC instructors  83  were interviewed at the end of the second term after two classes had been observed.  All interviews were semi-structured and guided by a list of general questions (see Appendices D and E). I also conducted a total of 22 classroom observations, fourteen in OC classes and eight in MSU classes. Of the fourteen OC classroom observations, nine were conducted in academic language courses and five in subject-area courses. For all observations, I took notes about the layout of the classroom, instructor or participant discourse, what participants were doing, the primary objective of the lesson, and any teaching and learning activities conducted (see Appendix F). This information informed subsequent interviews with student participants and aided in contextualizing students’ perceptions and reported experiences across contexts and over time. All interview and document discourse collected for this study was analyzed thematically (for additional details on data collection and analytical procedures, see Section 3.5).  4.4 Results and discussion The results of this study are organized as follows. First, I report and discuss salient themes unique to non-student discourse (the program perspective), then themes that were shared across non-student and student discourse (the shared perspective), and finally themes unique to student discourse only (the student perspective).  4.4.1 The program perspective Non-student discourse included interview data collected from a consultant, two administrators, and two instructors working in the program as well as institutional and program documents collected throughout the study and my research notes. Three salient themes were identified in non-student discourse:   84  • program uniqueness and innovativeness  • integration within OC and MSU • program and student advocacy  a) Program uniqueness and innovativeness  The unique and innovative design of this program was a salient theme identified in non-student discourse. For the consultant and both administrators, the potential to be innovative was greatly facilitated by, amongst other factors, the internal design of the program. According to an early consultant for the program, who was also an MSU faculty member, contracting out to external companies like Navitas (https://www.navitas.com /partnerships) was initially considered but ultimately abandoned. The consultant reported that one important factor in the decision was recognition of the benefits of internal program design for the institution, administrators, instructors, and students. This was a relief, in the consultant’s opinion, since an externally-designed and administered program would not adequately meet the educational needs of the multilingual learners: Those [external] programs are, they’re a non-functional approach, it’s the you know teaching grammar, academic word list, … they’re everywhere, they’re like, I’m careful with this metaphor, but they’re like a cancer I think on institutions and for international students. I think they do an injustice, they’re there truly only to generate revenue. (Consultant, in interview, before term 1) However, in the consultant’s opinion, an internally-designed program afforded the opportunity to design a truly innovative or “cutting-edge” program:   85  There’s just not a program like this in North America that I’m aware of and I think it would be a huge, pardon my language, [expletive] you to [names two well-known external companies]. We can do this in a scholarly way, mobilize our intellectual resources in implementing a really rigorous, cutting-edge, theoretically and methodologically cutting-edge program. (Consultant, in interview, before term 1) Administrators also described the unique design of the program and how this encouraged innovation, which according to one administrator was “supposed to happen” because it was built into the program mandate (Administrator 1, term 1). Similarly, in program documents, OC was described as a having a “unique” approach to program design, characterized as a “living laboratory” and a “catalyst” for new pedagogical and research initiatives (from program documents). Innovation was associated with, among other theoretical and pedagogical approaches, genre-based pedagogies (e.g., move analysis), content and language integration (e.g., CLIL), functional approaches to language (e.g., SFL), discipline-specific language pedagogy, task-based learning, and several others. Instructors also commented on several innovative aspects of the program (e.g., content and language integration, smaller classes, interdisciplinary collaboration, etc.) that enabled them to tailor instructional practices to the needs of multilingual learners in a less competitive environment than direct-entry students tend to experience in the mainstream.  We can tailor what we do, and it's for them at their level. That's very efficient and that's a good thing. I think that's the primary reason for it, because if we did throw them into the general population, they would be adrift in a very serious way. This allows us to connect with them a little bit easier without competition. (Instructor 2, term 2)  86  b) Integration within OC and MSU On the program website, in brochures, as well as in interviews with administrators, program features intended to integrate the program within MSU were mentioned including: credit-bearing courses taught by MSU faculty, customized academic English instruction and support, and the ability for students to transition directly into the second year in MSU upon successful completion of program requirements. The program was also positioned as a model or beacon for positive change that could serve to inform the wider university community about innovative pedagogical practices that have relevance outside the program (e.g., the benefits of collaboration across content and language specialists, more attention to language learner needs, a more disciplinary-specific approach to language instruction, etc.). For example, in one of the publicly-available brochures created to advertise the program, the program was described as a “reimagining” of the first-year student experience and one that emphasized “learning without borders.” Amongst other program design aspects identified were smaller class sizes, customized curriculum, and learning models that were responsive to students (from program brochure).   In interviews, both instructors described efforts they made to connect language and content courses for the benefit of their own development as teachers as well as to enhance the learning experiences of their students. For example, the language instructor described collaborating with a content instructor regularly to learn subject-area material more deeply. The instructor hoped this knowledge would assist in selecting appropriate subject-area material for more targeted language instruction and controlled practice. Both instructors also reported that it was necessary for them to integrate disciplinary-specific concepts and language into their classroom practices. However, these efforts were challenging for them:   87  I was still finding it challenging to be perfectly honest and I still feel that next year, and I think I will be doing [the content course] again, I think that I might have to go to all of the lectures to really learn the content. I’ve been so busy keeping up with the language part that I need to do. (Instructor 1, term 2) Administrators highlighted the importance of strengthening administrative partnerships with MSU faculties, making connections with MSU stakeholders, as well as facilitating a “collaborative integrated approach between our language instructors and our disciplinary instructors” (Administrator 2, term 1).  c) Program and student advocacy  The consultant and both administrators reported there was potential for the program to inform other university stakeholders about innovative teaching, learning, and assessment practices, and provide inspiration for positive change in mainstream university contexts. As one administrator explained: So, I think with this program, we maybe have unique drivers, but they apply across all the classrooms and that’s another aspect of why I am involved in the [OC] program is that in rethinking what a first-year experience looks like through these students, we will be learning a lot and the hope is that those practices spread. (Administrator 2, term 1) For the consultant and both administrators, it was necessary to advocate for the program and for the students enrolled to combat inaccurate or misguided assumptions (e.g., the program was designed exclusively for wealthy international students). Such advocacy efforts were viewed as essential to program success and important in educating the wider university community. Before  88  the program commenced, the consultant commented on both the “exciting” pedagogical possibilities available as well as a need to guard against competing (or conflicting) objectives:  So, this was very exciting, talking about this, on the one hand making this program for credit and integrating language and content, and then on the other hand, you know there were the alarm bells because it’s another initiative, it’s digging into the pockets of international students, it’s the whole international student as cash cow being mobilized. (Consultant, before term 1) The need for program and student advocacy was reinforced by the publication of a media article shortly after the program became operational. This was an online news article published in a prominent national news outlet that specifically identified the program and reported on some of the negative perceptions of various MSU and non-MSU stakeholders. To maintain confidentiality of the program and the participants, the details of the article cannot be reported. I became aware of the article a few days after its publication and only days before my interview with one of the administrators. I was unsure whether I should bring up the article during the interview and ultimately decided not to. However, the following comment regarding “speculation” and “mistrust” about the program was freely offered at the end of our interview:  So, on the one hand, because there is so much speculation and mistrust about what we are trying to do and you know, people who see this only as a business. And we on the academic side of things really see it as an opportunity to do something in our careers that is very meaningful to students. (Administrator 1, term 1)   89  In my interview with Administrator 2, conducted two weeks later, I had intended to discuss the article, but it was brought up before I had the opportunity. Unprompted, the administrator offered the following commentary about the need to advocate on behalf of students and “prepare” the university community: Administrator 2: So, you likely know that there was a [media name] article that put a very negative lens on the project.  John: It made my blood boil! Administrator 2: Yeah. It made our … our faculty reacted to it. I have even seen evidence that our students have reacted to it. And, so I think that there are issues, it had racist overtones as well as, not overtones, racist remarks, as well as issues of class and inclusiveness and equity. John: Assumptions everywhere. The consultant and administrators expressed a belief in the importance of program and student advocacy efforts to combat inaccurate and ignorant perceptions about multilingual learners and the objectives of the OC program within the institutional structure. For example, when discussing the article, Administrator 2 reiterated how important is was to “prepare the campus” for OC students:  Yeah, [the article] was full of misinformation, but there is a big question to ask when we’re doing initiatives, when we’re really trying to think about internationalization and how we rethink what a first-year international student experience looks like, we need to think about the student within the program, but we also need to think more  90  widely about the campus. How do we prepare the campus? Is our campus ready to receive these students? (Administrator 2, term 1) Both instructors reported on the need to improve their preparedness to effectively and appropriately teach in this kind of program. They reported that there were numerous advantages for students who enrolled and affirmed there was great potential in the pedagogical approaches being incorporated.  In the design and delivery of this program, upper level stakeholders emphasized the importance of tailoring instruction to suit learner needs, remaining open to innovative ideas, maximising integration within and across contexts, and positioning students positively within the institutional structure. The internal design of this program, in close collaboration with several MSU-affiliated staff and faculty-members, was instrumental in enabling these beliefs and the concrete actions that were taken to achieve these objectives.  4.4.2 The shared perspective Three salient themes were shared across student and non-student discourse and are discussed below: • high quality of academic support • time constraints in meeting expectations • cultural and linguistic diversity  a) High quality of academic support The perception of a high quality of academic support provided within OC was shared across student and non-student discourse. Non-student participants cited several aspects of the program in support of this perception including: experienced and dedicated instructors, an interdisciplinary approach, integration within OC and MSU, smaller class sizes, content and language integration, collaborative efforts, and the wide availability of tutorials and office hours. Similarly, in program  91  documentation, OC was characterized as an “integrated experience with small classes and coordinated schedules” in which students work with “award-winning faculty” (from program document). Smaller classes were mentioned by administrators and instructors as one of the main factors enhancing the communication that occurs between students and instructors. Smaller class sizes also enabled more timely feedback and more personalized instruction. Other academic services made available within the program (e.g., extensive office hours, academic advisors for each Stream, etc.) were also cited as examples of the high quality of academic support available for OC students.    All student participants except Ashlee also commented on the high quality of academic language support they were receiving within the OC program. This was a perception that did not change after they entered the university mainstream and grew stronger for some. Aspects of the OC program from the students’ perspective included: friendly and dedicated instructors, smaller class sizes, availability of office hours, extensive feedback, personalized instruction, and the incorporation of digital technology. Most participants commented positively on the efforts of language instructors to make personalized connections with students, provide definitions and elaborations, and encourage student participation. Most participants reported that this kind of instruction was helpful to their academic language and content development.  A majority of participants reported that the academic support they received in their content classes helped them to better understand academic expectations and disciplinary-specific concepts. Most also reported that this helped them to complete the written assignments required of them in these courses. Reasons included the ability to understand lectures because instructors reduced their rate of speech, avoided the use of idioms and other expressions, or elaborated on course concepts. As Yvonne (Arts) stated, “because [OC professors] know that we’re all international students, if  92  they say something about cultural reference and nobody like knows, they will explain what that is” (Yvonne, term 2). This positive perception of academic support in the classroom context intensified for some participants upon entering the mainstream. For example, some students reported difficulties keeping up with the speed of communication in their mainstream courses and others had challenges understanding the English accents of some of their mainstream instructors. All participants noted that class sizes in most of their mainstream courses were larger and they did not feel they received as much personalized instruction, detailed feedback on assignments, or meaningful opportunities to discuss their assignments with instructors, TAs, or their peers.   b) Time constraints in meeting expectations All student and non-student participants commented on the lack of time they had to satisfy perceived expectations, whether their own or others. Administrators indicated there were time pressures involved in the rollout of the program. For example, one administrator commented that “everything happened in such a short time … I’m not sure if improvising or more like growing more organically … but never enough time to really do a fine-grained analysis of, and sort of iron-out all the wrinkles before we went ahead” (Administrator 1, term 1). The consultant also expressed concern about the time available, believing it was not “not doable, at least doing it in the way it should be done” (Consultant, before term 1). Both instructors indicated that time was a challenge for them in performing their roles as effectively as they would like. One instructor indicated there had been very little time to study the curriculum, which was sometimes received “in pieces” only days before class: Instructor 1: We were getting it at the 11th hour, sometimes on the Sunday night before teaching Monday morning.  John:  So, when you say in pieces, you mean like literally …  93  Instructor 1:  … unit at a time.  John:  A week or less before you have to actually teach it? Instructor 1:  Sometimes we would get it a couple of days before. John: Wow, challenging!  Both instructors commented positively on the innovativeness of the pedagogical approaches taken within the program (e.g., CLIL and SFL) and described the curriculum as very detailed. However, both instructors reported challenges in understanding and/or delivering the curriculum as designed, one instructor describing it as “hard, meta, and messy” (Instructor 1, term 2). For this instructor, the curriculum did not provide well-thought out opportunities for writing practice nor time to provide meaningful formative feedback, pedagogical aspects the instructor regarded as essential for students to improve. Instead, the instructor reported that the complexity and rushed pace of the curriculum felt like information was being “jammed down [students’] throats” (Instructor 1, term 2). This also had a direct impact on the instructor’s level of confidence in front of students:  John:  So then, as you are teaching, did you feel more frustration as the months went on, like how did you deal with … Instructor 1:  Well some of it was because I didn’t understand it, like I’ve never felt so stupid and underqualified, like I had to remind myself look you’ve done good work, but I … I had my teacher’s notes right at hand, if I had shown up for class forgetting to bring my teacher’s notes, I think I would’ve flown into a panic.  94  Despite these challenges, the instructor expressed a strong belief in the value of the curriculum for the learners in their classrooms. However, the instructor reported believing more time was needed for learners to develop meaningful connections between the academic language concepts being taught and the disciplinary discourse practices characteristic of their content-area courses. In my ongoing review of the curriculum provided by instructors, I also made notes about the complexity of SFL-informed lesson plans, tasks, and assessments as well as the extensive metalanguage included. I spent considerable time consulting additional sources to gain a better understanding of some of these concepts (e.g., nominalization, theme and rheme, interpersonal positioning, participants, processes, etc.). Having taught English for many years and having a workable understanding of English grammar, this background knowledge provided little assistance.  Instructors indicated that a lack of time influenced their ability to integrate language and content in meaningful ways for the full benefit of students. Instructors also reported that the academic language courses intended to be linked to content courses were not well-integrated from the start. However, they reported that this was an issue that was being addressed through collaborative efforts and continual program adjustments, and as a result had improved by the end of the second term. Both instructors also indicated that this became less of a factor in the second term, in part, because instructors and students had become more familiar with each other and program expectations. Student participants reported that time constraints were highly influential in their ability to meet their own and others’ expectations. This was most often associated with their course workload and the number of assignments they were expected to complete. They reported that this made it challenging to process and apply concepts being taught in their content and language classrooms, understand their required reading, and make connections between content and language courses.  95  This was a perception that did not change until the third (summer) term when they reported that their workload became more manageable. In this term, they took fewer courses and participated in an interdisciplinary student conference held in another city, which all participants reported to be a very useful, enjoyable, and rewarding experience. Five of six student participants (except Theo) cited time constraints as one of the main reasons they were unable to make connections outside OC, get involved in MSU organizations, or participate in community events. As Yvonne stated near the end of the second term, “I wanted to join the animal club but I just … I didn’t find any leisure time in my schedule” (Yvonne, term 3). Some students reported believing that their workload exceeded that of direct-entry students. However, Theo (Sciences) reported from the very first interview that he was “not that stressed out” by the workload because he took a very pragmatic approach to his studies (Theo, term 2). He also reported “not worrying too much” about understanding all the information presented (Theo, term 3). He also reported reading strategically, following assignment instructions closely, and adjusting his work based on feedback received. This pragmatic approach to his studies enabled him, in his view, to find time to make connections outside of OC (e.g., joining a sports club and volunteering in a local elementary school). c) Cultural and linguistic diversity  The final shared theme was a lack of linguistic and cultural diversity in the program. As researchers have discovered, there can be an appreciable effect when English language classrooms are primarily composed of learners who have the same first language, or what some have termed a “majority-L1 classroom” (Mori, 2014), although how this “affects socialization processes into English-only policies and classroom language use has not been as thoroughly investigated (p. 153). OC administrators discussed some of the potential challenges in the program influenced by the  96  majority-Chinese student population. For example, in my interview with one of the administrators, we discussed students’ previous educational experiences and the potential influence this may have on their transition to a Canadian university. While we both acknowledged the importance of avoiding generalizations, we both commented on the influence of students’ previous educational experiences, particularly their familiarity with Canadian higher education practices and the possibility these expectations are quite different in their home countries.  Administrator 1: We have heard some students say for example that they, in certain countries you just work really hard to enter university but once you are in university, it doesn’t really matter whether you go to class or not. You just need to show up to take your tests. John:  That was my experience in Korea. I heard that so many times. I think it’s like a socioeducational thing, but it’s hard to make that generalization.  Administrator 1:  Yes, right. It’s really hard. And you know we are informed by that literature as well. But I don’t have facts to prove that that’s the case and that that’s the main driver of student behavior here.  Administrators also mentioned some of the potential differences associated with the “values” of the program and the background experiences of students, many of whom attended high school in China. For example, Administrator 1 commented on the difficulties some students may experience in “adopting” practices designed to help them become “proactive learners.”   I do think that the fact that our students are mainly from one part of the world does not allow them to see other ways in which, other values, and sort of become familiar  97  with practices that we want them to adopt just because we know that it will help them to be more proactive learners, right? (Administrator 1, term 1) Administrator 1 discussed the challenge of encouraging the use of English in the program and expressed reservations about “forcing” people to use English. In interviews with instructors, the lack of English use in the classroom was mentioned as a concern. According to both instructors, a lack of cultural and linguistic diversity in the program made it challenging to encourage students to use English during group discussions and collaborative tasks. For example, one instructor reported that had there been more diversity in terms of language, “there would be no need to go around and say, speak English!” (Instructor 2, term 2). This was a practice the instructor reported doing often, but with little success. This was supported by two classroom observations in the second term in which I noted this instructor uttering this imperative on multiple occasions, either loudly to the class or softly to individual students. However, both the students and the instructor did not appear to take the imperative seriously as students often smiled or giggled and the instructor “smiled knowingly” (from observation notes). The instructor confirmed this in our interview and likened the situation to somewhat of a game. The instructor also reported that putting all the students in the same residence created additional problems because students had less opportunity to participate in diverse cultural and linguistic exchanges (Instructor 2, term 2).  All student participants commented on the lack of cultural and linguistic diversity in the program, but they expressed varied opinions about the influence this had on their educational environment or learning progress. All three participants in the Sciences Stream (Ashlee, Dawn, and Theo) indicated the lack of cultural and linguistic diversity in OC did not create issues for them, albeit for different reasons. Ashlee stated she preferred to work alone. For her, a more influential factor was the large difference she perceived in the knowledge level of OC students, particularly in  98  relation to scientific principles. Dawn mentioned that having many students from her home country enrolled in the program helped her to feel more comfortable in her first year and enabled her to make friends, something she said would have been much more difficult if she had entered the university directly. Theo maintained that it was not an issue for him personally. Instead, he reported being proactive in seeking out opportunities to interact outside of OC. However, Chloe, Jasmine, and Yvonne (Arts Stream) indicated that the lack of cultural and linguistic diversity in the program negatively impacted classroom communication, particularly during group work activities, and that this sometimes influenced their learning experiences negatively. Chloe and Yvonne reported that there was an apparent unwillingness to use English amongst many students unless it was forced or deemed necessary. They reported that this had a negative impact on some of their group work experiences.  Chloe described an ongoing challenge that occurred in one of her group projects in Term 2 which involved, in her words, “racist” attitudes (Chloe, term 3). She described how a student from a European country would blatantly ignore Chinese members of the group. While he would speak to her or other non-Chinese speaking students, he consistently ignored the opinions and contributions of Chinese group members. Similarly, Jasmine described how, in her opinion, a lack of language diversity in the program sometimes influenced group formation:  In [OC], 80% of the students are Chinese students. So, for every student who has the same nationalities, it's no different than they study domestically. Because we all know each other, because we are from the same country, and we know what our previous education is, and what our habit is, nothing changed. And if we could not understand in English, we could communicate in Chinese because we were all Chinese, we speak the same language. That also caused, for every class we need to  99  have group discussions, why students without Chinese nationalities gather in one group and Chinese students gather in another group. (Jasmine, term 3) Although Jasmine reported feeling comforted by having many other students in the program to talk to in her first language, she also indicated that the lack of cultural and linguistic diversity negatively impacted her ability to practice English and improve her English language abilities.  The above three themes identified as salient across student and non-student discourse provide insight into program design features that may have impacted all participants in the academic language program. This includes the degree to which academic support is seen as useful, the time available to meet expectations, and the impact of student demographics on instructional and communicative practices. In this study, most participants viewed the academic support provided in the program as extensive and useful, although there were significant time pressures reported in meeting objectives. Given the ambitious objectives established for this program, combined with the fact it was the inaugural year, it is perhaps unsurprising that time constraints were identified as a constraining factor. Nevertheless, lack of time to adequately absorb and apply information created frustration and stress for some stakeholders. These results suggest that a majority-L1 context can play an important role in shaping stakeholders’ perceptions and experiences and indicates there is need for much more discussion and research into its impact on language socialization practices (Mori, 2014).    4.4.3 The student perspective Three salient themes were identified in student discourse and are discussed below: • credit status of courses  • integration vs. separation • sheltered to mainstream   100   a) Credit status of academic language courses  Prior to entering the program, all student participants reported believing that their OC courses were credit-bearing and could be applied to their future programs of study. However, at some point during the second term, students became aware that some of their credits were not transferable to MSU in the way they had initially understood. Although two of their academic language courses were credit-bearing in that they were required to complete OC program requirements, they would not count as credits toward their mainstream programs of study. It should be noted that the total number of credits required to complete the program was 36 (or twelve courses worth three credits each). For first-year students in MSU, a full course load for the year was typically 30 credits (or ten courses worth three credits each). However, by completing the program, students would also satisfy English language admission standards for MSU and be able to register for mainstream programs and courses, something they had not been able to do previously.  After students became aware of the credit and non-credit status of two of their academic language courses, increasingly negative perceptions and experiences were reported about these courses by four of six student participants: Ashlee, Chloe, Jasmine, and Yvonne. In the first term, Ashlee (Sciences) reported that her academic language courses were not that useful to her but affirmed that she needed them. She indicated that her perception of their usefulness worsened when she discovered they were not for-credit in the way she initially understood it. In the second term. she reported struggling to understand instructional content, complete assignments, and communicate effectively with some of her academic language instructors. Over time, she reported increasingly negative experiences and perceptions associated with these courses.   Initially, both Jasmine and Yvonne (Arts) expressed positive perceptions of their academic language courses. Chloe (Arts) was initially skeptical about these courses but expressed the belief  101  that these courses were necessary. However, after learning about the credit status of these courses in the second term, these students reported more negative perceptions and experiences about their necessity and usefulness. Some aspects mentioned include: reduced attendance and effort by themselves or others, lack of attention to feedback received, lack of effort on revising assignments, getting reduced grades, and becoming more concerned about failing one of these courses. Some students also reported less interest, increased frustration, or increasing lack of concern for their assignments, grades, or the instructors of these courses. Chloe and Yvonne (Arts) reported believing this influenced their engagement with classroom instruction and their attention to the feedback they received from these instructors. Ashlee (Sciences) expressed the strongest and most persistent negative reaction to these courses. She remained unconvinced about their usefulness throughout the program and maintained this belief after she left the program and entered the mainstream. For her, learning that these credits could not be applied to her program of study further solidified her conviction.  Unlike the other participants, Dawn and Theo (Sciences) did not report increasingly negative perceptions or experiences specific to their academic language courses after learning of their credit status. Dawn remained very accepting of the instruction in these courses and tried to understand it to the best of her ability. It did not matter to her that these courses would not count towards her degree because she maintained a positive perception of their usefulness, even though she could not often apply what she was learning. From the outset, Theo reported that these courses offered valuable information that would be helpful to him. In the second term, he reported that these courses added to his workload and impacted his ability to make connections outside OC but maintained his belief in their usefulness. After these courses were complete, he reported not finding everything useful to him, but affirmed that what was being taught in these courses would be useful  102  to him later in his studies. He indicated he was not concerned about the credit status of these courses because he would still have the same number of credits as a direct-entry MSU student after the first year.  Credit-based disciplinary-specific academic language instruction in secondary and higher education contexts has been discussed in the literature and has been the subject of debate and even petitions from student stakeholders (e.g., B.C. migrant students want credit for English courses, 2013). Some researchers have noted that, although non-credit ESL/EAP courses are predominant, credit-bearing and content-based approaches have become more common (Fox, 2017; Melles et al., 2005). According to Melles et al. (2005), “in English-speaking countries such as Australia, the UK and North America the elevation of ESL to disciplinary (credit-based) status remains problematic” (p. 284). The status of ESL/EAP within the institutional structure and assumptions about the academic rigour of these courses continue to create challenges for those who advocate for change (e.g., http://freshvoices.ca/campaign/make-it-count). The results of this study suggest that much more attention is needed to address this issue for the benefit of administrators, instructors, and multilingual students. As much research has found, the need for multilingual students to have their work valued, for instruction not to become infantilizing, and for courses not to be viewed as remedial are all influential variables in how English language courses are perceived and responded to by instructors and learners at various educational levels (Deschambault, 2015; Fenton-Smith et al., 2017; Melles et al., 2005; Talmy, 2008, 2009). Given this, additional research is needed to better understand the role that credit status plays in mediating the teaching and learning practices that transpire in ESL/EAP programs.  103  b) Integration vs. separation  All student participants except Chloe arrived in Canada approximately a month before the start of the program and participated in an orientation program to familiarize them with each other, OC, MSU, and the wider community. Some students reported this helped them to feel a part of the MSU community. By the end of their first term, most students indicated they felt well-integrated into OC, and to a lesser extent, MSU. The reasons reported for holding this belief included: taking classes on MSU campus, using an MSU student card, and accessing MSU services. However, all participants indicated they felt separated or in some other way different from MSU students (e.g., taking courses with OC students, housed with OC students, etc.). In my initial interviews with students, I asked what they had heard or read about the program, whether before or after starting the program. Yvonne reported having read a Canadian news media article in Chinese. This was the same article I had discussed with one of the administrators of the program (see Section 4.4.1), an online article from a prominent national news agency that reported on the program in a negative way. When I asked what she thought of the article, she exclaimed, “It’s so not true!” and went on to explain that not all OC students are wealthy and that her family struggles to pay the high cost of studying in Canada (Yvonne, term 1). Dawn, who had also read the article, described an experience while walking with other OC students in which they were singled out by others because of their OC umbrellas:  Yesterday, I went to [location] and I was with two other [OC] students and we brought our [OC] umbrella. And I heard people say, “Oh, I see two more umbrellas here, two more [OC] umbrellas here” and blah blah blah blah blah. I didn’t hear it clearly, they were just staring at us and then they went away. (Dawn, 1st interview, term 1)  104   In the second term, most of the student participants stated they did not feel as integrated with or connected to MSU as they did in their first term. Some participants reported feeling more separated or segregated from the wider university community either because they were taking courses exclusively with other OC students or they felt restricted in developing wider social networks because of their academic workload. For Ashlee and Jasmine, perceptions of separation increased over the course of the program and remained strong after they left. Theo was the only student participant who did not report feeling separated from MSU or the wider community. In my research notes, I often commented on how proactive he was in making connections on his own (e.g., joining an MSU student club, volunteering in community schools, designing a software application with colleagues, etc.). c) Sheltered to mainstream In interviews conducted after students had started taking their first mainstream courses, participants were asked to reflect on their experiences in OC and comment on the extent to which OC was helpful in preparing them for mainstream study. Chloe, Yvonne, and Theo reported that participating in OC in their first year helped them to make successful transitions to MSU and indicated it would have been more challenging had they entered the mainstream directly. For example, reflecting on the positive aspects of the program, Chloe (4th interview, term 3) described the positive “psychological” effects of enrolling in OC for her foundational year: Chloe: I got some psychological beneficial effects, because, you know, when we think about like, oh, [MSU!], it has higher reputation, and people would be smart, and yeah, like the study would be hard. John: What's the positive psychological effect of being in the [OC] program?  105  Chloe: I think it's good rather than just encountering new environment all the time. You know, like, you will get used to it. But when you first come to Canada, and you just go to the new lecture room, tons of people who you don't really know. Rather than that, [being in OC] gives an opportunity to know people who you actually have a good chance to be close to, and you know, just hang around. And it's the same for the instructor. They're all same and they try to support us, so really, they will be more familiar too, so yeah. While reporting an overall benefit in participating in the OC program, most student participants indicated that some of their academic language courses were less useful, and for some became a source of confusion and frustration. Ashlee, while reporting that she was appreciative of several content courses and instructors, also stated that the “ridiculous language courses” she was required to take were subtractive to her development because “what they are teaching is not really what we need” (Ashlee, 5th interview, term 3). Jasmine reported appreciating many of her OC courses and instructors while in the program, but indicated she became overwhelmed with the pace and linguistic complexity of some of her academic language courses. Ultimately, she was not able to satisfy program requirements. After learning she could not enter MSU in her second year, she reported that some of her academic language courses “dragged on her steps” and made her entry to MSU more arduous and expensive than it needed to be (Jasmine, term 5). Dawn indicated that OC was helpful to her transition but wished there were more opportunities to take mainstream courses and engage with MSU in other ways. She reported that the quality of instruction and feedback in most of her courses was excellent, but that she wished there had been more choice in taking courses. Chloe and Theo, who stayed in touch into their fourth year of study, reported that they felt  106  more appreciative of the OC program as they progressed further into MSU. This was mostly because of the high quality of instruction and academic support provided at OC and the sense of community that was developed over the course of their first year. Both subsequently completed their undergraduate degrees within four years, and both reported believing that it would have been far more challenging had they entered MSU directly. These results align with others that have discovered that many multilingual learners appreciate the additional academic support provided to them and report an overall beneficial impact on their disciplinary studies (e.g., Cheng & Fox, 2008; Baik & Greig, 2009; Fox et al., 2014). The above themes identified as unique to student discourse played an important role in mediating students’ transitions into and out of the OC program. First, not knowing the credit status of some of their academic language courses caused frustration for most of the students in this study. It is not known if students who were initially positive about these courses would have had different reactions had they known the status of these courses from the beginning of the program. When asked in interviews, all participants reported they would prefer to have been made aware of this much earlier. Perhaps compounding the negative perception of these courses, a majority of participants also reported feeling increasingly separated from MSU in their second term (as compared to the first), a result that suggests there may be a detrimental impact on some students the longer they remain in a sheltered environment (Silva, 1994). However, all but one of the participants reported positive perceptions of sheltered instruction in their first year, particularly in their content courses. This finding supports other studies that have concluded that sheltered instructional approaches can have a beneficial impact on multilingual learners’ engagement and performance at the undergraduate level (e.g., Smollett, Arakawa, & Keefer, 2012). The results of this study also provide support for the pedagogical value of content and language integrated  107  approaches (e.g. CLIL, immersion, or CBI) to assist multilingual learners in making successful transitions to English-medium undergraduate contexts (Brinton, 2012; Cenoz, 2015); however, the time needed to make meaningful connections within and across instructional contexts needs to be carefully considered. The predominantly positive perception of sheltered instruction for the majority of participants in this study also suggests that these approaches can, at the undergraduate level, provide a “context in which ESL students are not isolated, where they are in a position to meet, work, and develop a sense of community with those in a similar situation” (Silva, 1994, p. 40). 4.5 Conclusion These results support those of Fox et al. (2014) who discovered that program features mediated multilingual learner perceptions in complex ways, and that these were further moderated (lessened or strengthened) by a range of social, cultural, and affective aspects. These results also support the conclusion that “language teaching which is situated within the university and offered by language teaching professionals” benefits multilingual learners because of the “ongoing interchange they occasion as participating members of the academic community” (p. 78). Academic language programs that are designed and delivered by internal stakeholders are well equipped to make the kinds of connections necessary to maximize students’ performance and encourage academic and social engagement in the university. In this study, most of the students reported they appreciated the efforts made to connect students to each other, the instructors, the content being delivered, and the wider institution. Given that five of the six student participants in this study transitioned to the mainstream in their second year, the results of this study provide support for the efficacy of this model.  However, these broadly positive perceptions can conceal significant challenges that multilingual learners may be experiencing and some of the negative consequences that can result  108  (Shi, 2010, 2011). It is therefore necessary to address these challenges particularly when, as Cheng and Fox (2008) have noted, there are “mismatched agendas.”  When students have defined expectations rooted in what they perceive are their most essential needs, but courses and services within the university (e.g., EAP courses, writing centres, etc.) discount or fail to address these needs, L2 students express dissatisfaction and frustration. In order for such courses and services to be useful, therefore, it is important that the students’ needs be addressed first. (p. 327)     Exploring the manner in which these themes intersected across non-student and student discourse, in both shared and unique ways, provides insight into the how program design features can mediate the beliefs and actions of the individual actors involved. This comparative analysis highlights how aspects of the program can become both enabling and constraining for various stakeholders. This adds to a growing body of research that has examined the perspectives and experiences of various socialization agents in language programs (Bronson, 2004; Godfrey, 2015; Morton et al., 2015; Séror, 2008; Zappa-Hollman, 2007a). While much research into the academic socialization practices involving multilingual learners has described the accomplishments, challenges, and often non-linear pathways that multilingual learners take, investigations have largely focused on teachers and/or learner’ beliefs and practices. What is often overlooked or glossed over is how program and curriculum design aspects (including their impetus) interact with perceptions, beliefs, and actions and how this can play a powerful role in how academic socialization practices are conceived, enacted, and changed over time.  The results of this study suggest that there is much to be gained in investigating how program design aspects mediate teaching and learning practices, whether they focus on the perceptions and experiences of the administrators of such programs (e.g., Fenton-Smith & Gurney,  109  2016), the instructors working in them (e.g., Marshall & Walsh-Marr, 2018), or the students enrolled in them (e.g., Fox et al., 2014). This research sheds much needed light on various program stakeholder perspectives and contributes to a more robust understanding of the complexities involved in assessing program effectiveness and appropriateness. Much more research is needed into the diverse ways that program design and delivery can mediate multilingual learners’ perceptions and experiences over time, including after they leave the program. In short, future research into academic discourse socialization would benefit from a deeper consideration of program design features and how they influence (and are influenced by) socialization practices.  In terms of implications, it is not possible to make generalizable statements given the design of this study, the number of participants involved, the uniqueness of this program, as well the fact that this program was in its inaugural year of operation. It is also important to bear in mind the challenges that inevitably occur in the rollout of any language program (see Brown, 2009), especially one as sophisticated as this one. It is also important to keep in mind the limitations that all language programs experience regardless of the interplay of theoretical and pedagogical approaches incorporated. As Spack (1997) warned over two decades ago in reference to the ability of language programs to meet their disciplinary-preparation objectives,  Given this complexity within a single discipline, given that students move through several disciplines as they fulfill graduation requirements, and given that ESL faculty cannot have expertise in all of the disciplines, we need to be realistic in our expectations of what can be accomplished. (p. 50)  The importance of remaining realistic in our expectations in language programs and courses is reinforced by the continued tendency for university administrators and others to seek a ‘silver  110  bullet’ that will quickly achieve quantifiable language gains, and that the responsibility to accomplish this falls exclusively on language specialists (Fenton-Smith et al., 2017, p. 471).  Nevertheless, there are always lessons to be learned that can be useful for those involved in similar program initiatives. Perhaps the lesson for upper-level stakeholders is to have realistic expectations in terms of the time frame needed to develop such an innovative program, especially ones intended to “revolutionize” the first-year student experience. As this research demonstrates, program features shape the immediate socioeducational context surrounding administrators, instructors, and students. The most meticulous plans (and curriculums) can be rendered ineffective or inappropriate if adequate time is not provided for administrators, instructors, and students to absorb and apply valued concepts, principles, and strategies. To help minimize the negative impact of unrealistic expectations on teaching and learning environments, discussion of learner expectations, beliefs, and abilities should occur as early as possible amongst instructors and students in an academic language program,. This is important as failure to do so can send unintended messages to learners regarding the rate at which one is supposed to learn. As Fenton-Smith et al. (2017) have suggested, students need to become “disabused of the quick-fix myth and, instead, formulate long-term learning objectives” (p. 471). In addition, the results suggest that more direct discussion with students is needed (as early as possible) regarding program policies and how they may influence students after they leave the program. In this study, student participants struggled to understand how their course credits would be applied (or not applied) to their chosen program of study and were confused about what prerequisites they needed to complete. More discussion of program policies (at various institutional levels) would help to mitigate the potentially negative impact on students’ transitions out of the program. However, students enrolling in these programs must also be encouraged (and enabled) to  111  ask questions as early as possible and to seek out multiple sources of information. Students need to become aware that the wider social context surrounding and motivating these programs plays an important role in the decision-making processes of upper-level stakeholders. This is a continually negotiated and often contested process that transpires largely out of view of the students. Nevertheless, these decisions will directly impact them. Efforts to maximize clear and transparent two-way communication can greatly assist students in navigating program policies that will shape their educational trajectories.  There are indeed many benefits to considering the social context surrounding and informing multiple stakeholders in academic language programs. In this study, I focused on multilingual learners’ changing perceptions and experiences over the course of their studies and included aspects of their immediate socioeducational context. However, this research was limited in terms of how much of the wider sociocultural context could be incorporated. Future research should aim to investigate the wider social, cultural, and historical aspects of academic language program design and delivery. In this way, we can come to a better understanding of the complexities involved in assessing and responding to multilingual learner needs in English-dominant academic language programs, as well as the importance of open, transparent, and socioculturally-engaged communication amongst all program stakeholders.  Benesch (1996) noted over two decades ago in relation to the field of EAP that few researchers had “critically analyzed academic content and teaching” nor “encouraged students to examine issues that affect their academic lives and future careers” (p. 730). The need for this is perhaps greater than ever given the theoretical and pedagogical sophistication informing academic language instruction and the profusion of academic language programs vying for international student capital in global education contact zones around the world.    112  Chapter 5: Multilingual writers’ disciplinary socialization in a sheltered undergraduate language program   5.1 Introduction   Academic writing has long been promoted as a foundational survival skill in English-medium higher education contexts (Flowerdew, 2016; Hyland, 2013). As Flowerdew (2016) has recently commented, academic writing “performs an important gate-keeping role, effective writing leading to success and ineffective writing to failure in the academy” (p. 6). Its perceived importance has led to the establishment of foundational year writing programs in many U.S. universities (often made a requirement for all undergraduate students), is required on internationally-administered English language tests such as the Test of English as Foreign Language (TOEFL), and has inspired a plethora of guidebooks (and accompanying academic writing courses) that purport to meet the unique writing needs of specific groups of learners (e.g., de Silva Joyce & Feez, 2012; Swales & Feak, 2012).  Given this perceived importance, it is perhaps unsurprising that so much attention has been devoted to the study of written academic discourse involving multilingual writers, most notably in the fields of second language (L2) writing (e.g., Belcher & Braine, 1995; Leki et al., 2008; Matsuda & Silva, 2005; Silva & Matsuda, 2001), English for specific purposes or ESP (e.g., Flowerdew & Costley, 2017; Hyon, 2018; Woodrow, 2017), and English for academic purposes or EAP (e.g., Bitchener, Storch, & Wette, 2017; Charles & Pecorari, 2015; Hyland & Shaw, 2016). Flagship journals have been established for each and a substantial corpus of research questions, foci, contexts, participants, and results have been developed associated with the teaching, learning, and assessment of multilingual writers. Scholars and researchers working within and across these fields have also drawn on work conducted by specialists in genre and rhetoric (e.g., Artemeva &  113  Freedman, 2008, 2015; Bazerman, 1981, 1988; Dias et al., 1999), advances made in systemic functional linguistics (e.g., Halliday, 2007; Martin & Rose, 2003), and increasingly, ethnographically-oriented research methods (Paltridge et al., 2016).  However, there remains an ongoing and sometimes contentious debate regarding how to best conceptualize, teach, and assess “academic writing” involving diverse multilingual learners around the globe (e.g., Benesch, 2001; Lillis, 2003; Tribble, 2017; also see Chapter 2 for additional discussion). Despite the proposed benefits of genre-aware and disciplinary-specific approaches to academic literacy instruction (Fang & Schleppegrell, 2010; Moje, 2008; Shanahan & Shanahan, 2008), some have questioned whether explicit instruction of such knowledge is practical or appropriate given the degree of student resistance, lack of institutional support, or level of instructor preparedness (e.g., Freedman, 1993, 1999; Hansen, 2000; Heller, 2011). Others have warned of the potential for these approaches to compel multilingual learners to accommodate to an “assimilationist model” of instruction (e.g., Jenkins, 2011).  Regardless of the various approaches, strategies, and techniques woven into an academic language program, students who enrol are expected to perform in ways deemed satisfactory to their assessors. Despite this reality, there remains a lack of research investigating how students respond to academic writing instruction over extended periods of time, especially after instruction has ended and pedagogical scaffolds are no longer in place (Fox et al., 2014; Kobayashi et al., 2017). Such investigations are especially relevant for numerous learners around the world currently transitioning into English-medium university contexts, many of whom have received most or all their previous education in non-English dominant countries.  In the present study, I investigate the perceptions and experiences of four multilingual writers (selected from a larger study involving six participants) who successfully made their  114  transition to a mainstream program of study in a Canadian university through a sheltered first-year academic language and content program. The foundational-year for-credit program included academic writing courses in one of two disciplinary Streams (Arts or Sciences). The following two research questions guided this study:    1. How do four successful participants characterize their experiences learning and performing academic writing in their language courses?  2. In what ways do participants report understanding and applying written academic discourse concepts from these courses in their written assignments? In undertaking this study, I aim to contribute to a deeper understanding of the diverse ways multilingual learners respond to academic writing instruction and the varied ways this mediates their complex transitions to new sociocultural environments. As Leki and Carson (1994) advised over two decades ago: We feel that as EAP writing teachers and researchers, we need to be making greater efforts to consult more with ESL students and former students about their needs and about the ways in which their EAP writing training articulates with cross-curricular writing demands. (p. 99) I believe the need for educators and researchers to meaningfully consult with multilingual writers is greater than ever given the range and sophistication of theoretical and pedagogical approaches now informing the design of academic language programs around the world.   115  5.2 Disciplinary socialization and multilingual writers Most students making the transition from high school to university face challenges in understanding and performing academic discourse. For many, this is a gradual transition that is far from linear or unproblematic. However, for high school students who have received most or all their previous instruction in a non-English dominant country, the challenges involved can be unique (Duff & Anderson, 2015; Kobayashi et al., 2017; Leki, 2007; Zamel & Spack, 1998, 2004). For this reason, several researchers have investigated socialization practices surrounding and informing multilingual writers who are going through this academic rite of passage, or what is sometimes referred to as the first-year experience (e.g., Brook et al., 2014).  Over the last few decades, there have been a number of in-depth investigations into sociocultural aspects informing the production of written academic discourse in higher education contexts and its relationship to multilingual writers’ successes, failures, and abilities to negotiate legitimate academic or writer identities (e.g., Anderson, 2015; Bronson, 2004; Casanave, 2002; Ivanič, 1998; Morton et al., 2015; Prior, 1998; Séror, 2008; Spack, 1997; Zappa Hollman, 2007a). These developments reflect a sustained desire to better understand how academic discourse is shaped by the historical, social, and cultural context in which it is produced and how this, in turn, shapes teaching, learning, and assessment practices. While conducted across diverse educational contexts and focused on a range of literacy practices, these longitudinal case studies provide valuable sustained glimpses into the kinds of challenges multilingual writers experience in response to the instruction and feedback they receive from more experienced others.  For the remainder of this section, and in the interests of space, I highlight qualitative studies of particular relevance to my study because they: 1) focus on multilingual learners making transitions into English-medium undergraduate contexts, 2) have included consideration of how  116  academic writing practices surrounding and informing participants mediated these transitions, and 3) have privileged the perspectives of multilingual writers and how these change over time.  Spack (1997) traced the experiences of Yoko, an undergraduate student from Japan, over three years in an American university. Yoko’s first-year experience was mediated by her previous learning experiences in Japan and her perception of the “Western” communicative practices she was being expected to perform. Yoko described her “Japanese style” of writing as “being reserved,” which was not “respectable” in her American university (p. 14). Despite a challenging first-year experience in which she dropped courses and considered quitting, Yoko was able to negotiate the reading and writing expectations of her courses in subsequent years. However, her academic trajectory was not predictable nor linear. Yoko’s “theories” about her own academic literacy development (or lack thereof) shifted over the course of three years, from initially attributing her difficulties to her language background and lack of cultural experience with “Western” educational practices, to later viewing her academic literacy developing because of meaningful engagement in “knowledge construction” across a range of disciplines and genres (p. 44). While Spack initially viewed Yoko’s acceptance of these practices in her first year as “accommodationist,” this perception changed in subsequent years as Yoko asserted agency by adjusting her reading strategies (e.g., reading for gist first), proactively seeking guidance from knowledgeable others, and strategically choosing her courses and program of study. Yoko’s improved reading strategies and deepening disciplinary and genre awareness enabled her to negotiate diverse writing tasks with greater facility. While Spack is careful to avoid making cross-cultural comparisons or generalizations, she points out that Yoko’s “perception of her first-language educational background cannot be ignored because it influenced her approach to learning in a second language and shaped the way she theorized about that learning (p. 47). However, the way in which this  117  mediated her perceptions and responses to academic writing practices was unique and not reducible to a cultural caricature representing the “essence” of a Japanese student (p. 54).  Zappa-Hollman’s (2007a) multiple case study investigated six multilingual exchange students from Mexico as they adjusted to their classes in a Canadian university. She focused on a broad range of academic literacy practices (e.g., reading strategies, group work, networking, etc.), but also reported extensively on her participants’ perceptions and experiences with academic writing. Among other findings, she reported that students struggled to adapt to culturally different academic writing expectations, some referring to this experience as a kind of “torture” they needed to survive (p. 113). Some of the main writing challenges they experienced included meeting format and length requirements, writing with a voice, adapting to the critical analysis expected, and responding appropriately to unfamiliar instructor feedback and grading practices. In conjunction with “overwhelming” reading challenges, they experienced a “cognitive overload” that mediated their affective reactions to instruction and the feedback they received. This impacted their ability to perform academic writing in ways that were valued and rewarded. However, by gradually developing their awareness of the differing academic cultures they were traversing, students “became agents of their own socialization by choosing to adjust to, resist, or ignore these practices” (p. 102). Despite this increasing sense of agency, Zappa-Hollman concludes that the “hasty manner” in which students were expected to learn and perform complex academic literacy practices was “problematic since, as L2 language socialization research shows, the process of becoming familiar with new practices and of gaining access to new academic communities takes time” (p. 114). These results highlight the complex and dynamic intercultural processes involved and how instructional content and pedagogical practices can become overwhelming for students “in spite of their advanced LI and L2 academic literacy proficiency” (p. 6).  118   Séror’s (2008) eight-month multiple case study investigated feedback practices involving five multilingual learners from Japan as they completed academic writing assignments in undergraduate content courses in a Canadian university. He focused on the “socializing power of feedback” that students received in and through completion of their various writing assignments (p. 5). In addition to more macro institutional and ideological forces shaping feedback practices, Séror identified several aspects from the students’ perspectives that led to confusion and frustration. For example, students reported unfamiliarity and inexperience with the diverse writing tasks required of them, in expressing their thoughts and ideas within diverse disciplines and genres, and in expressing themselves in writing under time constraints (e.g., impromptu written responses in exams). Students also valued the feedback they received on their written assignments as a valuable source of “tips and hints,” but expressed a preference for timely, easily understood, and applicable feedback. While grammar correction was appreciated when it was not the primary focus of feedback, students preferred engagement with the content of their writing since it represented “real communication, a real exchange of ideas with their professor” (p. 79). This influenced their changing conceptions of “who they thought they were and what they felt they would have to do or be able to write for courses” (p. 116). However, their idealized view of instructor feedback was not what participants experienced most often, which was often characterized as sparse, unintelligible, grammar-focused, or emphasizing “deficient linguistic skills” rather than “the validity of the ideas” (p. 123).  There is now a robust body of research available that explores various aspects of academic literacy (e.g., oral academic presentations, group work, classroom discourse, etc.). However, longitudinal qualitative studies that focus on multilingual writers’ transitional experiences remain scarce. While there have been useful studies investigating the efficacy of various pedagogical  119  strategies or interventions in relation to academic writing (e.g., Crosthwaite, 2016), very few studies track learners over longer periods of study, and especially after they have left the instructional context. In addition, very few studies privilege multilingual learners’ perceptions of written academic discourse in the context of their academic language courses, and how (or whether) these perceptions change over time.  5.3 Description of study This is a multiple case study (Duff, 2008, Yin, 2009) involving four first-year undergraduate university students as they transitioned into a specialized academic language and content program with the hope of entering the university mainstream. The participants in this study are part of the larger dissertation study but were singled out for focused analysis in this chapter because of their overall positive perceptions of the program and their ability to successfully complete all program requirements. This first-year, for-credit program included disciplinary-specific language and content courses in one of two main disciplinary Streams (Arts or Sciences). All four participants were enrolled in the same program and all had the aim of entering the university mainstream in their second year after completing program requirements. This enabled continual cross-reference of participants’ perceptions and experiences while they were learning and performing written academic discourse since students in each disciplinary Stream often shared the same courses, instructors, and writing assignments. The program also provided an opportunity to investigate the role of disciplinary-specific language instruction in mediating participants’ perceptions and performances of written academic discourse. Since program curriculum included both content and language instruction, it also offered an opportunity to investigate results across inter-connected classroom contexts, many of which were shared by participants.    120  Theoretically and methodologically, I draw on academic discourse socialization or ADS research (Duff, 2010; Kobayashi et al., 2017; Morita & Kobayashi, 2008), an emerging sub-discipline within language socialization theory (Bayley & Schecter, 2003; Duff & May, 2017; Duff & Talmy, 2011; Duranti et al., 2012; Zuengler & Cole, 2005). ADS researchers investigate how newcomers become socialized (to varying degrees) into the academic discourse practices of a larger group, a process which is multidirectional, negotiated, contested, and continually evolving over time and place (for more information on this approach, see Section 2.3.3). 5.3.1 Research context For the purposes of this study, the program has been given the pseudonym Oasis College (OC) and the host university in which it was housed, Main Stream University (MSU). MSU is a large research-intensive university in Canada with a highly-diverse university student population. OC was in its inaugural year of operation at the commencement of this study. This program offered prospective students credit-bearing first-year undergraduate credits that could be applied towards their mainstream course of study, provided they were able to successfully complete OC program requirements. In this study, courses were classified as “language” courses (as opposed to “content” or “other” courses) when curriculum focused on the appropriate and effective expression of academic or disciplinary discourse. Based on this operational definition, three academic language courses were identified in both disciplinary Streams. These courses have been given the following pseudonyms: LANGUAGE, WRITING, and RESEARCH. These courses were taken by student participants in their first two terms of the program (the program consisting of three terms).  These labels are intended to characterize the primary purpose of these courses based on my analysis of the descriptions provided by participants, relevant curriculum documents (e.g., course  121  syllabi, handouts, quizzes, etc.), and my classroom observations. To frame the ensuing discussion of participants’ descriptions and experiences, some description of these courses is needed. Below, I describe the main aims, common tasks, and some of the instructional concepts guiding each course. Course names and numbers are fictional and are included only to indicate the academic term in which the courses were taken. For example, a course designated as 100 was delivered in the first term, 101 in the second, and 102 in the third. In OC and MSU, there were a total of three terms each year, the third being in the summer. LANGUAGE 100-101  This course provided “sustained language support for linked content courses” intended to “advance students' disciplinary literacy through heightened understanding of the links between language and the construction of disciplinary knowledge” (from course syllabi). Classes consisted of separate tutorials, each consisting of disciplinary-specific language instruction and language awareness tasks that were explicitly linked to content courses. Written assignments for this course included periodic quizzes, self-assessments, reflections, and other short writing tasks completed in class. There were no extended assignments for this course and there was limited homework assigned. Based on my discussion with the instructor, curriculum documents collected, and my observations of these classes, I noted there was a high degree of metalanguage used, an explicit instructional approach, and largely controlled writing tasks (from research notes).  Metalanguage included functional language terminology informed by the field of SFL (e.g., nominalization, theme and rheme, verbal processes, etc.) which were explained through direct instruction and then practiced on worksheets or through individual or collaborative language tasks (e.g., locating language forms in text, sentence completion, phrase transformation, etc.). For example, in one of my observations of a 50-minute LANGUAGE course delivered in the Arts  122  Stream, students listened to a 15-minute lecture describing aspects of causal language (e.g., explicit conjunctions, implicit logic, etc.). For the next 25 minutes, students completed a three-page worksheet that included several language tasks: locating language features in text, matching terminology to definitions, describing the function of words or phrases from the text, identifying parts of sentences, or paraphrasing using targeted language features. The instructor aided individual students as necessary and, in the final 10 minutes, elicited answers or provided them when no answers were offered.    This course was structured around weekly themes that included general academic skills (e.g., vocabulary-building, note-taking, in-text referencing) often focused on written academic discourse concepts (e.g., packing and unpacking nominal groups, theme and rheme, verbal processes). According to one of the language instructors for this class, the pace of this instruction was quick and the content quite challenging for the students. This was confirmed in my observations of these classes in which I noted the structured format of the lessons, the amount and complexity of the content discussed, and controlled writing practice of targeted academic features (e.g., nominalization) that were challenging for me to understand (from research notes).  WRITING 100-101 This course was intended to prepare students to “write academically in English” by “being taught to read and analyze their writing and that of experts,” which was intended to increase “understanding of the language features and purposes of writing in the university” (from course syllabi). Writing tasks within this course included three main short compositions that focused on extended definition, data commentary, and problem-solution. For each assignment, students completed multiple drafts for which highly-detailed online feedback was provided. The final version submitted for each short writing task required students to provide examples of revisions  123  they had made and short reflections on why they made them. Within the Arts Stream, this course was delivered over two terms within the OC program (Sept-Apr). However, within the Sciences Stream the shorter written tasks for this course were completed in the first term only (Sept-Dec). For the second term, Sciences Stream students started a new WRITING course in which they designed and carried out their own survey study. They completed individual sections of their research report throughout the term (i.e., proposal, introduction, method, results, and discussion). They also received formative and summative feedback on drafts submitted for these sections.  RESEARCH 100-101 These courses were sheltered versions of mainstream first-year undergraduate courses that all MSU students were required to take. Within the Arts Stream, the primary aim was to “study and practice the kinds of research and writing you will encounter in the Faculty of Arts” in order to “analyze how different styles of writing arise out of different research contexts” and participate in the “scholarly conversations that define these contexts” (from course syllabus). Writing tasks consisted of summary writing based on required readings and additional sources students located on their own. In the second term, students completed a research report that required them to design a research plan, conduct data analysis, present their findings to the class, and write a research report. Within the Sciences Stream, the primary aim was to provide students with the opportunity to “explore what science is and how it is done” and to “construct and critically assess scientific arguments” (from course syllabus). Writing tasks included multiple in-class (and timed) responses written and submitted on paper. For this, explicit prompts were provided a day or two beforehand so that students could prepare for their response. Students were given written feedback on their in-class writing and were required to submit a typed final version. In addition, students completed a term paper on a chosen topic that integrated scholarly sources.   124  In all three courses, participants were exposed to several academic writing concepts and expected to complete of a range of writing tasks intended to scaffold their understanding and performance of written academic discourse within their disciplinary areas. It is not possible to list and describe all the concepts informing these courses as they are too plentiful, but those listed below represent some of the more complex terminology that appeared in curriculum documents or in feedback received on students’ assignments.    • Purpose; controlling genre; sub-purpose; embedded genre; topic; interactants; evaluation; formality; organization • Finding the gap; occupying the niche; move structure • Nominalization: packing and unpacking ideas to organize texts • Modifiers in nominal groups (pre-modifiers, head nouns, post-modifiers) • Theme and rheme; thematic progression (zig zag, fan, etc.) • Interpersonal positioning; interactional resources (hedges, boosters, attitude markers, self-mention, engagement markers, etc.) • Types of expository writing; hortatory and analytical expositions • Evaluating things, ideas, and people (attitude, modality, and graduation) • Processes (material, mental, attributive relational, verbal, existential, etc.) • Representing ideas (processes, participants, circumstances, etc.) • Strategies to strengthen stance by using the voices of others  It was in these three courses (LANGUAGE, WRITING, and RESEARCH) that participants were explicitly and implicitly learning how to write, although this was not all they were learning (e.g., locating and evaluating sources, effectively communicating with peers and instructors, giving academic presentations, etc.). However, all three courses were (to varying degrees) intended to help develop students’ academic and disciplinary writing through instruction and practice.   125  5.3.2 Participants The focal participants for this study are four multilingual students who enrolled in the inaugural year of the OC program. Chloe and Yvonne were enrolled in the Arts Stream and Dawn and Theo in the Sciences. Within these Streams, participants took many of the same courses together and had many of the same instructors. All participants except one (Chloe) had arrived from their home countries in the month before starting the program and all but one (Dawn) had attended an international high school. Chloe and Theo had direct experience with Canadian-designed curriculum prior to enrolling in OC.  Yvonne was born in Russia and moved to China prior to commencing primary school. She reported being a “foreigner” in China and for this reason not having to prepare for the same national university entrance exam (Gaokao) as students born in China. She indicated she became dissatisfied with her education in China and, with the advice and support of her parents, intended to study abroad in an English-medium university upon completion of high school. She reported that instruction in her international school was largely lecture-based with minimal participation from students. There were also a lot of tests that she had to prepare for and take monthly. She described her grades as “in the middle,” which she was satisfied with. She did not know much about the OC program prior to arrival as most of the arrangements were taken care of for her. She was aware of the need to complete additional academic language courses and that it would enable her to enter the university mainstream if she successfully completed program requirements.  Chloe had already been in the country for over a year while attending a Canadian high school and was living with her mother in a nearby city. She was born in South Korea and completed all her primary education in the public-school system. After two and half years in two different high schools in South Korea, she transferred to a Canadian high school. She reported  126  being unhappy at both high schools she attended in Korea because it was either too competitive or too unruly. She indicated that studying in Canada would help her find a better balance in her education, and not be so focused on preparing for tests. She indicated she was happy to come to Canada because it meant she would not have to prepare for or take the national university entrance test (Suneung), which she characterized as “unfair” and “so subjective” (Chloe, 1st interview, term 1). She received one and half years of high school instruction in Canada before entering OC and reported struggling in making an adjustment to a Canadian high school at first. However, with the assistance of “kind teachers” and “native speaker” friends, she reported feeling much more comfortable in her final year of high school.  Dawn attended a state-funded boarding school in China that she described as a typical high school which was “more strict than international school” with a lot of homework (Dawn, 1st interview, term 1). There were “very big tests” every month in all her subject areas and students always knew their ranking in relation to others. She received English instruction exclusively from Chinese teachers, whose skill level she described as “not that high” because of a lack of English proficiency and their tendency to rely heavily on test-preparation methods (Dawn, 1st interview, term 1). She described this instruction as “always focused on grammar” with little attention to content. She primarily wrote short one-page English compositions based on a series of exemplars (in various subject areas) designed to score well on the Gaokao. Although she did not describe these as based on a 5-paragraph structure, one of the exemplars she showed me on her laptop was structured as such (from research notes). She reported the atmosphere of her classes as “not very lively” because it was largely teacher-fronted with little groupwork, presentations, or active student participation. Dawn decided to study abroad because, in her words, “I didn’t want to waste my four years in Chinese university” (Dawn, 1st interview, term 1). She also reported knowing very little  127  about the program as her application was prepared by her teacher-advisor in high school in consultation with her mother.  Theo attended a private international high school that incorporated “Canadian” curriculum modelled on the “kind of courses that people [in Canada] study” and which helped him “absorb a Western way of education” (Theo, 1st interview, term 2). He described a range of activities and writing assignments that involved group work, learner choice, community involvement, and creative thinking. He reported not preparing for the Gaokao in his home country because he planned to study abroad, as did everyone else in his high school. He stated that he wanted to “expand his horizons” and reported feeling that his mind would be constricted if he went to university in China. His decision to study in Canada was influenced by the “Canadian” curriculum of his high school and he became aware of the OC program because a recruiter visited his school. He indicated he was interested given that this program would enable him to enter a well-respected Canadian university even though he was unable to obtain the minimum English test scores required for direct entry. He reported being active in the application process and trying to find out as much as he could about the program and the university before arrival.  5.3.3 Data collection and analysis Primary data collection methods for this study include: 1) multiple interviews with student participants; 2) students’ voluntarily-submitted written assignments, and 3) observations of participants’ classrooms. A total of 22 interviews were conducted with student participants over the course of the study (5-6 each), representing 25.5 hours of dialogue or an average of 1.2 hours per interview. The initial interview was conducted as early as possible in participants’ first year of study, and subsequent interviews were conducted at least once per term at participants’ convenience. One additional interview was conducted with Yvonne (Arts) and Dawn (Sciences)  128  because they expressed an interest in further discussing their experiences. Interviews were guided by probing questions based on students’ term of study (see Appendix D), but room for elaboration, clarification, and critical analysis of topics of discussion was also given. All interviews were transcribed for qualitative analysis. To better contextualize student perceptions and reported experiences in interviews, I asked participants to submit academic writing assignments of significance to them including outlines, drafts, and instructor feedback if possible. I also asked students to consider any specific aspects of these assignments they would like to discuss further in interviews. All participants submitted at least one assignment they wanted to discuss before every interview and in most cases, there were several. A total of 83 assignments of various lengths and types were submitted. Since written assignments were voluntarily submitted, there was considerable diversity in number and types of assignments collected from the participants. Chloe (Arts) submitted 30 written assignments, Yvonne (Arts) 24, Theo (Sciences) 15, and Dawn (Sciences) 14. I also collected curriculum documents related to these assignments (e.g., course syllabi, handouts, etc.) and conducted 16 classroom observations of content and language courses representing over 22 hours of instructional time (for additional details on data collection and analytical procedures, see Section 3.5).  5.4 Results In the design of this research, I privilege the learners’ perspectives. Data collection techniques are intended to provide deeper insight into how participants described, reacted to, and reported applying explicitly-instructed written academic discourse concepts in their writing assignments. I am less concerned with “performance” in terms of grades received and whether these improved over time, although this aspect is included when information was available and deemed relevant. Before discussing participants’ perceptions and performances in more detail, I report on  129  commonalities observed across most or all the participants. In relation to their participation in the OC program, all four participants reported overall positive experiences that were beneficial to them in multiple ways. Upon successful completion of the program, they all reported that the academic support provided to them was extensive and helpful. For Chloe and Theo, whom I stayed in touch with into their fourth year of undergraduate studies at MSU, this perception became stronger after they entered the mainstream and much of the support they received in the first year was no longer available. For example, Theo stated that he got “VIP service” at OC in terms of academic support, but that this support was “totally different” in the mainstream (Theo, 4th interview, term 4). These results, albeit generated by a small sample of focal participants in this study, are in line with other studies that have reported on the overall positive benefits of ensconcing academic English instruction within the university as much as possible (e.g., Fox et al., 2014; Keefe & Shi, 2017) and the potential benefits of sheltered ESL instruction in enhancing the performance and engagement of undergraduate-level multilingual learners (e.g., Brook et al., 2014). In their initial interviews, all four participants expressed their belief that they lacked the necessary academic English skills to succeed at the university level, particularly in relation to academic writing and reading. For Yvonne, Theo, and Dawn, this conviction was based in part on their low scores on large-scale standardized tests such as the Test of English as a Foreign Language (TOEFL) or the International English Language Testing System (IELTS). Compared to the others, Theo and Chloe expressed more confidence in their English abilities, a belief based on the preparation they received in their respective high schools, both exposed to teaching and learning practices they described as “Canadian.”  Throughout the study, all four participants reported positively on the high quality and dedication of most of their OC instructors. This perspective was often connected to the close  130  attention instructors and TAs paid to students’ language learning needs, including reducing their rate of speech and avoiding numerous pop culture references or elaborating on them when they were discussed. Some students also reported positively on the sense of community that developed because they had a chance to get to know each other and their instructors well over the course of the year. All participants also reported enjoying and benefiting from a student-led interdisciplinary conference held near the end of the program, Chloe and Theo later describing this experience as one of the highlights of their first-year experience.  The perceptions and experiences reported below focus more exclusively on participants’ interactions within their academic language courses in relation to written academic discourse socialization, and therefore should not be interpreted as representative of their overall perception of the program or their experiences within it. Since each participant’s transitional experiences were unique, they are described separately. Discussion of similarities and differences across participants and their disciplinary Streams is taken up in the next section. This discussion is guided by three themes identified as salient in participants’ written academic discourse socialization: a) reactions to explicit instruction and controlled writing practice b) responses to instructor feedback on written assignments c) understanding and applying academic writing concepts  5.4.1 Yvonne (Arts) Early in the program, Yvonne reported appreciating the pedagogical approach taken in her LANGUAGE and WRITING courses and that these courses were necessary for her to improve her academic writing abilities. She also reported that her RESEARCH course was helpful because she was learning how to summarize articles, something she indicated would be required of her in future writing assignments (1st interview, term 1). After completing her first term in OC, Yvonne reported  131  that the explicit instruction and detailed feedback she was receiving in her WRITING course was helpful, particularly in revising her writing at the sentence-level: If we are supposed to revise something, I think I prefer [WRITING] feedback. When I write the revised version, I felt it easier to revise with that kind of instruction. So, I know like what the problem with the sentence is. (2nd interview, term 2)  Her attention to learning these concepts was noted in my observations of her WRITING class in which she took notes and participated actively in classroom tasks. Also, in her LANGUAGE class she kept all the handouts and worksheets handed out and reported reviewing them outside of class. By the end of the second term, Yvonne reported a rudimentary understanding of some of the concepts being taught (e.g., nominalization, theme and rheme, hedges and boosters, and text patterns); however, she remained confused about several others (e.g., interpersonal positioning, implicit and explicit logic, and verbal processes).  In written assignments we discussed from her first term, Yvonne was able to demonstrate how instructional concepts from her RESEARCH course were being applied in her written assignments (e.g., outlining, collapsing notes, and varying reporting verbs). For example, in a research paper she wrote in her second term, she indicated several examples, some of which are reproduced below (applied concepts are highlighted): If a state wants to present itself as a peaceful, strong state, the host state might sanitize its history during the performance. (RESEARCH 101, term paper, introduction, para. 1)   132  My first move is to identify whether these groups with civil conflict history are presented during the Olympic Opening Ceremony (RESEARCH 101, term paper, introduction, para. 3)   Taiwan was written out of history during the Beijing Olympic opening ceremony. (RESEARCH 101, term paper, discussion, para. 2)  In the first case, the metaphorical expression “sanitize history” was suggested by her instructor. Yvonne liked the expression, understood its metaphorical use, and decided to incorporate it into her paper (3rd interview, term 3). She noticed the expression “moves” in assignment instructions. Although she was unaware that the term was not commonly used in research reports, she demonstrated her understanding of its rhetorical function and used it appropriately. She also learned “written out of history” from her instructor, which she used in her paper because it captured her intended meaning well. She also provided examples of strategies she had learned and applied (e.g., annotating readings and condensing notes).  In her third term, while taking her first mainstream course, Yvonne’s reported beliefs about the necessity and value of explicit instruction and controlled writing practice changed. She still affirmed that her academic language courses were useful to her, but she stated this was largely because these courses required her to read extensively. She reported spending a considerable amount of time reading and trying to more deeply understand the concepts and ideas being written about. For Yvonne, completing her academic writing assignments was less a matter of understanding, noticing, and performing rhetorical structures and grammatical forms (the focus of her LANGUAGE and WRITING courses) and far more a matter of meaningful engagement with concepts and ideas, something she stated her RESEARCH and OC content courses helped better  133  prepare her for. Yvonne reported believing that this deeper understanding enabled her to integrate the words and ideas of others into her writing in more meaningful ways.  In her second year, after completing all her academic language courses successfully, Yvonne reported that her RESEARCH courses were more helpful than LANGUAGE and WRITING courses because “you also learn how to read some materials, and how to find the important parts” (6th interview, term 5). This was particularly helpful in her content courses, which she characterized as reading intensive. She described writing tasks in her RESEARCH courses as more authentic and meaningful than those in WRITING and LANGUAGE courses, which she characterized as more “repetitive” and “boring” (5th interview, Term 4). She appreciated having an opportunity to write about her own understanding and reaction to course materials in her RESEARCH courses, something she indicated improved her comprehension of content and gave her more confidence to write.   Yvonne reported that her content courses at OC were most helpful because they provided instruction in foundational concepts at a pace that was tailored to the language needs of learners (e.g., reduced rate of speech, elaborating on concepts, and avoiding or explaining cultural references). She also gained experience completing a range of writing assignments (e.g., opinion pieces, collaborative reports, timed-writing exams, etc.), many of which she was required to do in her mainstream content courses. However, reflecting on her academic language courses, she reported there was too much redundancy in her WRITING and LANGUAGE courses, something that reminded her of her experiences preparing for the Scholastic Aptitude Test in high school. She came to believe that the time she spent learning and applying academic writing concepts at the sentence and paragraph level could have been spent more productively doing applied tasks.   134  During Yvonne’s first term in MSU, she shared her weekly journals with me for her HISTORY 201 course, including the feedback she received from her TA. On her first graded journal, her feedback focused exclusively on grammar and vocabulary with no substantive feedback. The summary feedback she received was as follows: You need to address quite a few issues here. Your English needs serious improvement. Some of your sentences do not make sense. Please seek help at the writing services centre or with an ESL supporter. You make some good observations, but you did not communicate them very effectively. I cannot assign higher grades to your assignments when your writing is not at the university level. 60% (TA feedback given to Yvonne in HISTORY 201, term 4) When we discussed this feedback in a subsequent interview, Yvonne said she felt it was “unfair” because “it's actually not that hard to understand what I'm trying to say” (5th interview, term 4). In our 6th and final interview conducted near the end of her fifth term (her second term in MSU), Yvonne reported feeling much more appreciative of the efforts of her OC instructors and TAs (6th interview, term 5). Reasons for holding this belief included the faster pace of her mainstream lectures, fewer elaborations of concepts by mainstream instructors, increased pop culture references, unclear assignment instructions, reduced academic support, and less detailed feedback. She indicated that by participating in the OC program (as opposed to direct-entry), it helped to better prepare her for the academic challenges she experienced in the mainstream.  5.4.2 Chloe (Arts) Unlike Yvonne, Chloe expressed dissatisfaction with her LANGUAGE and WRITING courses from the outset and maintained this perspective throughout the study. She did not believe  135  that her academic writing was improving because of the explicit instruction and controlled writing practice she was receiving. The complexity of concepts being taught and practiced in these courses, combined with a lack of time available to meaningfully understand and apply them, also contributed to her negative perceptions. She reported not having experienced this style of language instruction in the Canadian high school she attended prior to entry and described it as more characteristic of the test-preparation style of instruction she experienced in South Korea.  After completing her first term at OC, Chloe questioned the usefulness of the feedback she had received on her LANGUAGE tasks and WRITING assignments. The feedback provided was extensive, but Chloe indicated that much of it focused on grammar and sentence-level concerns and very little was related to content. While she tried to understand and apply feedback from these courses to her revisions, she reported often not understanding what she was doing or why.  Yeah, I don’t really recognize that my writing is improving because I get instruction, like the help from [LANGUAGE] or [WRITING], because they don’t give like real practice for writing, you know … You know it’s not really academic writing, well it’s academic writing, but you know it’s not an assignment for real. (2nd interview, term 2)  In contrast, Chloe reported the approach being taken in her RESEARCH course was more effective in developing her academic writing abilities because of the “real practice” she was getting. She reported that the feedback provided in this course was from a “reader’s perspective rather than just assessing” (3rd interview, term 3). For her, this feedback engaged her thinking processes (e.g., generating ideas, making logical connections, finding appropriate support, etc.). In interviews, Chloe was able to demonstrate her understanding of genre-based concepts explicitly taught in her RESEARCH course (e.g., creating research questions, establishing a territory, occupying a niche,  136  etc.), but reported that these strategies were challenging to apply (2nd interview, term 2). However, she indicated that other concepts and strategies she learned in this course were helpful in her content courses. For example, she started to focus more on her use of reporting verbs (e.g., states, suggests, observes, etc.) when citing sources in her content courses.   Chloe indicated in several interviews that she was not confident in understanding and applying several instructional concepts from her WRITING and LANGUAGE courses (e.g., nominalization, theme and rheme, interpersonal positioning, verbal processes, etc.). For example, in the second term, she describes her lack of understanding of some of the concepts recently taught and practiced: Do you know the process, like with the verb, you need to kind of, umm, identify like … umm, identify like attributive relational process, identifying relational process, material process, and verbal process? You need to distinguish, like that’s the thing in LANGUAGE and WRITING, for those two courses they really wanted all the students to practice … and I don’t know why we need to actually distinguish those things. (2nd interview, term 2) In our fifth interview, conducted after she completed one term in the mainstream, Chloe indicated she was not able to consciously apply many of the academic writing concepts she had practiced in her WRITING and LANGUAGE courses because they were too advanced for her, and in her opinion for many others taking the course: John: So what did [WRITING] and [LANGUAGE] focus on in your opinion? What were they trying to do?  137  Chloe: They're trying to do like it's, you know when we learn grammar, but it doesn't really teach grammar itself, but it's kind of teaching how writing is composed. Like the components, like clause and phrases or paragraph. John: Like analyzing paragraphs? Chloe: Yeah it's more like analyzing, but I think it would be helpful when we actually get to some degree when we actually can see and have the ability to see and analyze more without problems, you know. It was a first-year course and most of them didn't really learn this kind of thing. They're not really ... Well I wasn't even really familiar with English paragraphs and things, and then they're like okay analyze them! I don't even know what you're talking about! Yeah. Unlike Yvonne who did not often submit assignment or course grades, Chloe contributed all her grades for her first year of study at OC. Table 5.1 lists all courses taken (pseudonyms) separated by course type (OC Language courses, OC content courses, OC other courses, and MSU courses). Individual and course averages by course type (e.g., OC content courses) are also included.  As can be seen, Chloe generally obtained higher grades in her OC content courses than her OC language courses. Also, overall course averages for OC language courses were lower than OC content courses, suggesting several students in these classes were struggling to achieve a passing grade. It should be noted that in the MSU grading scheme, students must achieve at least 50% or better to pass; however, in OC, students must achieve 60% or better. It’s also interesting to note that despite Chloe’s strong preference for her RESEARCH courses, it was her lowest grade (70%). Despite the challenges she experienced in her WRITING and LANGUAGE courses, she maintained a very favourable view of the OC program and appreciated the dedicated efforts of her instructors.   138  Table 5.1: Chloe's first-year grades Course (pseudonym) Chloe’s grade (%) Chloe’s grade by type (%) Course avg. (%) Course avg. by type (%) OC language courses     LANGUAGE 100-101 76  74 (B) 57 54 (D) WRITING 100-101 76 52 RESEARCH 100-101 70 54 OC content courses     GEOGRAPHY 100 85 80 (A-) 69 67 (C+) POLITICS 100 78  61 PSYCHOLOGY 100 82  61 GEOGRAPHY 101 80  74 POLITICS 101 76  67 PSYCHOLOGY 101 80  69 OC other courses      I-DISCIPLINE 100-101 78 77 (B+) 66 64 (C) PROJECT 102 75 62 MSU (mainstream)     ANTHRO 102 78 78 (B+) 66 66 (C) OVERALL 78  63  *60% required to pass OC courses, 50% required for MSU courses   Arts Stream students were permitted to take one MSU course in the summer term of their program and Chloe elected to take an introductory course in anthropology (note: this option was not available to students in the Sciences Stream). This was a large lecture-style classroom composed of approximately 100 students. Chloe achieved 78% (B+) in this course but was disappointed she was not able to get an A- (80%). In my observations of this class, I noted that the instructor’s rate of speech was quite fast at times and there was a lot of participation from some students. According to Chloe, there were several other OC students in the class, but few participated openly, including her. Instead, Chloe reported that she contributed to the class by answering questions posted by the instructor on the online discussion board.   139  In an extended discussion with the instructor for this course after it had finished, the instructor expressed frustration at not having been told there were OC students in the class. When we discussed the potential of the OC program, the instructor indicated it was a “bad idea” because it created difficulties for learners and instructors, particularly in courses like hers that depended on active participation. The instructor also indicated that it was “cruel” to many students who fail, sometimes with very low grades (one OC student in this course received 20%). The instructor also indicated that it “feels like people are buying their way in,” which is unfair to others. Interestingly, while not knowing who the participants in my study were, the instructor singled out Chloe as an OC student who received the top mark among all OC students in the class. Despite not speaking in class “even once,” the instructor indicated Chloe worked very hard in the course, participated actively on the discussion board, and put a lot of effort into her written assignments. However, the instructor indicated the same was not the case for most other OC students (from research notes). Chloe continued to stay in touch with me after the end of my data collection period and we sometimes met for coffee on campus to discuss her progress in the mainstream. She was also willing to share her transcripts for the courses she was taking. In her second year, she achieved a 79% average across all her mainstream courses. In her third year, Chloe achieved an average of 80% across all her courses and, for the first time in her undergraduate studies at MSU, achieved 90% (A+) in one of her courses. For Chloe, the transition to the mainstream was successful and she reported feeling confident in her ability to graduate within four years. She reported believing her writing was continually improving through applied practice. However, she did not credit the explicit language instruction and controlled writing practice she received in her LANGUAGE and WRITING courses as contributing much to this success. In her view, these courses did not offer enough pedagogical value for the additional workload, cognitive load, and stress they created for  140  her and others, as well as the potentially negative consequences this created for students unable to pass these courses (e.g., failing the program and not qualifying for the second year in MSU). Overall, Chloe reported that the OC program enabled a smoother transition for her than would have been possible had she entered MSU directly (5th interview, term 5). After the end of her third year of undergraduate study, she anticipated completing her BA in her fourth year. I asked her to write a written reflection about her first-year experience at OC. A portion of this reflection is provided below.   As an Arts student, one thing I appreciate from [OC] is that they made me read a lot of scholarly articles and got me to conduct my own scholarly research and to perform a presentation in a conference as a novice scholar. It was really a challenging process as a first-year student who just started out my university life. But I believe the hardships and stressful environment that I encountered certainly helped lay groundwork for the rest of my university years. (Chloe, written reflection in her fourth year) 5.4.3 Dawn (Sciences)  In our first interview in term 1, Dawn expressed her desire to improve her English academic writing and indicated that this would be her main challenge in completing her undergraduate degree. Dawn reported being very attentive in her academic language courses and working hard to understand and apply the writing concepts being taught and practiced. Early in the program, Dawn identified the concepts of “theme” and “rheme” as especially useful because they enabled her to stay focused on the main idea, thus making her academic writing less “messy” (1st interview, term 1). She reported that by learning and applying these and other concepts, she was improving her understanding and performance of academic writing. Unlike Chloe in the Arts Stream, Dawn  141  indicated that what she was learning was not the kind of “grammar” she learned in her “strict” high school in China. Instead, “here we focus on the idea, how you express it” (1st interview, term 1). In her second term, Dawn reported being very satisfied with her academic language courses and her instructors. She indicated she was making progress in her academic writing because of the explicit instruction and controlled practice of academic writing concepts, the range of assignments she was assigned, and the extensive feedback she received on her assignments (2nd interview, term 2). She characterized her RESEARCH courses as providing the most beneficial feedback as it was extensive and focused on various aspects of her writing, including grammar, vocabulary, structure, formatting, and content. She also reported that her WRITING course was helping her make logical connections in extended text: Before I took [WRITING], I didn’t know how to write academic papers and I could only use some descriptive expressions, everything started with ‘I’, ‘We’, or ‘The’. I didn’t have many logic connections between sentences, it’s my weakness. But after I take this class, I fixed a lot of my problems (2nd interview, term 2). Dawn reported following assignment instructions very closely and relying on her feedback to make revisions on written assignments:   Dawn: Most [of the revisions] are from feedback and sometimes I will go to see the instructors and they may point out some changes. John: Did you make any changes on your own because you felt, “Oh I want to change this now?” Dawn: I know they will give us feedback, I just wait for the feedback and then I will change.  142  In several interviews in which we discussed some of her written assignments, Dawn was able to identify concrete examples where she consciously applied instructional concepts. For example, she was applying “zig zag, fan, or parallel” text patterns to help structure her paragraphs (3rd interview, term 2). These patterns described how clauses and sentences within paragraphs could be structured differently. In the zig zag pattern, “information at the end of each clause is taken up as the theme in the next clause,” whereas in the fan pattern, “new information is presented early in the text and this is taken up in separate points as the text unfolds,” and in the parallel pattern, “information presented early in the text is repeated in different variations” (de Silva Joyce & Feez, 2012, pp. 7-8). Dawn was able to demonstrate her understanding of several academic writing concepts taught and practiced in her language courses such as hedging, active and passive sentences, and theme and rheme. However, she reported challenges understanding concepts such as nominalization, interpersonal positioning, verbal processes (e.g., material, mental, verbal, etc.) and others. Unlike Chloe, and to a lesser extent Yvonne in the Arts Stream, Dawn reported being highly receptive to the instruction and feedback she was receiving in these courses and she did not express any frustration with the instructional concepts explicitly taught and practiced. When I asked her to express her own views on learning and applying these concepts, she explained that it represented an “academic” way of writing, a way of writing she needed to learn (3rd interview, term 2). In the third term, after successfully completing all her courses up to that point, Dawn indicated that the WRITING course was the easiest to fail for many students. Therefore, she was happy to pass the course, even though it was her worst grade (67%). When I asked her about this course again in our sixth and final interview, she offered the following explanation (6th interview, term 5):    143  John:  Why was it so hard, do you think? Dawn: I don't know, the standard they used to grade might, maybe, also apply to locals, English speakers, and then maybe they used that standard to grade us. But we still have a lot of difference between the locals. Probably that's why we didn't get a good grade. While applying concepts she understood in a largely direct fashion during her first two terms, the more Dawn read the more she became aware of how these writing techniques can be applied strategically. For example, while she had been trying to use theme and rheme in her writing as much as possible, especially when revising, she reported “breaking the rule” because she noticed this rule was not always followed in her required reading (4th interview, term 3). Dawn’s reported beliefs about science and how to write scientifically did not change throughout the study. She indicated that “science is about the truth and we can’t add too much personal opinion in science articles” (4th interview, term 3). For this reason, she avoided self-mention (e.g., I, we, us) and used passive sentences as much as possible in writing assignments she perceived to be scientific (e.g., a research report). In several assignments submitted to the study, I noted her tendency to write passive sentences and avoid self-mention unless the assignment was a reflection or personalized response (e.g., journal). For example, in her 1224-word term paper for RESEARCH 101, in which wrote about the possibility of extra-terrestrial life, she incorporated thirteen sources and avoided any reference to her own position until the very end.  During her summer term at OC, Dawn reflected on her experience at OC as contributing to her academic and disciplinary writing development. She stated she “didn’t know how to write an academic essay before [OC],” something she attributed to not attending an international school (5th interview, term 3).  144  I was in public school in China and they didn’t teach us a lot about academic writing. They only teach us about descriptive writing. Right now, what impresses me most is that when you’re writing some essay about some report or experiment, try to be selfless.   Dawn chose to major in computer science in her second year, in part because she did not do as well as she hoped in some of her content courses and indicated there was still much she did not understand well. She also reiterated that she was still not good at writing and thought it wiser to avoid it as much as possible if she was to successfully graduate. In her first term in the mainstream, she reported doing very little writing in her courses. She was pleased with this, but worried she was forgetting everything she learned about academic writing at OC. In discussing her experiences after one term in MSU, she reported that she had more opportunities to speak English and more time to do other activities like volunteering. However, she reported a far more impersonal experience in the mainstream than at OC: In [OC] courses, we don't have like 300 people sitting in the same room, you have more chances to speak out, to ask questions or something like that. And now [in the mainstream] every course I have, there are more than 100 people sitting together. It's a big classroom. It doesn't care about you. (6h interview, term 5)  Reflecting on her experience in OC, after completing one term in the mainstream, Dawn reported that the academic writing instruction she received throughout the program helped her “form a good habit to write” even though she had limited opportunity to apply what she had learned and practiced (5th interview, term 4).   145  5.4.4 Theo (Sciences) Like Yvonne (Arts) and Dawn (Sciences), Theo reported being very receptive to learning the academic writing concepts being taught and practiced in his WRITING and LANGUAGE courses. However, like Chloe (Arts), he was not fully convinced of their usefulness after one term of study. While he stated these concepts were “important ideas to have in our minds when we write scientific research papers,” he also stated that writing was not a matter of following “patterns” but rather “a piece of art that we can elaborate on our own” (1st interview, term 2). When describing his understanding and application of these academic writing concepts, he remarked, “those things actually kind of help me” but “I don’t deliberately think of those things, I just write fluently, like water flows from a tube” (1st interview, term 2).  As did Chloe, Yvonne, and Dawn, Theo reported positive experiences in his RESEARCH courses, which he characterized as focused more on ideas, “rather than nominalization, rather than those details,” and the course was beneficial to him because “you had to really talk to your peers and share your opinions with them” (1st interview, term 2). He reported applying academic writing strategies taught in these classes and was able to demonstrate this application in assignments we discussed in interviews. For example, he learned how to follow an “I” structure, which was described on a handout he received in class and which he shared with me: The vertical bar of the ‘I’ represents the main body of the essay where each of the points presented in the development part of the thesis, is presented and discussed. Examples and references are generally included in these paragraphs. Remember that each paragraph contains one idea and examples or references that justify it. In terms of internal structure, its shape is a smaller ‘I’. (RESEARCH 101, handout)  146  In his second term, Theo reported reading assignment instructions very carefully and applying academic writing concepts in his academic language courses as required to maximize his grades.  Nominalization, theme and rheme, structure, those things are kind of criteria, that is how instructors mark our work. So, if I don't follow those requirements, I might get a bad mark. So in that way, I have to follow. So, I just deliberately, every time I write, I just think of those concepts in my mind. (2nd interview, term 2) At first, Theo was “uncomfortable” with it, but he reasoned that, “Yeah, maybe this is the way that university will require us to do and I have to accept the fact that I need to learn those new techniques” (2nd interview, term 2). His conceptualization of “scientific” writing was also changing in response to instruction and feedback he was receiving in his RESEARCH course, which sometimes confused him regarding whether and how much he should include the first person in his writing.    Theo reported not consciously applying these concepts outside of his academic language courses in the first term, but this changed by the end of his second term. For example, he reported applying nominalization strategies while completing a writing task in his physics class because it was important to be “concise” and “condensed,” something that nominalization could help with as it provided “a chance to compact sentences in a meaningful way” (2nd interview, term 2). For the first time, he also reported thinking about other academic writing concepts taught and practiced in his academic language courses in the process of writing assignments (e.g., hedging, theme and rheme, finding a gap, and establishing a niche).  Like Dawn, Theo also expressed similar disciplinary perceptions in relation to scientific writing. He characterized scientific writing as objective, depersonalized, and focused on what can be proven as fact. For example, in discussing hedging and boosting, Theo stated that he “hardly  147  ever” used boosters because in “scientific writing, you have to be objective, so words like those are not favored by instructors” (2nd interview, term 2). He also indicated that scientific writing, as he understood it then, “confines” and “suppresses our imagination” and was a style of writing “no people like to do” (3rd interview, term 3). Despite this, he stated that this was a style of writing he needed to learn to perform well in the mainstream. For this reason, he stated he attended to the instructions given and tried to satisfy instructors’ expectations as much as possible.     Interestingly, Theo decided to major in an interdisciplinary field (cognitive sciences) that required him to complete courses in various disciplines (e.g., psychology, linguistics, computer sciences, etc.). When Theo decided upon his major near the end of his first year, he described himself as “no longer really a pure science student” (3rd interview, term 3). In his first term in the mainstream, he proclaimed himself to be “both an arts and science guy” (4th interview, term 4). He reported that he felt more comfortable writing in ways appropriate to different disciplines and genres. He reported believing that the explicit instruction and practice he received in his academic language courses helped him to make these distinctions. Examples he provided of how he altered his disciplinary text include adjusting the degree of nominalization, hedging, and self-mention included to alter “how much of myself I put in my writing” (4th interview, term 4). After completing one term in the mainstream, Theo (like Dawn) reported that the amount of writing he was doing in MSU courses had reduced markedly compared to OC. Instead, group projects, quizzes, and timed exams with multiple-choice and short answer questions became far more common. Nevertheless, he reported consciously thinking about and using many of the concepts he had learned in OC to inform the limited writing he was doing.  Theo contributed all his final grades for the language and content courses he completed at OC, as well as interdisciplinary courses that included lectures, a group research project, and a  148  conference presentation in the summer term. Table 5.2 lists all courses taken (pseudonyms) separated by course type (OC Language courses, OC content courses, OC other courses). Individual and course averages by type are also included.  Table 5.2: Theo's first-year grades  Course (pseudonym) Theo’s grade (%) Theo’s grade by type (%)  Course avg. (%) Course avg. by type (%) OC Language courses     LANGUAGE 100-101 74   73 (B) 64 65 (C+) WRITING 100-101 62  57  RESEARCH 100-101 83  74 OC Content courses     CHEMISTRY 100 63  70 (B-) 62 69 (B-) PHYSICS 100 79  72 MATH 100 74  68 CHEMISTRY 101 65  71  PHYSICS 101 76  76 MATH 101 61 68  COMPUTER 102 72 68 OC Other courses      I-DISCIPLINE 100-101 87 88 (A) 71 71 (B-) PROJECT 102 88 70 OVERALL 74  68  *60% required to pass OC courses  As can be seen, Theo’s average grade across all three OC language courses was slightly higher than his OC content courses. Compared to Chloe (Arts) who obtained higher grades in content courses than language courses (see Table 5.1), there was also greater variability in Theo’s grades. In his language courses, Theo did quite well in RESEARCH (83%), achieved a satisfactory grade in LANGUAGE (74%), but barely passed WRITING (62%). Theo also struggled in several OC content courses, achieving higher grades in physics, but struggling with some of his math and chemistry courses. However, he did exceptionally well in other OC courses (I-DISCIPLINE and  149  PROJECT) that encouraged inter-disciplinary understanding, collaborative research projects, and presentations of research findings.  As with Chloe in the Arts Stream, I was able to stay in touch with Theo as he transitioned further into the mainstream. He continued to report that the academic writing instruction he received at OC was helpful in making his adjustment to MSU. Even though he did not use all the information, he was able to use some of that knowledge to take more control over his writing. Like Chloe, he also shared his transcripts for all the courses he took in his second and third year. He achieved an average of 69% in his second year and 67% in his third year in MSU. Theo reported being satisfied with his performance because he was equally concerned with other important aspects of his education, including socializing, running an MSU student club, developing professional contacts, and volunteering in the community.  In his third year, Theo discussed his experiences working on an independent-study project for a linguistics course. He decided to research his use of nominalization in his writing assignments and concluded that the explicit instruction and controlled practice he received in his LANGUAGE and WRITING courses in the first year had helped him to take more control over his disciplinary-specific writing in his third year. When we last met, he informed me that he expected to graduate at the end of his fourth year.    5.5 Discussion and conclusion While participants’ transitions into MSU were unique, they also shared several experiences by virtue of being in the same university, first-year program, disciplinary Stream, and often the same course. As a group, it is notable that all four participants reported an overall positive perception of the OC program, their OC content courses, and their OC instructors, a perception that appeared to strengthen after they entered the university mainstream. All students also reported  150  positive perceptions of their RESEARCH courses because of the extensive content-focused feedback received and writing assignments perceived to be more authentic than those in other academic language courses. For the remainder of this chapter, I contrast participants’ changing perceptions of their LANGUAGE and WRITING courses within their disciplinary Streams before discussing some implications and pedagogical suggestions. Chloe and Yvonne (Arts Stream) reported more negative perceptions of LANGUAGE and WRITING courses than Dawn and Theo (Sciences Stream). They also reported challenges in understanding and applying many of the academic writing concepts they were learning and practicing in these courses. Chloe became increasingly disengaged with what she came to see as a didactic style of instruction, cognitively complex academic writing concepts, and an inauthentic writing context. Yvonne also reported difficulty in understanding and applying these concepts in the time provided given her overall workload. She remained more engaged than Chloe and more willing to try to apply what she was learning. However, she became more skeptical about its usefulness over time as she was not able to consciously apply much of what she was learning and practicing outside the context of these courses. For both, classroom writing tasks became largely an exercise of trying to satisfy assignment instructions and pass these courses to avoid the negative consequences of failure (e.g., not proceeding to the second year in the mainstream).    Based on our discussions of written assignments and supported by their reported experiences in their LANGUAGE and WRITING courses, Chloe and Yvonne did not develop much facility in applying many of these academic discourse tools to shape their written communication in their OC or MSU content courses. While Yvonne reported trying to apply some of them (e.g., nominalization, hedging and boosting, and text patterns), she later preferred to write in what she referred to as a “more natural way.” Yvonne attributed most of the progress in her  151  writing to the extensive reading required throughout the program which enabled her to better understand instructional concepts and integrate them in more meaningful ways. Chloe attributed her progress in writing to practical strategies, content-focused feedback, and meaningful discussions with instructors. Over time, Chloe and Yvonne became more convinced that LANGUAGE and WRITING courses were not useful to them because of the lack of freer writing practice and meaningful engagement with content. Both had also expressed a desire to escape what they perceived as similar instructional approaches in high school in their home countries. They expressed little faith in the efficacy of an explicit and controlled approach because it was intended to prepare students to do well on standardized tests, but not how to communicate in the real world. While there were no such tests in OC, their background experiences with this kind of instruction shaped their perceptions and reactions. Instead, Chloe and Yvonne responded more positively to writing instruction and practice they perceived as more authentic and feedback that engaged with the content of their work more than its form.  However, Dawn and Theo (Sciences) reported appreciating the explicit writing instruction and controlled writing practice they were receiving in their LANGAUGE and WRITING courses because it would help them improve their academic writing abilities. This was the case even though they did not understand the concepts deeply at the time. They also appreciated the structured form-focused feedback they received in these courses because it helped them revise their writing. Like Chloe and Yvonne in the Arts Stream, both attended to the feedback they received and revised largely to satisfy instructor expectations and program requirements. However, they also tried to apply these concepts outside of these classes to the extent that they could. As a result, their disciplinary awareness was enhanced in various ways as they progressed into the mainstream.  152  While in the mainstream, Dawn reported noticing that certain structures she had learned (e.g., theme and rheme) were not always applied by more experienced writers in her assigned academic reading. She indicated she was becoming more aware that instructional concepts and strategies were not prescriptive rules to be blindly followed, but instead tools that enable her to shape written academic discourse for different communicative purposes. For Dawn, this awareness took time to develop to the extent it did, but the lack of writing practice in the mainstream made it difficult for her to deepen this awareness. Theo was able to discern disciplinary differences in how written academic discourse was communicated across various subject areas, a realization that was further enhanced by his decision to pursue an interdisciplinary major and study his own use of nominalization in a third-year research project. Theo reported taking more control over his writing and more deeply understanding the disciplinary genre and audience for which he was writing.  In line with several other studies that have investigated academic discourse socialization involving multilingual learners in higher education contexts, the trajectories of all four participants in this study were non-linear, always negotiated, and unpredictable (Duff, 2010; Kobayashi et al., 2017; Morita & Kobayashi, 2008). The influence of power and culture were omnipresent in mediating participants’ perceptions of academic writing instruction and their responses to the feedback they received, but these were not the most challenging factors involved in their transition. For these participants, the complexity involved in what they were learning and in how they could apply it were far more influential factors. Throughout this process, participants were at times more active in this process (e.g., questioning, resisting, or ignoring) and at other times more passive (e.g., following instructions despite not understanding or agreeing with them). These results are in line with several other studies that have investigated the teaching and learning of written academic discourse involving multilingual learners and have discovered dynamic, contested, and  153  multidirectional socialization practices (e.g., Bronson, 2004; Casanave, 2002; Ivanič, 1998; Leki, 2007; Prior, 1998; Séror, 2008; Spack, 1997; Zappa-Hollman, 2007a).  The results of this study suggest that explicit instruction of academic discourse concepts, the cognitive complexity involved, and the time available to learn and practice are impactful when not aligned well with students’ perceptions, needs, or abilities. As Bronson (2004) discovered in relation to graduate-level learners, students were less successful when academic resources were insufficient to meet expectations and when they were unable to visualize a pathway to meet these expectations. As Zappa-Hollman (2007a) has discovered, academic writing can be viewed as “torture” (p. 113) by multilingual learners who may experience significant levels of stress when expectations are not well-understood and sufficient time is not made available to make complex linguistic, cultural, and social adjustments. This can lead to avoidance, resistance, or rejection as it did for some of the participants in this study. However, more importantly, it can render inert highly useful information and practice that has great potential to aid multilingual learners later in their disciplinary studies, as happened in the case of Theo and to a lesser extent Dawn.  In addition, the degree of authentic writing practice and the balance of content and form-focused feedback provided can become confounding factors in written academic discourse socialization for multilingual learners. As Zamel (1983) discovered many years ago in her multiple case study of the composing processes of six “skilled” multilingual writers, they “developed strategies that allowed them to pursue the development of their ideas without being sidetracked by lexical and syntactic difficulties” (p. 175). Séror’s (2008) study of feedback practices indicated there were difficulties in participants’ ability to respond to feedback when it did not meaningfully engage with the content of their assignments. However, all participants in the present study appreciated and engaged with “rich dialogic feedback” (Lillis, 2001) whenever it was available.  154  This brings up several concerns regarding the degree of controlled writing practice and form-focused feedback that is appropriate or effective. While this study cannot speak directly to the efficacy of various approaches in relation to learners’ language abilities, it does suggest that much further research is needed to ensure that the cognitive complexity involved remains aligned as much as possible with students’ perceptions of their language needs and abilities. As much research has shown, feedback that is not understood is not applied, and at worst frustrates or alienates learners (Hyland & Hyland, 2006).  These results also point to some of the challenges involved in integrating multiple theoretical and pedagogical approaches (e.g., genre-based pedagogy, SFL, CLIL, etc.) given the unpredictability of student response in localized contexts. However, they also point to some of the possibilities that integrating approaches has to offer for multilingual learners that may not be realized for some time after instruction and practice has ceased. More research is needed into the longer-term influence of explicit instruction and controlled practice on learners’ understanding and performance of complex academic writing concepts across disciplines, particularly those derived from socially-informed approaches to language (e.g., SFL). Essential areas in need of further investigation include: 1) to what extent multilingual learners are prepared for this kind of instruction, 2) what level of metalanguage and how much time is needed to achieve instructional aims, 3) what scaffolds need to be put in place to develop conceptual knowledge across diverse classrooms, 4) what kinds of writing practice and feedback can enable learners to meaningfully apply these concepts over time, and 5) what assessment strategies can best enable instructors to check understanding and provide additional (targeted) scaffolding when necessary.  Some of the limitations of this study should be discussed. First, this study focuses on the mediating influence of participants’ academic language courses. These courses, while impactful,  155  represent a partial view of their overall academic literacy socialization (Duff, 2007a, 2019; Kobayashi et al., 2017). While the focus of this study is on participants’ perceptions and performances of academic writing, there was a range of literacy practices they were involved in throughout the course of the study (e.g., group projects, academic presentations, field trips, social networking, etc.). Undoubtedly, their many experiences (both academic and non-academic) played into their perceptions and performances of academic writing. Although there was often discussion of these various activities in interviews and discussions, these were not directly observed.  In addition, while numerous written assignments were collected for this study, it was not possible to discuss all of them in detail with participants. While I encouraged participants to identify specific sections of assignments they would like to discuss, only Chloe and Theo regularly did so. Since assignments were submitted voluntarily, there was considerable variability in the number and types of assignments submitted. Also, some assignments included outlines, drafts, and instructor feedback, which provided additional context for some participant interviews but not for others. Also, some participants did not often include feedback or grades with the assignments they submitted. Finally, while the contributions of longitudinal multiple case studies such as this one are many, they are unique to the participants involved and cannot be (nor are they intended to be) generalizable. Notwithstanding these limitations, the results of this study provide theoretical and pedagogical insights that are likely to be relevant for many administrators, teachers, and students in similar contexts around the world. For the remainder of this paper, I discuss some implications, suggestions, and directions for future research.        As the results of this study demonstrate, it is vital to investigate changes in learner perceptions and experiences over longer periods of time. As Spack (1997) noted in her three-year case study, Yoko’s first-year writing practices appeared to be “accommodationist” as she struggled  156  to adapt to “Western” expectations. However, over time Yoko’s engagement with disciplinary content improved, especially through reading, and she began to re-conceptualize academic writing as knowledge construction which enabled her to take more control of her writing. These experiences align well with my interpretations of Dawn’s and Theo’s changing writing practices. Initially, they appeared to be passive in accommodating pedagogical practices. However, over time, this gave way to deeper disciplinary connections indicating, as with Yoko, they did not have to be taught to become “emancipated learners,” just be given the opportunity to be one. As Theo remarked in one of our last meetings, “we first have to learn the rules to be able to break them” (from research notes). Finally, these results suggest that more explicit discussion with learners about the longer-term objectives of instructional approaches and the written tasks they are assigned would help students appreciate their full potential over time. Less “coverage” of concepts and reduced use of metalanguage might help to make abstract ideas more intelligible to students and reduce the risk of overwhelming them. More authentic writing practice, even when controlled, can help students remain engaged and see the value in what they are learning. Academic language programs that serve a large and diverse student population are often delivered in a one-size-fits-all manner. However, the ability of such programs to adequately address diverse multilingual learner needs continues to be questioned (Cheng & Fox, 2008; Fox et al., 2014). For this reason, open, transparent, and socioculturally-engaged communication is necessary across multiple program stakeholders as much as possible. For those working or studying in similar educational contexts, diagnostic assessment of language proficiency (and possibly attitudes towards academic or disciplinary writing) could help in identifying students who might experience challenges with the style or complexity of instruction  157  and help develop academic support strategies more attuned to individual learner needs. A growing body of research now exists that endeavors to improve alignments between instructional practices and students’ academic needs through incorporating post-admission diagnostic language assessment (Read, 2015). Some researchers have pushed for assessment initiatives that are embedded within content areas and utilize local disciplinary knowledge and practices (e.g., Fox, Haggerty, & Artemeva, 2016; Palmer, Levett-Jones, & Smith, 2018). Such approaches, when adjusted to suit local contexts, help to identify students at potential risk and better inform targeted academic support strategies that can be presented as optional rather than mandatory. Such strategies may help to address some of the challenges multilingual learners experience in performing written academic discourse by raising awareness and discussion of disciplinary expectations while also differentiating and targeting instruction. One-size-fits-all approaches to writing instruction that are guided by unproblematized assumptions about what students need to learn may be expedient for organizational or economic purposes, but they can be quite inappropriate and ineffective for some multilingual learners (Canagarajah, 1999, 2013; Heng Hartse & Kubota, 2014; Horner et al., 2010). As Zamel (1997) cautioned over two decades ago, reducing multilingual learners to passive recipients of language subject to didactic and controlled approaches has the potential to alienate and disengage students from participating in meaningful academic discourse practices:  This reductive perspective on students and their work no doubt accounts for the ongoing tendency to teach and assign formulaic representations of academic discourse and models of discipline-specific discourses and to resist engaging students in the messiness and struggle of authentic work that begins, values, and builds on their own "ways with words." (p. 343)   158  Chapter 6: Agency, appropriation, and identities in first-year undergraduate writing in the Arts and Sciences  6.1 Introduction Many university students are unable to cope with the technical and scholastic demands made on their use of language as students. They cannot define the terms which they hear in lectures or which they themselves use. (Bourdieu, Passeron, & de Saint Martin, 1994, p. 4)  The above description of novice university students’ comprehension and application of academic discourse accurately captures, in my opinion, the experience of a great many first-year undergraduate students around the world, including myself many years ago. The language of the academy is historical, abstract, negotiated, contested, and continually changing in response to micro, meso, and macro social contexts (Bazerman, Bonini, & Figueiredo, 2009; Duff, 2007b, 2010; Hyland, 2009). This renders any attempt to establish ironclad and permanent academic discourse conventions across disciplinary genres fruitless (Artemeva & Freedman, 2008, 2015; Bazerman, 1981, 1988; Dias et al., 1999; Freedman, 1993, 1999). Such hegemonic conceptualizations of academic or disciplinary discourse also have the potential to punish “violations” of poorly-understood academic conventions or supposed “native” language norms and encourage a deficit discourse that hinders inclusion of some learners as legitimate speakers or writers (Heng Hartse & Kubota, 2014; Horner et al., 2010; Rose, 1998; Shi, 2010, 2012). Regardless of these challenges and dangers, students are generally expected to grasp complex meaning-making practices and incorporate new ways with words into their discoursal repertoires in ways deemed appropriate by their assessors. Through participation in the socialization practices of the academy, novice students are expected to develop ever greater facility communicating with  159  more experienced others and become (should they wish) accepted members of academic, disciplinary, or professional communities.  Most students in English-medium universities are likely to face some difficulties, particularly in their first year (Mann, 2001; Perry & Allard, 2003). This is because “learning the language of rational, abstracting, academic discourse and processes” can become an alienating experience for newcomers (Mann, 2001, p. 12). Students may experience intense dissatisfaction, isolation, or alienation, all of which can have negative academic, social, and personal consequences, or what some have described as identity discontinuity or loss (Scanlon, Rowling, & Weber, 2007). This transition can be especially arduous for those with less experience communicating in similar social, cultural, and linguistic contexts (Casanave & Li, 2008; Duff & Anderson, 2015; Hyland, 2012; Kobayashi et al., 2017; Morita & Kobayashi, 2008; Zamel & Spack, 1998, 2004).  For these reasons, it is necessary to investigate multilingual learner transitions while also acknowledging the backgrounds of students, the prevailing ideologies influencing the context, and any power differentials involved. There is also a need to privilege learner perspectives as much as possible to better understand the kinds of challenges they face and the complex reasons underlying their responses within these programs and the trajectories they take. This longitudinal study investigates the experiences of two multilingual students learning and performing disciplinary-specific academic writing in a Canadian university. These students are part of a larger study investigating six multilingual students; however, they became the focus of this study because they reported considerable confusion, frustration, and conflict in meeting academic and disciplinary writing expectations.  This study investigates the socialization of written academic discourse and privileges the learners’ perspectives. Of specific interest in this study is learners’ changing beliefs about the  160  academic language instruction they received in this program, their experiences performing academic writing in their various courses, their responses to feedback on their written assignments, and how these experiences enabled or constrained them in transitioning into the university mainstream. While there are undoubtedly numerous literacies required of university students in need of further investigation (Duff, 2010; Lea & Street, 1998), the importance of academic writing continues to be characterized as a foundational skill required for success (Flowerdew, 2016; Hyland, 2013). While much scholarly research attention has focused on academic writing in terms of textual products produced across various genres and disciplines or the cognitive processes involved in its production, there is still a paucity of research that explores, from the student perspective, the sociocultural context that surrounds and informs what multilinguals learners write and why they write it (Paltridge et al., 2016).  6.2 The socialization of written academic discourse Researchers have been investigating multilingual learner transitions for some time, drawing on various theoretical, pedagogical, and methodological orientations. While large-scale survey studies have provided valuable overviews of multilingual students’ perceptions in relation to the various programs they are enrolled (e.g., Barton & Donahue, 2009; Fox et al., 2014), several researchers have utilized longitudinal qualitative methods to focus on individual multilingual learners within their localized educational contexts and investigate how perceptions and practices change over time (e.g., Atkinson & Ramanathan, 1995; Belcher & Braine, 1995; Blakeslee, 1997; Braine, 1996; Bronson, 2004; Casanave, 2002; Godfrey, 2015; Harklau, 2000; Ivanič, 1998; Leki, 2007; McKay, 1993; Morton et al., 2015; Prior, 1998; Séror, 2008; Shi & Cumming, 1995; Spack, 1997; Zamel & Spack, 1998, 2004; Zappa-Hollman, 2007a). These studies provide rich and nuanced views of the successes and challenges that multilingual learners experience, how they react  161  to and contribute to the socialization practices occurring, and how this mediates their perceptions and learning trajectories over time. In the interest of space, I have restricted my discussion to studies that have most directly informed and guided me in carrying out this research. Bronson (2004) conducted a multiple case study of four graduate students transitioning into various disciplines in an American university, two from Japan, one from Chile, and one from Laos/Thailand. The researcher focused on instructors’ feedback practices and students’ ongoing responses to them as expressed in interviews and in their writing. Among many findings, Bronson reported that instructor feedback was highly idiosyncratic and shaped by institutional and ideological forces, which often left students not knowing what to do to improve their performance. For two of the students taking an anthropology degree, they were encouraged to continually “re-negotiate” and “re-align” their academic objectives and literacy strategies with those of their professors, influenced by the critical approach taken in the department. However, for two students enrolled in non-arts-based programs, there was “no such space for negotiation or consensus building,” which resulted in a “submersion situation, one without any accommodation to their educational needs as (advanced) English-learners” (p. 373). Student responses to feedback were influenced by their individual backgrounds, but because the socioacademic context in which they studied offered differential opportunities for scaffolding and meaningful literacy engagement, it had a direct impact on their sense of agency and identity within their respective disciplinary communities. Leki’s (2007) multiple case study investigated the long-term academic literacy development of four multilingual learners, two from China, one from Japan, and one from Poland, who were enrolled in a first-year credit-bearing ESL writing course in an American university. Leki tracked students’ perceptions and experiences after they completed the course and continued in their  162  respective disciplines (nursing, engineering, social work, and business). Among other findings, the researcher discovered a lack of explicit detail and authenticity in participants’ required writing assignments, which provided “few writing experiences that promoted their intellectual or disciplinary growth” (p. 235). Influential in mediating participants’ successful transition into their respective disciplines was the relationships they were able to mould with their peers and instructors, which were directly impacted by the institutional structures and ideological assumptions mediating classroom discourse and behaviour. Salient in her findings was the importance of acknowledging the “unequal power balance generally between newcomers and more experienced others” that may become magnified for multilingual learners in postsecondary settings (p. 275). Godfrey’s (2015) multiple case study focused on one instructor and two multilingual learners (one from China and one from South Korea) enrolled in an academic writing course in an American university. Godfrey investigated the instructional practices and pedagogical beliefs of the instructor and the various ways students responded to instruction and feedback on written assignments. Results indicated the instructor was instrumental in creating a “favorable site and space of socialization,” influenced by her personal, professional, and educational background (p. 167). The instructor’s sensitivity to the cultural and linguistic differences in the class and the treatment of all students as “worthy and capable” helped to foster a safe learning environment. Despite this favourable socialization space, the two student participants achieved quite different results, one, described as more accommodating, achieving an A, and the other, a self-proclaimed “loud girl” who preferred “random writing,” almost failing the course (p. 162). For Godfrey, this highlights “how socialization is not guaranteed and that learner agency to enact and resist socialization further complicates the socialization process” (p. 169). However, unlike the present study, the researcher did not privilege the (emic) perspective of the student, opting instead to focus  163  on the beliefs and practice of one instructor as the primary socializing agent in the course. Because of this, and the rather short duration of the study (one term), the researcher was limited in investigating students’ changing perceptions and practices, the degree to which they understood instructional concepts, or the common and unique ways they responded to instruction and feedback over time (e.g., avoidance, accommodation, resistance, etc.).  Morton et al. (2015) conducted a multiple case study involving three multilingual learners, two from China and one from Poland, who were transitioning into an Australian university. Participants were taking undergraduate content courses concurrently with credit-bearing EAP courses, although these courses were not linked (as in CLIL approaches). The researchers focused on participants’ “perceptions of academic writing and their development as academic writers” (p. 2) as they pursued programs of study in business, arts, or science fields. The researchers discovered that individual participants’ perceptions and trajectories were unique and dynamic, influenced by their previous educational experiences as well as the academic writing practices that occurred within their respective disciplinary areas. For one participant (Laura), who was in an arts-related field, academic assignments were longer and required more critical thinking and stance-taking compared with the other disciplines. As a result, Laura struggled to express herself in many writing assignments and grappled considerably with issues of identity. Her learning trajectory was impacted by the writing tasks she was assigned, especially those that “encouraged her to reflect upon her own views” (p. 9). The researchers conclude that participants pursuing business and science-related fields focused on “textual and rhetorical conventions of their disciplines with the aim of accommodating successfully,” while Laura adopted a “questioning, critical stance,” influenced by the academic discourse practices occurring in her arts-based program (p. 10).  164  Several other scholars and researchers have incorporated socially-informed theories and methodologies to explore the social context surrounding and informing the production of academic text (see Kobayashi et al., 2017, and Paltridge et al., 2016, for recent reviews). These studies have deepened our understanding of socialization practices and the vital role that language, culture, power, and identity plays in various contexts around the world. They have also inspired and informed the present study, which aims to contribute to deeper understandings of these multifaceted socialization processes.  6.3 Description of study This study is part of a larger longitudinal multiple case study (Duff, 2008; Yin, 2009) involving six multilingual writers transitioning into a Canadian university, given the fictional name Main Stream University (MSU). In this chapter, I focus on two multilingual learners who experienced considerable confusion, frustration, and conflict in making their transitions. In this study, I am interested in better understanding the complex reasons for their changing perceptions and performances of written academic discourse in their respective disciplinary areas. I focus on the degree to which they understood and applied the academic writing concepts they were being taught and some of the confounding factors that contributed to their inability or unwillingness to apply them. The following research questions guided this study: 1. How do two less successful students characterize their experiences learning and performing academic writing in their language courses? 2. How do social, cultural, or linguistic challenges mediate their transitions to their respective programs of study?   165  6.3.1 Context and participants Both participants, Ashlee and Jasmine, were enrolled in the same inaugural year of a specialized first-year undergraduate program, given the fictional name Oasis College (OC) for this study (for additional details about this program, see Section 3.2). Jasmine was enrolled in the Arts Stream and Ashlee in the Sciences. Both were nineteen at the start of the study and had arrived directly from their respective international high schools in China. (for additional details about these participants, see Section 3.3).  6.3.2 Theoretical and methodological approach This research draws on academic discourse socialization (hereafter ADS), an approach to language and literacy development that has informed a growing body of research into a variety of educational and processional contexts (Duff, 2010; Kobayashi et al., 2017; Morita & Kobayashi, 2008). ADS researchers are informed by Language Socialization (LS) theory in their investigations of “how newcomers to an academic community negotiate not only their academic and linguistic needs and expectations, but also their personal goals and histories, their roles and identities, as well as social, cultural, and historical aspects of their academic socialization” (Kobayashi et al., 2017, p. 9). LS/ADS perspectives share much in common with sociocultural theory in that “learning, knowledge-construction, and socialization – that is, the development of the human mind and the socialized individual – are seen to be processes that are mutually engaged in by members in a community over time” (Duff, 2007b, p. 312). An LS/ADS approach is also compatible with post-structural approaches to identity (e.g., Darvin & Norton, 2015; Norton, 2013; Norton & Toohey, 2011) that affirm "it is through language that a person negotiates a sense of self within and across different sites at different points in time" and that "it is through language that a person gains access  166  to – or is denied access to – powerful social networks that give learners the opportunity to speak" (Norton Peirce, 1995, p. 13). Post-structural notions of identity have informed the work of several scholars and researchers investigating the teaching and learning of academic discourse in multilingual contexts (Casanave, 2002; Harklau, 2000; Ivanič, 1998; Lui & Tannacito, 2013; Marshall & Walsh-Marr, 2018; Morita, 2002, 2004; Pavlenko & Blackledge, 2004; Prior, 1998). This is because, according to Duff (2010): Identity work and the negotiation of institutional and disciplinary ideologies and epistemologies are core aspects of the production and interpretation of academic discourse. Academic discourse is therefore a site of internal and interpersonal struggle for many people, especially for newcomers or novices. Considerable emotional investment and power dynamics may therefore be involved. (p. 170) Interest in the notion of identity and its significance in the production of academic discourse has inspired research across a wide range of disciplines. Of relevance for this study is research into academic discourse socialization involving multilingual learners, particularly studies that have focused on academic writing practices in higher education (e.g., Bronson, 2004; Godfrey, 2015; Séror, 2008; Zappa-Hollman, 2007a). In addition, post-structural conceptions of identity have also provided more socially and critically-relevant conceptualizations of the range of identity options available for administrators, instructors, and students who interact in these sociohistorical and sociocultural zones, and the complex and dynamic ways that socialization practices can enable or constrain (i.e., legitimate or censure) the identities that can be performed. These theoretical perspectives provide powerful analytical tools that enable researchers to move beyond strictly text-linguistic or cognitive conceptions to include a consideration of how individuals are positioned by  167  others, how they position themselves, and how this impacts who they believe they are and who they believe they can be.   The notion of agency has also been taken up by several scholars and researchers interested in better understanding the teaching and learning practices occurring in multilingual language learning contexts (e.g., Douglas Fir Group, 2016; Duff & Doherty, 2015; Gao, 2013; Larsen-Freeman, 2019). Larsen-Freeman (2019) has long been concerned about the portrayal of multilingual learners as “nonagentive” in the teaching and learning process and points out that several second language research agendas have implicitly represented learners in this way (e.g., universal acquisition order, comprehensible input, etc.). Instead, as the Douglas Fir Group (2016) describes, language learners must be seen as active agents in the socialization process in which their “social identities, subjectivities, and sense of agency are further significant to the development of their multilingual repertoires” and that “their growing repertoires and abilities will influence their identities, and their roles, rights, status, means, and agency” (p. 32). This symbiotic relationship is mediated by (macro, meso, and micro) social contexts and the inherent power relationships involved in all communication (Duff, 2019).  Agency has been defined in diverse ways, and conceptions overlap with other influential concepts in language teaching and learning such as investment, intentionality, motivation, and locus of control, among others (Duff & Doherty, 2015). Ahearn (2001) described agency as the “socioculturally mediated capacity to act” (p. 112) and warned of the need to define the term beyond being synonymous with “free will or resistance” (p. 130). More recently, Duff (2012) has characterized agency as “people’s ability to make choices, take control, self-regulate, and thereby pursue their goals as individuals leading, potentially, to personal or social transformation” (p. 413). These perspectives on agency provide a valuable lens to consider not only how structural conditions  168  and sociocultural aspects impinge on language learners, but also how learners exert a substantial influence on the social context and their own language learning trajectory, although not always in ways that are beneficial to them in the learning situation. As always, the ability to exercise control over one’s practices is interconnected with and directly influenced by the (more macro) sociohistorical and sociocultural context as well as the (more micro) social roles, expectations, practices, and power relations associated with the individuals involved.     6.3.3 Data collection and analysis Primary data collection methods include: 1) multiple semi-structured interviews, 2) voluntarily-submitted written assignments, and 3) observations of classrooms for which these assignments were written. A total of twelve interviews were conducted, six with each participant (avg. of 1.2 hours per interview). The first interview was conducted as early as possible after the start of their program and subsequent interviews were conducted at least once per term at the participants’ convenience (see Appendix D for guiding questions for each interview).  To better contextualize discussion in our interviews, I asked participants to submit writing assignments (including outlines, drafts and any feedback if possible) that they wanted to discuss further. Both participants submitted several assignments per term. A total of 31 assignments were submitted by Jasmine (Arts) and 33 were submitted by Ashlee (Sciences). It was not possible to discuss all assignments in detail in interviews, but they provided valuable artifacts that enriched our discussions and enabled us to locate examples in the text that relate to the various topics and themes being discussed. I also conducted ten classroom observations for which assignments were being written (see Appendix F for the observation sheet used), collected curriculum documents relevant to these assignments, and held numerous discussions with various stakeholders working in OC and  169  MSU for which research notes were taken (for additional details on data collection and analytical procedures, see Section 3.5).  6.4 Results  6.4.1 Jasmine’s delayed transition to the mainstream Jasmine went to an international high school in which she was taught by Chinese and Canadian teachers. She attended this school because she did not want to prepare for the Chinese college entrance test (Gaokao), something she described as a “crazy exam” that, in her opinion, negatively impacted the personalities of teenagers (1st interview, term 2). She reported an overall negative experience in her middle school, which she described as highly test-oriented with very few opportunities to communicate in English. She decided that she wanted to learn something new and, with her parents’ support, enrolled in a non-traditional high school that specialized in preparing students to study abroad. She reported completing a variety of writing tasks in her English classes in this high school, including shorter and longer compositions.  She indicated she had learned the “5-paragraph essay” format in her test-preparation courses, a structure she described as a “burger” (1st interview, term 1). She also indicated she was encouraged to use a thesaurus to improve her vocabulary. Jasmine reported very positive experiences throughout high school. She stated her Canadian teachers were like “best friends” to her and this made her feel very comfortable to communicate in English. Jasmine described herself as “one of the special ones” in her English classes because of her strong communicative abilities, particularly in speaking (1st interview, term 2). She reported that her experiences (and grades) greatly improved in this school (compared to her middle school) because of the additional “room” she was given:  170  [In high school] I really did much better than my junior performance, because I think I have more room to adjust myself and to find the weakness and fix them. And because I could focus on every assignment I got, I learned from them, I really learned a lot (1st interview, term 2) In her first term at OC, Jasmine described feeling challenged by the workload, but also believing that her academic language courses were useful. For this reason, she tried to “accept all the knowledge the teachers give me” (1st interview, term 2): I think if the teacher spends a lot of time on that, and tries to let us understand, that must be useful. Maybe I cannot get the point of that because I cannot understand it now, but if you keep going on, maybe you can catch some part of that and explore more. But if you hide from that, you will know nothing.  In her second term at OC, Jasmine described feeling more frustrated about her ability to understand and apply the academic writing concepts being explicitly taught and practiced in her language courses. For example, in our second interview she reported feeling “pretty scared” after learning about the concept of nominalization:  [Nominalization] was really hard, and then even after the class I learned about this, I was totally confused. I was more confused, I just didn’t know it and just pretty scared. Think about that, just after your class, you become more confused and not less confused. (2nd interview, term 3)   171  She reported that she learned the importance of nominalization in making writing “sound more academic” and was trying to incorporate this and other instructed academic writing concepts into her writing to “push” herself:  Sometimes, I will try to push myself to write in these ways that I’m mostly unfamiliar with. That's a hard process, because you are not used to writing that way, and everything comes out really slow. Sometimes your flow will be disturbed by that. You forget what your previous thought is, and I just ... I hate that because it's really hard. (3rd interview, term 3) In interviews, Jasmine was able to demonstrate a superficial understanding of some of the academic writing concepts being discussed in her language classes. Despite not understanding the concepts deeply, she was able to specify some areas in her writing she was applying this knowledge. For example, below are three examples from one of Jasmine’s writing assignments in the second term that she indicated (in bold) were attempts to incorporate more nominalization in her writing: • Under a serious situation of vast anthropogenic carbon emission and high energy consumption, nowadays, the world is facing an unprecedented problem …   • The expanding urbanization and industrialization also drove the need of commercial fuel constantly ascend, especially …  • In the view of positive contributions from the low-carbon revolution, more advanced evaluation was pointed out …       She reported learning and practicing how to transform sentences in various ways at the sentence level and then applying this strategy in her extended writing assignments for other courses. While  172  she was able to apply many of the concepts at a sentence level, she reported struggling to gain control over this in her extended writing.  In all our interviews, Jasmine was able to express her opinions on many subjects and comment substantively on several issues related to her courses and her experiences in China, in the OC program, and in Canada more generally. In my notes and in interviews, I often remarked on how impressed I was with her oral communication ability, and how this differed from many of the academic writing assignments she shared with me. Jasmine often reported considerable difficulty in comprehending abstract concepts from her lectures and assigned readings and then being able to communicate intelligibly about them in her writing assignments. This was a challenge in both her language and content courses. To compensate, Jasmine incorporated writing strategies she was learning in her academic language courses to improve her writing performances. This appears to have further complicated some of the difficulties she was having and negatively impacted her conception of herself as a writer and what identities she believed she could legitimately claim. At the end of her second term at OC, Jasmine was not able to pass one of her academic language courses, a result that was a surprise to her “because actually I feel good about the things I learned from that class” (4th interview, term 4). Because she was also unable to pass one of her content courses, she was informed she would not be able to enter the mainstream in her second year. Jasmine continued to take courses in the third (summer) term, including her first mainstream course in anthropology, which as an Arts Stream student she could take as an elective while still in OC. Jasmine reported that the pace of instruction was much faster and the class size much larger than most of her classes at OC. Nevertheless, she decided to do her best and improve upon her abilities. She continued to apply some of the academic writing concepts she learned such as nominalization, despite not yet having a solid understanding:  173  I cannot give a clear enough definition for [nominalization], but if you want me to describe for that, I think it’s a process to, maybe just kind of ways to use nouns or noun groups to paraphrase your sentence and make it shorter with the same meaning. (4th interview, term 4) In her mid-term written assignment for her first mainstream anthropology course, Jasmine struggled to understand the required readings and incorporate them into her writing. This impacted her ability to establish a clear focus for her mid-term take-home exam. Because Jasmine was unable to develop her conceptual understanding through lectures, she was unable to incorporate information from sources to guide and support her paper. As a result, she reported a lack of confidence in writing about course topics and incorporating readings appropriately. She stated she was dealing with two related issues: “I need to deal with English language at the same time as overcoming the fear of knowing deeply about that theory” (4th interview, term 4). She was able to pass, but just barely (52%). It should be noted that 50% or higher was required to pass MSU courses, but 60% was required for OC courses. Jasmine sought advice from OC advisors on how she could continue her studies and enter MSU in the future. This was a “dream” she was not willing to give up (4th interview, term 4). However, she struggled to understand what she was doing wrong and what she needed to do to perform better. Jasmine was advised by OC to enroll in another MSU-affiliated college on campus for her second year. She was told she could then enter MSU in her third year and retain some of the credits she had accumulated. The college she enrolled in for her second year was smaller than OC and included both domestic and international students. In the first term in this college, she reported that the pace of instruction and her overall workload was reduced, which was less overwhelming for her than her first year at OC. She reported benefitting greatly from having the opportunity to  174  communicate in small classes with domestic students and reported that this was very helpful in improving her understanding of course content. While she no longer took academic language courses, she was receiving one-on-one tutoring on her academic writing assignments from her instructors during office hours.  Because of this change in instructional style, exchanges with peers, and reduced workload, Jasmine reported feeling that her writing was finally starting to improve. She shared several assignments with me from this term and I noted that her use of nominalization had reduced considerably. When I asked her about this, she indicated that she was encouraged to write more naturally by her instructors and to focus first on being understood. One of the most important changes for Jasmine that term was the additional time she was given to develop her understanding of concepts and to work on being more explicit, organized, and connected, without trying to make it sound “crazy academic” with too many “complicated structures” (5th interview, term 4).  Jasmine characterized her experiences in her academic language courses throughout her first year in the OC program as a continual effort to meet expectations that were not well understood, and despite trying, constantly feeling she was missing the mark. This contributed to her inability to satisfy program requirements and delayed her entry to the mainstream. However, in her second year in another university-affiliated college that did not include required academic language courses, the slower pace, reduced workload, culturally and linguistically diverse context, and increased peer collaboration provided more “room” for her to focus on reading comprehension and clarity in writing, which gave her a sense of renewed confidence to write about concepts learned, organize the content in a more reader-friendly manner, and express her opinions.  When reflecting on her experiences at OC, she was still very appreciative of the efforts of her instructors and advisors to assist her in both academic and personal ways. However, in relation  175  to her academic language courses, her opinion changed somewhat. She stated that while the information may have been helpful, it “dragged her down” because “If I didn't take so many courses in the first year, I might do better” (6th interview, term 5). Because of what for her was a “rushed” academic language curriculum in OC, it became too complex for her to absorb in the time available:  The only problem is it's too rushed, and this is the only problem, but that is also the biggest problem. Especially the first-year student, everything they touch is all brand new, they need time to absorb it. (6th interview, term 5) Based on her willingness, I kept in touch with Jasmine for two additional years after our last interview to informally discuss her progress. In our meeting early in the summer of her second year, she informed me she was able to successfully complete all her courses at her second-year college. She was very happy with her experiences there and indicated that her academic writing skills had improved as result of the slower pace. She had applied to MSU and was looking forward to finally studying in the mainstream. Two weeks after this meeting, I received a frantic phone call from Jasmine explaining she did not qualify because she did not satisfy MSU’s direct-entry English language admission standards and would need to submit a recognized English language test score (e.g., TOEFL iBT, IELTS Academic, etc.). She informed me that she was planning to give up and return home to China. She was devastated that she had “wasted two years of [her] life for nothing” (from research notes).  Before giving up, I convinced Jasmine to go the registrar’s office with me to try to find out more about the decision and to appeal for a review. At the office, the clerk confirmed to Jasmine that she did not qualify and would need to submit a test score. Jasmine explained her academic history and that her college advisors had indicated she qualified. She was again told she did not  176  qualify. At this point, I interjected and asked to speak to someone else with more authority on these matters. After some time, another person came to discuss the situation with us. I explained what I knew about Jasmine’s academic history as I understood it. After this explanation, the person left to consult with others. After about half an hour, the person returned to inform us that Jasmine had in fact satisfied English language admission requirements and would receive a notice about the status of her application soon. Jasmine was understandably relieved and very happy to hear that all her hard work would not go to waste. We were never told the reason for this change in their decision. In our next campus coffee chat near the end of her second year, Jasmine informed me she was accepted to MSU and would start to take mainstream courses at the start of her third year. She decided to major in sociology because it allowed her to explore her interest in psychology as well as learn more about society. She continued to share her writing assignments with me throughout her third year and reported continuing to struggle with cohesion and coherence. However, in my reading of these assignments, I noted she improved in establishing a clear purpose for her paper, which was now explicitly stated in her introductions and signposted at various stages. She also displayed more facility in constructing complex sentences. For example, in a research report completed for her SOCIOLOGY 200 course, she provided the following statement of purpose at the end of her nicely-contextualized introduction: My aim in this research is to explore how long working hours influence the academic performance of fulltime students in Canada and how the situation is varied among international students compared to domestic students. Knowing the effect of in-term employment on full-time students in different populations could help the school and government to produce better policies that could bring more benefit to the students in the future (research paper, SOCIOLOGY 200, para 1).  177  In our last discussion, Jasmine reported that there was limited academic writing support available in mainstream courses. She reported challenges in understanding assignment instructions and in knowing what to do based on the minimal feedback provided by instructors and TAs. These were positive aspects she reported in the MSU-affiliated college and at OC, and something she was much more appreciative about after a few terms in the mainstream. Jasmine passed all her MSU courses, although she reduced her course load to four courses each term, a full course load typically being five. For this reason, it would take her longer than five years to complete her BA, something she had yet to tell her family. Regardless, she planned to finish her degree despite all the struggles she had faced. She stated her experiences made her a “stronger” and “well-balanced” person (from research notes).   6.4.2 Ashlee’s disciplinary conflict Prior to entering OC, Ashlee had attended an international high school that adopted a “British” style of curriculum which included a variety of courses designed to develop content knowledge and English language abilities. This was a private school intended for students who wanted to study in English-medium universities abroad. Similar to Jasmine’s high school, students did not prepare to take the college entrance test (Gaokao) which was needed to attend a Chinese university. Instead, Ashlee prepared for standardized English language assessments such as the International General Certificate of Secondary Education, the Cambridge English First, and the International English Language Testing System (Academic). She described the educational approach in her English courses as very structured and largely test-oriented. She indicated she did not like this style of English instruction. In her final year of high school, she reported doing more communicative activities (e.g., presentations and debating) and writing some longer compositions she described as “research papers, but not real research papers” (1st interview, term 1). Like  178  Jasmine, Ashlee learned the 5-paragraph essay structure and reports having used this structure often, especially in test-preparation classes. She indicated she struggled in her English courses and usually received low grades on her written assignments.  Ashlee was very interested in science, especially physics. She indicated that studying in an English-medium university would provide her a better education in this discipline than in China. She reported not knowing very much about OC before applying because this was taken care of by her school in consultation with recruiters with little information provided to her or her parents. However, Ashlee indicated she was aware that completing this first-year program would require her to complete English language courses in addition to content courses and that she would be able to use the credits achieved towards her degree. She also reported believing additional language courses were necessary, not only because of her inability to achieve a satisfactory score on her IELTS test, but also because “I am a science student and I am not good at language” (1st interview, term 1). Ashlee reported feeling overwhelmed with the workload early in the program as well as the complexity and pace of the academic language instruction she was receiving in the Sciences Stream. While she reported being able to understand most of the concepts being taught in her academic language courses and believing it to be useful, she was having difficulty applying this information.  In her second month in the program, Ashlee received feedback on her first writing assignment completed in one of her academic language courses. This was a first draft that required students to provide an extended definition of a scientific concept. She indicated she was quite confused about the feedback she received and what she should do to revise. I examined the feedback with her because it was only available online and required her university login information to view. The online feedback was quite complex to navigate, but it provided extensive  179  feedback that could be displayed multiple ways (e.g., normal view, statistics, feedback, score, all annotations, and marking key). There were text-embedded annotations that referred to short written comments provided at the end. There were also hyperlinks embedded in the text that linked to comments based on various categories, including those associated with structure (e.g., organization, formatting, etc.), meaning (e.g., content, logic, etc.) as well as form-focused components (e.g., punctuation, preposition, word form, verb tense, etc.). In addition, the number of instances in which a linguistic feature (e.g. subject-verb agreement, articles, etc.) was incorrect was indicated. A written summary comment was also provided that indicated it was a “good first draft that included the required stages,” that the “content is quite good,” and that “sequential explanation was interesting.” Areas for improvement centered on improving formatting, language issues, avoiding specialized vocabulary, and expanding in some areas. Her grade for this assignment was calculated as follows:   • Organization: Whole text level: 20/30  • Organization: Clause, sentence, paragraph level: 20/30  • Content: 15/20  • Interpersonal positioning: 0/10  • Presentation: 0/10 Overall: 55%  When I asked her about these assessment criteria, Ashlee indicated she did not know what interpersonal positioning meant nor did she understand why she received a zero on this component or the last one (Presentation). When I asked her what she thought she should focus on in her revisions, she reported not knowing what to do to improve other than to fix some of the “simple” language errors (e.g., verb tense, pluralizing nouns, etc.). When I asked if she was taking advantage of academic support provided, she stated, “I think I need more help here, in high school we can find  180  anyone that can help, but here we can only book 30 minutes [per week] for help” (2nd interview, term 2).   After completing her first term at OC, Ashlee expressed dissatisfaction and frustration with two of her three academic language courses. These courses included explicit instruction and controlled practice of several academic writing concepts (e.g., nominalization, theme and rheme, interpersonal positioning, etc.). Much of her dissatisfaction with these academic language courses was based on a perceived lack of connection to her content courses, redundancy of material within and across language courses, and instruction that was not related to “scientific writing” as she understood it. As a result, she reported feeling that these courses added to her workload but did not provide useful and relevant information for her planned program of study (physics). She received her lowest grades in both courses but reported being confused about her grades since she was following the instructions properly. In the assignments she submitted from her first term, for which she included feedback on her drafts, there was little evidence that she was understanding or responding to the feedback she received. In our second interview, when I asked her about some of the feedback on her assignments in the first term (as described above), she indicated she did not understand it, so she “skipped it” because there was not enough time to figure it out given her workload (3rd interview, term 2).   Despite her challenges understanding and applying these academic writing concepts, Ashlee was able to pass all her courses in her first term. However, in her second term in one of her academic language courses, the focus changed from completing shorter controlled writing tasks to an applied research project in which she was required to individually write a proposal, develop research questions, design a survey, analyze the results, and write a referenced research report. This term-long assignment was separated into smaller writing tasks that focused on sections of the report  181  (introduction, method, results, and discussion). These sections were written separately and submitted as drafts so that feedback could be provided by the instructor. In this course, Ashlee reported mounting tension, frustration, and conflict throughout the term as she tried to satisfy the requirements of the course.  There was also intertextual tension evident in Ashlee’s writing in this course and the feedback she received on her drafts. For example, Table 6.1 displays portions of the first draft of Ashlee’s introduction along with the focus of instructor feedback provided in the margin of this assignment and the grade she received.  Table 6.1: Ashlee's introduction (first draft) Section Ashlee’s text (first submission)  Margin feedback  Introduction  (from paragraph 1) More and more international programs set up recent years, and [Oasis College] is one of them. As a new program, [OC] need to concern if the students is suitable for the program. How students fit into their chosen program might be reflected by attitudes towards different streams, i.e., Arts or Science. Academic vocabulary, specificity, tense, subject-verb Introduction  (from paragraph 2) Also some students of [OC] are offered admittance automatically without applying as they were rejected by other faculties at [MSU]. Thus, the faculties in [OC] (i.e., Arts and Sciences) might not be preferable for some students. Then if the students suitable for the stream at [OC] becomes a question. “This is a lie,” inconsistent terminology, confusing Introduction  (from paragraph 5) In this research, I will do a survey on students’ feeling about both streams [OC]. Then the data will be analyzed in the aspect of attitude, and focusing on the stream. This will help understanding how students fit in the program, and how to improve the program.  Use of first-person, logical connections *No summary feedback provided Grade: 46.6% The introduction was written and submitted in Word and feedback was provided by the instructor using the comment feature. There were five paragraphs and 384 words in the complete  182  draft, not including three references. In the interests of space, only a portion of Ashlee’s text and the feedback she received is provided in the table. Extensive margin feedback was provided throughout this introduction, including advice to use theme and rheme (not shown in table), correct tense errors, improve logical connections, etc. In reaction to her claim that “some students of [OC] are offered admittance automatically without applying as they were rejected by other faculties at [MSU]” (Introduction, para. 2), the instructor remarked that this was a “lie”. In a later comment, the instructor questioned how she thought sentences are connected to one another, “by magic?” There were several comments expressing clarification, expansion, or confusion, but no advice offered on how she might address these issues. She was admonished for not following previous instructions and was advised to use more recent references (including a comment asking how old she was in 1985). One of her sources was marked as “fake” (I later investigated this source and discovered that it did indeed exist). In my review of her feedback, I noted only one positive comment that indicated there was a good idea, but also that it was unclear and disconnected. Nevertheless, she resolved to try to improve the paper because she needed to pass the course. However, she also stated, “I feel bored about this course and I feel bored about him. And it’s just like we are playing games and whatever I do, I just don’t want to care about it” (4th interview, term 3). For Ashlee’s second and final submission, she sought the assistance of another instructor she had for a course the previous term (term 1) who was someone she respected. She reported meeting with her previous instructor for over an hour as they worked on making her introduction better. After this meeting, the introduction was greatly revised and subsequently submitted. The opening two paragraphs are shown in Table 6.2 along with the feedback she received from her (second-term) course instructor. I noted several changes in terminology from her first draft such as  183  the use of hedges, removal of self-mention, as well as the focus of the study (indicated in bold). When we discussed the changes made to the introduction, Ashlee indicated that she accepted the changes recommended by her previous instructor on this assignment but did not understand them deeply. She reported that none of these changes were her choice and that most of the writing was “not really mine” (3rd interview, term 3). Despite this, her grade on the final submission was lower than her first draft (no margin feedback was provided).  Table 6.2: Ashlee's introduction (final submission) Section Ashlee’s text (final submission)  Margin feedback  Introduction (paragraph 1) Despite the perception that scholars are objective, in the academic world, scholars are typically deeply invested in their area of study. For university students, the level of investment in the chosen area of study has great potential for increasing the level of success. As reported by Richardson (2012), students’ attitudes have an effect on academic achievement. One group that is likely to experience challenges with investing in the discipline is first-year international students. First-year students would face challenges since they do not know all the choices of discipline (Schuitema, 2014) and international students can become disoriented as a result of difference in academic culture (Zhang 2014). This study seeks to understand the attitudes of first-year international students in two general academic areas (i.e., Arts and Science).  None Introduction (paragraph 2) The purpose of this study is to better inform stakeholders, including students, parents, instructors and administrators about students’ attitudes to their chosen academic area. A specific purpose is to help students understand their attitudes and adopt positive strategies for succeeding in their studies. We investigate these in a first year alternative entry program for international students; the program is called [OC]. This is an entirely new program that is still flexible in terms of how to help prepare students learn in and appreciate their chosen area of study.  None  *Summary comment focused on lack of attention paid to previous feedback and questioning why an entirely new introduction was written.  Grade 45.3%   184  Throughout our interviews, Ashlee expressed firm opinions on the nature of “science” and commented on the irrelevance of her “non-scientific” academic language courses. When we discussed one of her academic language courses that focused on how to think about science and express oneself appropriately in writing, she stated, “I don’t think it’s a science course even though it’s called science” (5th interview, term 4). She indicated that her academic language courses were guided by “social science” assumptions which were decidedly more subjective in her view. As she stated, “Like we are supposed to have a science program, but actually none of us are doing science because we have to do a survey and that’s not science” (4th interview, term 3).  Ashlee was able to successfully pass the OC program and she transitioned into MSU at the start of her second year. She majored in physics and math and began taking mainstream courses in both subjects. She reported there was much less writing in these courses than at OC. She was also struggling in her math courses and reported that this subject had been challenging for her since high school (6th interview, term 4). She was also required to take two courses that focused on the philosophy of science, and she struggled with the writing assignments in these courses. Once again, Ashlee’s disciplinary perceptions would cause her to question the usefulness of course content and assignments. Ashlee submitted her writing assignments for one of these courses prior to our last meeting, which included three short compositions in which she was required to provide extended definitions about scientific notions and a longer expository essay on the double helix structure. In her text, I noted she primarily used passive voice and avoided any mention of self, sometimes using a collective “we” instead. For example, her introduction to the term paper is shown below:  In this paper, we are going to talk about double helix structure, single strain structure of DNA, in the aspects of the structures and how the structures function. And why  185  the view of the double helix structure is a better model will also be discussed. (Mainstream term paper, para 1) In her summary feedback on this assignment, for which she received a B (72%), there was no mention of language issues, and instead the instructor’s comments focused on content. Suggestions included expanding on concepts, ideas, or specialized terminology, making more explicit connections between her main argument and supporting details, and avoiding description of too many details. She reported appreciating this focus on content, but that it was still challenging for her to improve her performance because she didn’t know what else to write. She also reported not seeking any assistance on how to improve her writing because there were only two options available (e.g., MSU writing center or instructor’s office hours) and she did not think either would be helpful. She did not believe the course was helpful for her because it focused on topics that were about science, but were not scientific (6th interview, term 4).  When reflecting on OC and whether it helped prepare her for MSU, Ashlee indicated that her content courses were helpful and that she appreciated the dedicated efforts of many of her content instructors. However, her opinion of two of her academic language courses remained the same. She stated that the instructional concepts were too challenging to understand and apply and this made things more stressful for her. Furthermore, she did not feel these courses were relevant to her disciplinary studies and she questioned the scientific knowledge base of one of her language instructors. Increasingly, Ashlee did not regard her academic language courses as “real science,” which impacted her response to feedback and her reactions to grades she perceived as unfair or highly subjective. Over time, this mitigated her willingness to invest in the language practices of these courses. For Ashlee, this became a critical issue in her evaluation of the “scientific” quality of all her courses and her instructors. For her, the issue revolved around the need to be objective in  186  science and to avoid personal opinions as much as possible, something she associated with the social sciences. The influence of disciplinary perceptions in students’ willingness to accept and attend to feedback has been noted in the literature (Leki & Carson, 1994), some scholars referring to such learners as "resisters" who “perceive the language instructor as a person skilled merely in the grammar of the target language and its explication but not skilled in its principles of rhetorical organization” (Radecki & Swales, 1988, p. 364). 6.5 Appropriating written academic discourse  In relation to academic writing, the socialization of academic discourse was mediated by participants’ ability and willingness to exercise agency in their writing, the extent to which they appropriated language or had their text appropriated by others, and their ability to enact or claim legitimated identities. Each of these aspects will be discussed before describing the interconnected manner in which they influenced the participants in this study. The results of this study affirm the importance of agency and identity as salient notions mediating written academic discourse socialization for these participants. However, of equal significance to the participants in this study was the degree to which they reported being in control of their writing or that their writing was being controlled by others. The notion of appropriation has been discussed and debated by composition and L2 writing scholars for decades, and in the process has undergone considerable change. For example, Reid (1994) addresses what she views as the “myth of appropriation” in terms of teachers taking too much control over students’ ways with words. She argues the notion (as construed at the time) had failed to adequately consider the wider social context and that “the rights and responsibilities for making meaning in texts are shared by writers and readers” (p. 275). In her opinion, all teachers are in fact “cultural informants” who have a pedagogical responsibility to share their “cultural and rhetorical knowledge” (p. 278). Instead of  187  teachers representing the expert and students working towards pre-determined models, the objective should instead involve facilitating a “community of writers” in the classroom in which teachers become “liaisons” for students (p. 279). Similarly, Canagarajah and Matsumoto (2017) argue that there is a need to “design pedagogies that move beyond form-focused and teacher-led processes, and treat students as agentive and aware, able to chart their own trajectories and desired voices in their writing” (p. 404). Tardy (2006) proposes moving beyond unidirectional (or monologic) conceptions of appropriation by incorporating a deeper understanding of the social context that directly informs and influences text construction. Drawing on Bakhtin’s (1981, 1986) arguments that all texts are multi-voiced and co-constructed, she argues for a dialogic approach to the concept, one that foregrounds the “writer’s agency and the contingency of expertise, integrating issues of investment, power, and social alignment” (Tardy, 2006, p. 73). In this way, appropriation is viewed as multidirectional, as something imposed on writers by more experienced and powerful others, as well as a tool for writers to promote change and attempt to enact their abilities and preferred identities. Conceptualized in this way, the notion is highly compatible with notions of agency and identity as conceived in numerous second language socialization studies. A dialogic notion of appropriation, and its interconnected relationship with agency and identity, helps to clarify and highlight the tension that exists in global education contact zones, and the way this is perceived and reacted to by those who are less experienced and who must satisfy the linguistic and cultural expectations of more experienced others.   The three themes identified as salient for these participants are listed in Table 6.3, along with a working definition, a description of relevant experiences that aided in its identification, and the outcome or effect this theme had on students’ transition into mainstream university studies.   188  Table 6.3: Salient themes in participants’ written academic socialization Theme  Working definition Relevant experiences Outcome or effect Agency  Willingness and ability to exert control over one’s own or others’ words, dispositions, actions, etc.  Academic writing concepts were not well understood, and obfuscated students’ attempts to communicate in personally meaningful ways  Reduced sense of individual agency in producing appropriate disciplinary discourse Appropriation (dialogic) The degree to which words or actions are controlled by others or in one’s own control Performance of academic writing features became largely curriculum- and instructor-driven to satisfy perceived expectations only  Discourse increasingly controlled by more experienced others Identities The discursive and situational identities made available that can be appropriately and legitimately performed Participants’ sense of belonging within their academic program or specific courses became increasingly strained and created conflict and tension in transitioning to the mainstream Academic and disciplinary identity options were restricted  Although presented separately here, I do not envision these themes as mutually exclusive but rather as overlapping and mutually reinforcing. For the remainder of this chapter, I will discuss these interconnected themes and how they enhance our understanding of written academic discourse socialization in global education contact zones around the world. It should be noted that my analysis and interpretation is not informed by the instructor’s perspectives on these two students’ written performances, and in this way represents only a partial perspective on complex and multifaceted processes. With this limitation in mind, I offer some implications and tentative  189  suggestions that can be of great benefit to multilingual writers studying in similar sociolinguistic, socioeducational, and sociocultural contexts.  The above themes of agency, appropriation, and identity were interconnected for Ashlee and Jasmine, but expressed in decidedly unique ways. Both Jasmine and Ashlee reported considerable challenge in understanding and applying the writing concepts taught and practiced in their academic language courses. Because these were not well understood, this situation worsened over time and led to increasing confusion and frustration for both participants, which combined with what they perceived to be a rapid pace of instruction and high workload, further obfuscated their ability to communicate (in writing and otherwise) with more experienced others in intelligible ways or make meaningful connections to what they were learning in other courses. This result aligns well with other socialization studies that have demonstrated the importance of multilingual learners’ ability to understand and respond appropriately to socioculturally-mediated instructional practices and concepts (e.g., Bronson, 2004, Godfrey, 2015; Leki, 2007; Morton et al., 2015; Séror, 2008; Spack, 1997), and the potential for “cognitive overload” (Zappa-Hollman, 2007a) when practices are not aligned well with learners’ cultural expectations or linguistic abilities (see also Currie, 1988, and Mohan & Lo, 1985). Over time, both participants’ performance of academic writing in their academic language courses also became more curriculum- and instructor-driven (i.e., appropriated by more experienced others) instead of driven by meaningful engagement with instructional practices, written assignments, and the feedback received throughout the writing process. This finding aligns with the results of other studies that indicate learners desire authentic writing experiences and meaningful engagement with the content of their writing (Bronson, 2004; Séror, 2008; Zappa-Hollman, 2007a).   190  Focusing on multilingual learners’ responses to “Western” instructional practices, Canagarajah (2004) identified five heuristic categories (avoidance, accommodation, opposition, appropriation, and transposition) that characterize strategies L2 writers may utilize. He argues that students who primarily use avoidance, accommodation, or opposition strategies tend to become “univocal” in their L2 writing by adopting only “one strand of the conflicting discourses without negotiating an independent voice” (Canagarajah, 2004, p. 264). However, L2 writers who use appropriation or transposition find a “favorable space for one’s own voice in the established discourses or form a new discourse that transcends the earlier dichotomies” (p. 265). The results of this study suggest that for novice multilingual writers, the “choice” of what strategies to employ may be limited by the instructional context, classroom practices, types of assignments, intelligibility of feedback, and the ability of students to comprehend and apply concepts as instructed in the time made available. Both participants used a mixture of avoidance, accommodation, and opposition strategies, but were unable to take more control of their academic writing (i.e., appropriation or transformation). More discussion and research into these strategies, and what influences them at various educational levels and in various instructional contexts, would help to clarify how and why such strategies are taken up.  The results of this study imply that academic writing instruction that does not align well with learner expectations or their perceived abilities can become a source of confusion, frustration, and personal conflict for multilingual writers. The explicit instruction of complex academic concepts under time constraints can negatively impact multilingual writers’ development. In addition, how feedback is delivered and the extent that it meaningfully addresses the content of students’ writing can further complicate multilingual learners’ transitions. Another important implication from the results of this study is the need to better align the use of metalanguage in  191  instructional and assessment practices with the time needed for learners to process and begin to apply what they are learning. While this is always a delicate balance given the needs of various stakeholders, and learners’ willingness to engage can be a mitigating issue, numerous studies have found that when expectations are not in reach, they can have deleterious consequences for those unable to keep up, as well as for the instructors and administrators who are under pressure to prepare learners for the academic and disciplinary language demands of the university mainstream.  There are several pedagogical and programmatic implications that arise from these results. First, diagnostic assessment can be of great assistance in identifying students who, based on their language attitudes and abilities, might benefit from targeted discipline-specific academic language support (Fox et al., 2016; Palmer et al., 2018; Read, 2015). In addition, more open discussion incorporating notions of agency and appropriation and the role this plays in enabling or constraining the potential identities multilingual learners envision would help to address some of the challenges multilingual writers face in performing academic and disciplinary discourse. Shi and Beckett (2002) have suggested that those involved in educating multilingual writers need to become more aware that all writers carry with them different perspectives and traditions in their writing. If instructors focus primarily on deficit, or on how to best eradicate “non-native” features, they risk overlooking meaningful and culturally-valid methods of meaning making. Singh and Doherty (2004) argue that “teachers of EAP, ESL, and foundation studies need to critically engage with assumptions about teacher, student, and cultural identities” and suggest that “communicative relations in such contact zones need to be renegotiated, reworked, and remade in new and contingent ways” (p. 10). Similarly, Canagarajah and Matsumota (2017) have argued that safe spaces need to be created to encourage meaningful dialogue that frames the classroom as a contact zone that provides “ecological affordances for the negotiation of competing norms and the  192  emergence of new genres” (p. 390). However, the need to demonstrate progress in language learning to upper level stakeholders, often in quantifiable ways, can make such efforts impractical or impossible in instructional contexts. While developing more democratic practices for the benefits of multilingual writers will no doubt be challenging, failure to do so runs the risk of sweeping aside any meaningful consideration for the complex multidirectional process of socialization taking place, much of it not observable in class, understood by instructors, or visible in the writing that students do for them.     193  Chapter 7: Conclusion 7.1 Teaching and learning in global education contact zones In conducting this research, I was inspired and informed by my experiences teaching and assessing academic writing for a wide variety of purposes in South Korea and Canada. These experiences have provided me with a unique window into the experiences of multilingual learners as they negotiate English academic discourse in situated educational contexts (e.g., for standardized language tests, university preparation courses, mainstream undergraduate courses, etc.). Through these experiences, I have learned a great deal in my communications with multilingual learners, both through their writing and in conversations about their writing. These experiences have impressed on me the need, perhaps now more than ever, to better understand the changing perceptions and experiences of multilingual learners who are traversing countries, cultures, and languages in the continuance of their academic studies. This is needed to better inform the educational decisions made by program designers, administrators, instructors, learners, and other influential stakeholders in global education contact zones around the world (Canagarajah & Matsumoto, 2017; Pratt, 1991; Scotland, 2014; Shi, 2009; Singh & Doherty, 2004).  Before reiterating the main findings of each manuscript chapter and discussing their individual contributions and pedagogical implications, I would like to return to the three overarching research questions that guided this dissertation study: 1. How does the academic language program mediate students’ transitions into mainstream studies? 2. How do academic language courses mediate students’ perceptions and performances of written academic discourse?    194  3. What are some of the struggles experienced by multilingual writers as they transition through this program? Each manuscript addressed these overarching research questions in various ways. However, more focused research questions were developed for each chapter, some of which addressed one overarching research question more than the others. For example, Chapter 4 addresses the first question most directly as it investigates the mediating role of program design and delivery. Chapter 5 addresses the second question most directly since emphasis is placed on four successful participants’ responses to academic and disciplinary-specific writing instruction in their academic language courses. Finally, Chapter 6 addresses the third question most directly since it examines sociocultural aspects mediating the experiences of two multilingual writers who experienced considerable confusion, frustration, and conflict transitioning to their respective disciplinary areas of study. Collectively, these manuscripts triangulate understanding of macro, meso, and micro aspects (Douglas Fir Group, 2016; Duff, 2019) influencing written academic discourse socialization for the participants in this study. These sociohistoric, sociocognitive, and sociocultural processes, as several language socialization scholars and researchers have discovered, are continually in flux and negotiated over time and across settings (Duff, 2010; Duff & Talmy, 2011; Kobayashi et al., 2017; Morita & Kobayashi, 2008).   The individual experiences of participants in this study reveal complex written academic discourse socialization processes that are highly individualized and resistant to broad characterizations. These results align with numerous studies that have examined the long-term transitions of multilingual writers and found them to be characterized by both opportunities and conflicts that transcend strictly linguistic explanations (Anderson, 2016; Bronson, 2004; Casanave, 2002; Ivanič, 1998; Leki, 2007; Morton et al., 2015; Prior, 1998; Séror, 2008, Spack, 1997; Zamel  195  & Spack, 1998, 2004; Zappa-Hollman, 2007a). Each participant in this study responded to administrative and pedagogical aspects of the sheltered program in unique ways, which was influenced by their previous learning experiences, their evolving beliefs about effective and appropriate language teaching practices, and the most optimal way to achieve their goals. However, there were also many commonalities across these socialization experiences which were mediated by aspects of program design and delivery, pedagogical approaches, instructional practices, as well as the more personal relationships developed in individual classes. It must be noted that this study was conducted in the first year of a newly-designed academic language program. Program designers drew on a sophisticated range of theoretical and pedagogical approaches to provide a state-of the-art learning experience for the multilingual learners who enrolled. Given this was a for-credit foundational-year program that was designed to prepare learners for the second year of mainstream study, it was ambitious, and the stakes were arguably much higher than most other established academic language programs in Canadian universities. This was also a unique program in Canada (and possibly the world) in that it was 11-months in duration and included content and language courses delivered by MSU faculty members. Given the breadth and complexity of this program, it is important to remember the challenges that all academic language programs experience, particularly in their first year (Brown, 2009; Morris, 2006).  In addition, while I was conducting my research during this inaugural year, there were numerous other internal research projects and ongoing needs analyses surveys being conducted that were designed to assess what was happening on the ground to better respond to any issues discovered. My research, as it was limited in scope and restricted to the first year of the program, was not able to report on how these challenges were being addressed as well as the program and  196  curriculum changes occurring that year, and undoubtedly in the subsequent years of the program. This study focuses on the perceptions of six students who were willing to participate in this longitudinal study, and therefore, represents only a partial glimpse into the overall student experience in the program in its first year. With these important caveats in mind, the next section summarizes the main findings and contributions of each manuscript chapter.    7.2 Manuscript chapter summaries This section summarizes each manuscript chapter (Chapters 4-6) and their main contributions. While each manuscript is separate in a textual sense, it remains integrated within the overall multiple case study reported in this dissertation.  7.2.1 Chapter 4 In this chapter, I focused on how aspects of program design and delivery mediated students’ first-year apprenticeship into their respective programs of study. I also drew on the perspectives of other program stakeholders (an early consultant, two administrators, and two instructors) as well as institutional and media discourse related to the program, observations of language and content classrooms, and discussions with sheltered and mainstream instructors. Participants experienced enabling and constraining influences based on several aspects of program design and delivery. Salient themes identified in student and non-student discourse were unique but also shared, suggesting there are differences and commonalties in how program design and delivery influences administrators, instructors, and students. All stakeholders experienced challenges in responding to aspects of the program, including time constraints imposed upon them, the complexity and pace of the curriculum, and a perceived lack of cultural and linguistic diversity in the program. However, there was general agreement regarding the high quality of academic support provided. Most students reported positive benefits related to their participation in the program and indicated the  197  experience helped to scaffold their learning. However, students experienced several challenges in relation to program design features which impacted their transitions into the university mainstream. This included confusion and frustration regarding the credit status of their academic language courses and feelings of separation from the larger university community. However, most students reported that participation in the sheltered program (OC) in their first year helped them to successfully transition to the university mainstream (MSU). This research provides a nuanced perspective in that it explicitly investigates program features from a variety of stakeholder perspectives as well as the changing perspectives of the multilingual learners enrolled.  This research helps to fill a gap in the available literature by more explicitly considering the role of program design and delivery. While there is a large body of research that has investigated multilingual learner needs, much of this research focuses on pre-determined program-level outcomes with little interrogation of the language learning assumptions or power relationships involved. As a result, these initiatives often overlook the individual experiences and voices of the multilingual learners, particularly those who may be less successful or who are experiencing challenges not captured by preferred measurement tools. Large-scale survey studies have added to our understanding of some of these challenges and opportunities, but these provide a limited window into individual multilingual learner experiences. Socialization studies have provided valuable in-depth investigations into the complexities involved in making transitions to global education contact zones; however, much of this research overlooks the mediating role of program design and delivery. The results of this study affirm the value of including programmatic aspects and the perspectives of multiple program stakeholders as much as possible when investigating multilingual learners’ individual perceptions, experiences, or needs.  198  7.2.2 Chapter 5  This chapter investigated the written academic discourse socialization experiences of four multilingual learners. These four students (Chloe, Dawn, Theo, and Yvonne) were identified as successful in that they met all the requirements of the OC program and reported an overall positive experience. They also reported believing that participating in this specialized program helped them to make successful transitions to the university mainstream, a belief that strengthened over time. This investigation focused on participants’ experiences in their academic language courses, including their perceptions of their academic writing instruction, their challenges in completing their written assignments, and their reactions to the feedback they received. In addition to multiple interviews conducted with students (5-6 each), I collected several assignments and associated curriculum documents and drew on observational data to better understand participants changing perceptions about academic and disciplinary discourse and how this mediated their academic and disciplinary socialization. In two of three of their academic language courses, several students reported challenges in understanding instructional concepts or applying them in the time available. Students in the Arts Stream did not respond positively to the controlled writing practice and structured feedback they received in these courses. However, all students appreciated what they considered to be more authentic writing experiences as well as the engaging feedback they received in one of their academic language courses.   The results of this study suggest that explicit instruction of academic discourse concepts and the cognitive complexity involved can be highly impactful when objectives are not aligned well with students’ preferences, needs, and abilities. The results suggest that multilingual learners may experience mounting confusion, frustration, and stress when instruction and feedback is not well-understood and insufficient time is made available to compensate for this lack of understanding.  199  This can lead to avoidance, resistance, or rejection, as it did for some of the participants in this study. This indicates that the degree of authentic writing practice, the balance of content and form-focused feedback provided, and the time made available are potentially confounding factors in written academic discourse socialization for multilingual learners. This brings up several concerns regarding the degree of controlled writing practice and form-focused feedback that is most appropriate or effective. Such concerns include the potential for students to become overwhelmed with instructional complexity or for them to disengage with what they perceive to be repetitive or inauthentic writing tasks. These results also point to some of the challenges involved in integrating multiple theoretical and pedagogical approaches as well as some of the possibilities that integrating approaches has to offer for multilingual learners that may not be realized for some time after explicit instruction and controlled practice has ceased. These results suggest that frank discussion with learners about the purpose of their instruction, how students may react, and how it is designed to scaffold them for the future might help students appreciate its potential and may help to mitigate some of the cognitive, social, and sociocultural challenges involved in global education contact zones around the world. 7.2.3 Chapter 6  This chapter focuses on two student participants (Ashlee and Jasmine) who reported considerable confusion, frustration, and conflict in making their transitions to the mainstream. Ashlee became highly resistant to learning any content she did not view to be based on sound “scientific” principles. This led her to question the content as well as the pedagogical approaches utilized to deliver it. She did not generally read feedback from instructors she viewed as not in the realm of science (including the philosophy of science) or from instructors she did not believe had enough scientific background. No other participant displayed this level of resistance to pedagogical  200  content, disregard for feedback, and strong sense of personal agency. At the other extreme, Jasmine tried to incorporate almost everything that her instructors provided in feedback and revised much of her writing on this basis. She took advantage of additional assistance whenever it was available to improve her writing performances and the grades she received (although this was not her primary focus). Unfortunately, this often resulted in greater confusion for her and a continued inability for her instructors to follow the logic and organization of her writing. No other participant displayed this level of accommodation of pedagogical content and attention to feedback in their academic language courses.  Both Ashlee and Jasmine struggled to achieve passing grades, and both were at risk of failing courses early in their first year of study. However, Ashlee was able to satisfy program requirements and enter the mainstream in her second year. Jasmine was not successful and completed another year in an affiliated college to qualify for MSU in her third year (which she was successful in doing). For Jasmine, her experiences improved in the second-year college because of the reduced workload, a slower-paced curriculum, more individualized and personalized support, and being able to communicate and work with domestic students. For these participants, the interconnected themes of agency, appropriation, and identities helped me to interpret the changing perceptions and experiences of these learners and provides a useful heuristic for investigating multilingual learner transitions in future studies of written academic discourse socialization in global education contact zones. 7.3 Limitations and additional challenges Case study research affords many advantages in understanding individual “experiences, issues, insights, developmental pathways, or performance within a particular linguistic, social, or educational context” (Duff, 2014, p. 233). However, there are also practical and methodological  201  limitations that often arise (Duff, 2008; Yin, 2009) and this study was no different. First, recruitment and data collection methods were impacted by several factors. Since this was a new program and time was needed for students and other stakeholders to adjust, it was not possible to observe classes or interview instructors in the first term. My efforts during the first term were focused on gathering and analyzing institutional and curriculum discourse, interviewing administrators, and recruiting and interviewing student participants. While I was able to collect assignments written in the first term and conduct an initial interview with most participants, I did not start observing classes or interviewing instructors until the second term. It would have been preferable to have observed courses earlier in the program to better contextualize student participants’ early perceptions and experiences upon arrival to OC (and to Canada). Nevertheless, the support I received and the access I was granted was extensive given: 1) the high stakes involved in the rollout of this new program, 2) the level of sophistication and breadth involved in its design, and 3) my status as a doctoral researcher. This level of access was possible because program administrators and instructors were committed to open discussion and recognized the need for applied research from multidisciplinary perspectives. The shared institutional objective that became evident to me was the desire to provide an effective and appropriate first-year experience for the students enrolled and prepare as many of them as possible for the academic and disciplinary demands of their disciplines.  I initially hoped to select a representative sample of the student population from a larger group of interested students; however, I found it challenging to recruit students and keep them in the study. Three students who initially signed informed consent letters decided to drop out of the study early in their second term. Thankfully, I was able to add one more participant in the second term. Given a great deal of focus in this study is placed on students’ academic language courses and  202  their written assignments, it represents a partial view of students’ overall academic literacy socialization. In terms of the data collected for this study, there were some limitations created by participants’ schedules and their degree of willingness to contribute assignments, grades or feedback, or keep a journal. Challenges in collecting written data from student participants is not uncommon, particularly in longitudinal studies (e.g., Anderson, 2015; Séror, 2008). My original plan was to have students submit a written reflection each term on their experiences completing a writing assignment. Unfortunately, I only received one short reflection by the middle of the second term, so I decided not to press students further given their busy schedules. While several participants contributed their grades and feedback, some participants were more reluctant. In addition, some participants submitted many of their writing assignments and others comparatively few. Since assignments were submitted voluntarily, there was considerable variability in the number and types of assignments submitted to the study and discussed in interviews. I was also restricted in terms of providing facsimiles or screenshots of instructors’ feedback on student writing, which would have helped to visualize what the students were receiving. There were also some challenges in working around the schedules of students (who were exceptionally busy) to conduct timely interviews as close as possible after the end of each term so that they did not interfere unduly with their studies. Instructors were also exceptionally busy throughout the program and, as a result, they were often unable to conduct a formal recorded interview. However, given the time constraints everyone was under in this program, the access and engagement I did receive from many dedicated program stakeholders was instrumental in providing the valuable contextual information I collected, which enhanced my ability to represent students’ perceptions and experiences and triangulate my interpretations.    203  7.4 Questions, suggestions, and closing thoughts  While the design of this study does not permit bold conclusions or recommendations to be made, there are several lingering questions and suggestions that arise from these results. Firstly, these results raise some concerns about the potential benefit of explicit academic writing instruction and controlled writing practice in preparing multilingual learners for disciplinary-specific academic writing practices, particularly in contexts that are fast paced and high stakes. The explicit instruction of academic discourse concepts, depending on their complexity and the pace of instruction, may cause considerable stress and identity conflict for some learners when not aligned well with their (perceived or actual) needs and abilities. As has been discovered in several studies reviewed throughout this dissertation, when instructional concepts are not well understood this can have negative consequences for some multilingual learners.  However, the reasons why some learners react differently to similar instructional practices is highly complex, as discovered in Godfrey’s (2015) study of Jason and Nancy, who were both studying in a “favorable site and space of socialization” (p. 167). In the present study, Theo (Sciences) was highly pragmatic in adopting and adapting to written academic discourse features and Dawn (Sciences) was very accommodating but less engaged with instructional concepts than Theo. Although Theo and Dawn struggled with the complexity of written academic discourse concepts taught and practiced, over time they reported gaining more awareness and control of some of these features in their mainstream writing. Chloe and Yvonne (Arts Stream) struggled with the complexity and pace of academic language instruction, which led them to attend to revisions in a largely non-engaged, mechanical fashion. They sought more meaningful engagement with the content of their writing and more authentic writing experiences. The remaining students, Ashlee (Sciences) and Jasmine (Arts) responded to the challenges they were experiencing in decidedly  204  different ways; however, the imbalance both experienced in agency and appropriation had negative consequences in terms of how they saw themselves as learners and writers. As Séror (2008) and Anderson (2015) discovered, instruction and feedback serve important socializing functions that impact what learners believe they can do and what identities they believe are possible to take up.   Given the complexities involved in aligning theoretical and pedagogical approaches with diverse multilingual learner needs and institutional expectations, some may question what should (or can) be done. There are several suggestions that offer a potential way forward. First, these results suggest it is necessary to better understand learners’ background experiences (e.g., language learning beliefs, high school curriculum, test-preparation experiences) and how they may impact learner response to instruction and assessment practices. This process requires continual communication that is sensitive to linguistic and sociocultural differences. As Zappa-Hollman (2007a) discovered with her participants, a “new way of learning” (p. 119) in Canada poses considerable challenges for some learners, which can lead to significant confusion and mounting frustration when left unaddressed. For some learners, this can contribute to cognitive overload as they attempt to perform in ways valued and rewarded in their new academic discourse community without the necessary scaffolds in place to make these adjustments in a reasonable (and well-reasoned) manner.  These results also point to some of the challenges involved in integrating multiple theoretical and pedagogical approaches (e.g., genre-based pedagogy, SFL, CLIL, etc.). However, they also suggest there are numerous benefits that may not be realized for some time after instruction and practice has ceased. Duff (2007a) has noted that isolated examples of the “after-life” of socialization are insufficient evidence that “earlier discourse socialization experiences were inappropriate, or their efforts were misplaced” since they may provide “a solid foundation on the  205  basis of which students could then adapt their practices and introduce innovation and efficiency as needed” (p. 14). This appeared to be the case for Theo and to a lesser extent Dawn. However, the results of this research also suggest that inappropriate or unrealistic expectations can create considerable pressure for multiple stakeholders and negative consequences for learners. This finding underscores the need to acknowledge and carefully consider the time it takes to achieve ambitious educational objectives and the time multilingual learners need to understand socioculturally-influenced academic discourse expectations. Finally, more critical approaches to assessing learner needs are needed to better enable multilingual learners in global education contact zones around the world to appropriate English academic discourse in ways that celebrate their linguistic repertoires, draw on their background knowledge, enhance their confidence, enable their agency, and promote positive perceptions of the identity options available to them and how they may go about laying claim to them.   206  References Agosti, C. I., & Bernat, E. (2018). University pathway programs: Local responses within a growing global trend. Cham, Switzerland: Springer International Publishing AG. Ahearn, L. M. (2001). Language and agency. Annual Review of Anthropology, 30, 109–137. Anderson, T. (2015). Seeking internationalization: The state of Canadian higher education. Canadian Journal of Higher Education, 45, 166-187. Anderson, T. (2016). Negotiating academic discourse practices, ideologies, and identities: The socialization of Chinese PhD students. Unpublished doctoral dissertation. University of British Columbia, Canada. Artemeva, N., & Freedman, A. (Eds.). (2008). Rhetorical genre studies and beyond. Winnipeg, Canada: Inkshed Publications. Artemeva, N., & Freedman, A. (2015). Introduction: Everything is illuminated, or genre beyond the three traditions. In N. Artemeva & A. Freedman (Eds.), Genre studies around the globe: Beyond the three traditions (pp. xix-xxviii). Winnipeg, Canada: Inkshed Publications.   Atkinson, D. (2002). Toward a sociocognitive approach to second language acquisition. The Modern Language Journal, 86, 525-545. Atkinson, D. (2003). Language socialization and dys-socialization in a South Indian college. In R. Bayley & S. Schecter (Eds.), Language socialization in bilingual and multilingual societies (pp. 147–161). Clevedon, UK: Multilingual Matters,  Atkinson, D., & Ramanathan, V. (1995). Cultures of writing: An ethnographic comparison of L1 and L2 university writing/language programs. TESOL Quarterly, 29, 539–568.  Atkinson, P., & Silverman, D. (1997). Kundera’s Immortality: The interview society and the invention of the self. Qualitative Inquiry, 3, 304–25. B.C. migrant students want credit for learning English. (2013, Oct. 3). CBC. Retrieved from https://www.cbc.ca/news/canada/british-columbia/b-c-migrant-students-want-credit-for-learning-english-1.1894633 Baik, C., & Greig, J. (2009). Improving the academic outcomes of undergraduate ESL students: The case for discipline-based academic skills programs. Higher Education Research & Development, 28, 401–416. Bakhtin, M. M. (1981). The dialogic imagination (M. Holquist, Ed.; C. Emerson & M. Holquist, Trans.). Austin, TX: University of Texas Press.  207  Bakhtin, M. M. (1986). The problem of speech genres. In C. Emerson, & M. Holquist (Eds.), Speech genres and other later essays (V. McGee, Trans., pp. 60–102). Austin, TX: University of Texas Press. Barton, D., & Donahue, C. (2009). Multiple assessments of a first-year seminar pilot. The Journal of General Education, 58, 259-278. Basturkmen, H. & Lewis, M. (2002). Learner perspectives of success in an EAP writing course. Assessing Writing, 8, 31-46.  Bawarshi, A. & Reiff, M. J. (2010). Genre: An introduction to history, theory, research, and pedagogy. West Lafayette, IN: Parlor Press. Bayley, R., & Schecter, S. (Eds.). (2003). Language socialization in bilingual and multilingual societies. Clevedon, UK: Multilingual Matters. Bazerman, C. (1981). What written knowledge does: Three examples of academic discourse. Philosophy of the Social Sciences, 11, 361-387.  Bazerman, C. (1988). Shaping written knowledge: The genre and activity of the experimental research article in science. Madison, WI: University of Wisconsin Press.  Bazerman, C., Applebee, A., Berninger, V., Brandt, D., Graham, S., Matsuda, P. K., Murphy, S., Rowe, D., & Schleppegrell, M. (2017). Taking the long view on writing development. Research in the Teaching of English, 51, 361-360. Bazerman, C., Bonini, A., & Figueiredo, D. (Eds.). (2009). Genre in a changing world. Fort Collins, CO: Parlor Press. Bazerman, C., & Prior, P. (Eds.). (2004). What writing does and how it does it: An introduction to analyzing texts and textual practices. Mahwah, New Jersey: Lawrence Erlbaum Associates. Belcher, D., & Braine, G. (Eds.). (1995). Academic writing in a second language: Essays on research and pedagogy. Norwood, NJ: Ablex.  Benesch, S. (1996). Needs analysis and curriculum development in EAP: An example of a critical approach. TESOL Quarterly, 30, 723–738. Benesch, S. (2001). Critical English for academic purposes: Theory, politics and practice. New York: Routledge. Bereiter, C., & Scardamalia, M. (1987). The psychology of written composition. Hillsdale, NJ: Lawrence Erlbaum.  208  Berkenkotter, C., & Huckin, T. (1995). Genre knowledge in disciplinary communication: Cognition/culture/power. Northvale, NJ: Lawrence Erlbaum.  Berlin, J. (1987). Rhetoric and reality: Writing instruction in American colleges, 1900–1985. Carbondale, IL: Southern Illinois University Press. Bitchener, J., Storch, N., & Wette, R. (Eds.). (2017). Teaching writing for academic purposes to multilingual students: Instructional approaches. New York: Routledge. Blakeslee, A. M. (1997). Activity, context, interactions, and authority: Learning to write scientific papers in situ. Journal of Business and Technical Communication, 11, 125–169. Block, D. (2003). The social turn in second language acquisition. Edinburgh: Edinburgh University Press.  Bourdieu, P. (1977). The economics of linguistic exchanges. Social Science Information, 16, 645-668. Bourdieu, P. (1991). Language and symbolic power. Cambridge: Harvard University Press. Bourdieu, P., Passeron, J-C., & de Saint Martin, M. (Eds.). (1994). Academic discourse: Linguistic misunderstanding and professorial power. Palo Alto, CA: Stanford University Press. Boyatzis, R. (1988). Transforming qualitative information: Thematic analysis and code development. Thousand Oaks, CA: Sage.  Braine, G. (1996). ESL students in first-year writing courses: ESL versus mainstream classes. Journal of Second Language Writing, 5, 91-107. Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3, 77–101. Braun, V., & Clarke, V. (2016). (Mis)conceptualising themes, thematic analysis, and other problems with Fugard and Potts’ (2015) sample-size tool for thematic analysis. International Journal of Social Research Methodology, 19, 739-743. Brinton, D. (2012). Content-based instruction in English for specific purposes. In C. Chapelle (Ed.), The Encyclopedia of Applied Linguistics (pp. 1-9). Hoboken, NJ: Blackwell Publishing. Bronson, M. (2004). Writing passage: Academic literacy socialization among ESL graduate students, a multiple case study. Unpublished doctoral dissertation. University of California, USA. Brook, H., Fergie, D., Maeorg, M., & Michell, D. (2014). Universities in transition: Foregrounding social contexts of knowledge in the first-year experience. Adelaide, Australia: University of Adelaide Press.  209  Brown, J.D. (2009). Language curriculum development: Mistakes were made, problems faced, and lessons learned. Second Language Studies, 28, 85-105. Brown, J.D. (2016). Introducing needs analysis and English for specific purposes. New York: Routledge. Bunch, G. C., Abram, P. L., Lotan, R. A., & Valdés, G. (2001). Beyond sheltered instruction: Rethinking conditions for academic language development. TESOL Journal, 10, 28-33. Canagarajah, S. (1997). Safe houses in the contact zone: Coping strategies of African-American students in the academy. College Composition and Communication, 48, 173-196. Canagarajah, S. (1999). Resisting imperialism in English language teaching. Oxford, UK: Oxford University Press.  Canagarajah, S. (2004). Multilingual writers and the struggle for voice in academic discourse. In A. Pavlenko & A. Blackledge (Eds.), Negotiation of identities in multilingual contexts (pp. 266-289). Clevedon, UK: Multilingual Matters.  Canagarajah, S. (2011). Codemeshing in academic writing: Identifying teachable strategies of translanguaging. The Modern Language Journal, 95, 401–417. Canagarajah, S. (2013). Translingual practice: Global Englishes and cosmopolitan relations. London: Routledge. Canagarajah, S., & Matsumota, Y. (2017). Negotiating voice in translingual literacies: From literacy regimes to contact zones. Journal of Multilingual and Multicultural Development, 38, 390-406. Casanave, C. (2002). Writing games: Multicultural case studies of academic literacy practices in higher education. Mahwah, NJ: Lawrence Erlbaum. Casanave, C., & Li, X. (Eds.). (2008). Learning the literacy practices of graduate school: Insiders’ reflections on academic enculturation. Ann Arbor, MI: University of Michigan Press. Cenoz, J. (2015). Content-based instruction and content and language integrated learning: The same or different? Language, Culture, and Curriculum, 28, 8-24. Charles, M., & Pecorari, D. (2016). Introducing English for academic purposes. New York: Routledge. Cheng, L., & Fox, J. (2008). Towards a better understanding of academic acculturation: Second language students in Canadian universities. Canadian Modern Language Review, 65, 307–333.  210  Cheng, L., Watanabe, Y, & Curtis, A. (2004). Washback in language testing: Research contexts and methods. Mahwah, NJ: Lawrence Erlbaum. Choi, I. (2008). The impact of EFL testing on EFL education in Korea. Language Testing, 25, 39-61. Coe, R. (2002). The new rhetoric of genre: Writing political briefs. In A. M. Johns (Ed.), Genre in the classroom (pp. 195–205). Mahwah, NJ: Erlbaum. Crosthwaite, P. (2016). A longitudinal multidimensional analysis of EAP writing: Determining EAP course effectiveness. Journal of English for Academic Purposes, 22, 166-178. Cummins, J. (2007). Rethinking monolingual instructional strategies in multilingual classrooms. Canadian Journal of Applied Linguistics, 10, 221–240. Currie, P. (1988). Staying out of trouble: Apparent plagiarism and academic survival. Journal of Second Language Writing, 7, 1-18 Darvin, R., & Norton, B. (2015). Identity and a model of investment in applied linguistics. Annual Review of Applied Linguistics, 35, 36–56. de Silva Joyce, H., & Feez, S. (2012). Text-based approaches to language and literacy education: Programming and methodology. Putney, Australia: Phoenix Education.  Deschambault, R. (2015). Economizing education: Fee-paying ESL students in a public school. Unpublished doctoral dissertation. University of British Columbia, Canada.  Dias, P., Freedman, A., Medway, P., & Paré, A. (Eds.). (1999). Worlds apart: Acting and writing in academic and workplace contexts. New York: Routledge.  Dias, P., & Paré, A. (Eds.). (2000). Transitions: Writing in academic and workplace settings. Cresskill, New Jersey: Hampton Press. Douglas Fir Group. (2016). A transdisciplinary framework for SLA in a multilingual world. Modern Language Journal, 100 (Supplement), 19–47. Duff, P. A. (1995). An ethnography of communication in immersion classrooms in Hungary. TESOL Quarterly, 29, 505-537. Duff, P. A. (2007a). Problematising academic discourse socialization. In H. Marriott, T. Moore, & R. Spence-Brown (Eds.), Learning discourses and the discourses of learning (pp. 1–18). Melbourne, Australia: University of Sydney Press. Duff, P. A. (2007b). Second language socialization as sociocultural theory: Insights and issues. Language Teaching, 40, 309-319.  211  Duff, P. A. (2008). Case study research in applied linguistics. New York: Lawrence Erlbaum. Duff, P. A (2010). Language socialization into academic discourse communities. Annual Review of Applied Linguistics, 30, 169-192. Duff, P. A. (2012). Issues of identity. In A. Mackey & S. Gass (Eds.), The Routledge handbook of second language acquisition (pp. 410-426). London: Routledge. Duff, P. A. (2014). Case study research on language learning and use. Annual Review of Applied Linguistics, 34, 233–255. Duff, P. A. (2019). Social dimensions and processes in second language acquisition: Multilingual socialization in transnational contexts. Modern Language Journal, 103, 6-22. Duff, P. A., & Anderson, T. (2015). Academic language and literacy socialization for second language students. In N. Markee (Ed.), The Handbook of classroom discourse and interaction (pp. 337-352). West Sussex, UK: John Wiley & Sons.  Duff, P. A., & Doherty, L. (2015). Examining agency in (second) language socialization research. In P. Deters, X. Gao, E., R. Miller, & G. Vitanova, (Eds.), Theorizing and analyzing agency in second language learning (pp. 54–72). Bristol, UK: Multilingual Matters. Duff, P. A., & Kobayashi, M. (2010). The intersection of social, cognitive, and cultural processes in language learning: A second language socialization approach. In R. Batstone (Ed.), Sociocognitive perspectives on language use and language learning (pp. 75-93). Oxford: Oxford University Press. Duff, P. A., & May, S. (Eds.) (2017). Language socialization: Encyclopedia of language and education (3rd ed.). Cham, Switzerland: Springer International Publishing AG. Duff, P. A., & Talmy, S. (2011). Language socialization approaches to second language acquisition: Social, cultural, and linguistic development in additional languages. In D. Atkinson (Ed.), Alternative approaches to second language acquisition (pp. 95-116). New York: Routledge. Duranti, A., Ochs, E., & Schieffelin, B. (Eds.). (2012). The handbook of language socialization. Malden: MA: Wiley-Blackwell. Echevarría, J., & Graves, A. (2010). Sheltered content instruction: Teaching students with diverse abilities (4th ed.). Boston: Allyn & Bacon. Edwards, H., Wesche, M., Krashen, S., Clement, R., & Kruidenier, B. (1984). Second language acquisition through subject-matter learning: A study of sheltered psychology classes at the University of Ottawa. Canadian Modern Language Review, 41, 268-282.  212  Elbow, P. (1998a). Writing without teachers (2nd ed.). New York: Oxford University Press. Elbow, P. (1998b). Writing with power: Techniques for mastering the writing process (2nd ed.). New York: Oxford University Press. Elsom, S., Greenaway, R., & Marshman, M. (2017). Experiences of bridging program students at a regional satellite campus. Australian Journal of Adult Learning, 57, 242-265. Engeström, Y. (2014). Learning by expanding: An activity-theoretical approach to developmental research (2nd ed.). Cambridge: Cambridge University Press. Fang, Z., & Schleppegrell, M. J. (2010). Disciplinary literacies across content areas: Supporting secondary reading through functional language analysis. Journal of Adolescent & Adult Literacy, 53, 589-597. Fenton-Smith, B. & Gurney, L. (2016). Actors and agency in academic language policy and planning. Current Issues in Language Planning, 17, 72-87. Fenton-Smith, B., Humphreys, P., Walkinshaw, I., Michael. R., & Lobo, A. (2017). Implementing a university-wide credit-bearing English language enhancement programme: Issues emerging from practice. Studies in Higher Education, 42, 463-479. Fernandez, R., Peyton, J. K., & Schaetzel, K. (2017). A survey of writing instruction in adult ESL programs: Are teaching practices meeting adult learner needs? Journal of Research and Practice for Adult Literacy, Secondary, and Basic Education, 6, 5-20.  Ferris, D., & Hedgcock, J. (2014). Teaching L2 composition: Purpose, process, and practice (3rd ed.). New York: Routledge.   Firth, A., & Wagner, J. (2007). Second/foreign language learning as a social accomplishment: Elaborations on a reconceptualized SLA. Modern Language Journal, 91, 800-819. Flower, L. S., & Hayes, J. R. (1981). A cognitive process theory of writing. College Composition and Communication, 32, 365–387. Flowerdew, J. (2016). English for Specific Academic Purposes (ESAP) writing: Making the case. Writing & Pedagogy, 8, 5-32.  Flowerdew, J., & Costley, T. (Eds.). (2017). Discipline-specific writing: Theory into practice. New York: Routledge. Flowerdew, J., & Wan, A. (2010). The linguistic and the contextual in applied genre analysis: The case of the company audit report. English for Specific Purposes, 29, 78-93. Flyvbjerg, B. (2006). Five misunderstandings about case-study research.  Qualitative Inquiry, 12, 219-245.   213  Foreign Affairs, Trade and Development Canada. (2012). International education: A key driver of Canada’s future prosperity. Retrieved from http://www.international.gc.ca.   Foreign Affairs, Trade and Development Canada. (2014). Canada’s international education strategy: Harnessing our knowledge advantage to drive innovation and prosperity. Retrieved from http://www.international.gc.ca.  Fox, J. (2009). Moderating top-down policy impact and supporting EAP curricular renewal: Exploring the potential of diagnostic assessment. Journal of English for Academic Purposes, 8, 26–42. Fox, J., Cheng, L., & Zumbo, B. (2014). Do they make a difference? The impact of English language programs on second language students in Canadian universities. TESOL Quarterly, 48, 57-85. Fox, J., Haggerty, J., & Artemeva, N. (2016). Mitigating risk in first-year engineering: Postadmission diagnostic assessment in a Canadian university. In J. Read (Ed.), Post-admission language assessment of university students (pp. 43-65). New York: Springer. Fox, J. M. (2017). Transitioning to a U.S. university: Case studies of Chinese students and an intensive English program. Unpublished doctoral dissertation. University of Nebraska, USA. Freedman, A. (1993). Show and tell? The role of explicit teaching in the learning of new genres. Research in the Teaching of English, 27, 222-251. Freedman, A. (1999). Beyond the text: Towards understanding the teaching and learning of genres. TESOL Quarterly, 33, 764-767. Fujioka, M. (2014). L2 student–U.S. professor interactions through disciplinary writing assignments: An activity theory perspective. Journal of Second Language Writing, 25, 40-58. Galante, A., Okubo, K., Cole, C., Abd Elkader, N., Carozza, N., Wilkinson, C., Wotton, C., & Vasic, J. (2019). Plurilingualism in higher education: A collaborative initiative for the implementation of plurilingual pedagogy in an English for academic purposes program at a Canadian university. TESL Canada Journal, 36, 121-133.   Gallagher, C., & Haan, J. (2018). University faculty beliefs about emergent multilinguals and linguistically responsive instruction. TESOL Quarterly, 52, 304-330. Gao, X. (2013). Reflexive and reflective thinking: A crucial link between agency and autonomy. Innovation in Language Learning and Teaching, 7, 226–237.  214  Gee, J. P. (2000). The new literacy studies: From ‘socially situated’ to the work of the social. In D. Barton, M. Hamilton, & R. Ivanic (Eds.), Situated literacies: Reading and writing in context (pp. 177-193) London: Routledge.  Godfrey, L. B. (2015). Becoming a writer, becoming a student: The pedagogy of academic discourse socialization in a first-year writing course for non-native English speakers. Unpublished doctoral dissertation, University of Minnesota, USA.  Green, J., & Bloome, D. (2014). Ethnography and ethnographers of and in education: A situated perspective. In J. Flood, S.B. Heath, & D. Lapp (Eds.), Handbook of research on teaching literacy through the communicative and visual arts (pp. 181-202). New York: Routledge.  Haggerty, J., & Fox, J. (2015). Raising the bar: Language testing experience and second language motivation among South Korean young adolescents. Language Testing in Asia, 5, 1-16.  Halliday, M. A. K. (2007). Language and society: Volume 10 in the collected works of M.A.K. Halliday. New York: Continuum. Halliday, M. A. K., & Hasan, R. (1985). Language, context, and text: Aspects of language in a social-semiotic perspective. Victoria, Australia: Deakin University Press.  Hansen, J. (2000). Interactional conflicts among audience, purpose, and content knowledge in the acquisition of academic literacy in an EAP course. Written Communication, 17, 27-52. Harklau, L. (2000). From the ‘good kids’ to the ‘worst’: Representations of English language learners across educational settings. TESOL Quarterly, 34, 35-67. Hauptman, P., Wesche, M., & Ready, D. (1988). Second language acquisition through subject‐matter learning: A follow‐up study at the University of Ottawa. Language Learning, 38, 433-475. Heller, R. (2011). In praise of amateurism: A friendly critique of Moje's "call for change" in secondary literacy. Journal of Adolescent & Adult Literacy, 54, 267-273. Helmer, K. (2013). Critical English for academic purposes: Building on learner, teacher, and program strengths. Journal of English for Academic Purposes 12, 273–287. Heng Hartse, J., & Kubota, R. (2014). Pluralizing English? Variation in high-stakes academic texts and challenges of copyediting. Journal of Second Language Writing, 22, 355–373.  Herrington, A. (1985). Writing in academic settings: A study of the contexts for writing in two college chemical engineering courses. Research in the Teaching of English, 19, 331-361. Holliday, A. (2005). How is it possible to write? Journal of Language, Identity & Education, 4, 304-309  215  Horner, B., Lu, M., & Matsuda, P. K. (Eds.) (2010). Cross-language relations in composition: Understanding the multilingual nature of composition courses. Carbondale, IL: Southern Illinois University Press. Howe, K. (2004). A critique of experimentalism. Qualitative Inquiry, 10, 42-61. Hutchinson, T., & Waters, A. (1987). English for specific purposes: A learner-centered approach. Cambridge: Cambridge University Press.  Hyland, K. (2003). Genre-based pedagogies: A social response to process. Journal of Second Language Writing, 12, 17-29. Hyland, K. (2009). Academic discourse: English in a global context. London: Continuum. Hyland, K. (2012). Disciplinary identities: Individuality and community in academic discourse. Cambridge, UK: Cambridge University Press. Hyland, K. (2013). Writing in the university: Education, knowledge and reputation. Language Teaching, 46, 53–70.  Hyland, K., & Hyland, F. (2006). Interpersonal aspects of response: Constructing and interpreting teacher written feedback. In K. Hyland & F. Hyland (Eds.), Feedback in second language writing: Contexts and issues (pp. 206-224). New York: Cambridge University Press. Hyland, K., & Shaw, P. (Eds.) (2016). The Routledge handbook of English for academic purposes. New York: Routledge.  Hyon, S. (1996). Genre in three traditions: Implications for ESL. TESOL Quarterly, 30, 693-722. Hyon, S. (2018). Introducing genre and English for specific purposes. New York: Routledge. Ivanič, R. (1998). Writing and identity: The discoursal construction of identity in academic writing. Amsterdam, the Netherlands: John Benjamins. Jenkins, J. (2011). Accommodating (to) ELF in the international university. Journal of Pragmatics, 43, 926-936. Johns, A. M., Bawarshi, A., Coe, R., Hyland, K., Paltridge, B., Reiff, M. J., & Tardy, C. (2006). Crossing the boundaries of genre studies: Commentaries by experts. Journal of Second Language Writing, 15, 234-249.  Kachru, B. (1985). Standards, codification, and sociolinguistic realism: The English language in the outer circle. In R. Quirk and H.G. Widdowson (Eds.), English in the world: Teaching and learning the language and literatures (pp. 11-30). Cambridge: Cambridge University Press.  216  Kalantzis, M. & Cope, B. (2016). Debating Functional Literacy. Retrieved from http://newlearningonline.com Katznelson, H., Perpignan, H., & Rubin, B. (2001). What develops along with the development of second language writing? Exploring the ‘‘by-products.’’ Journal of Second Language Writing, 10, 141-159. Keck, C. (2006). The use of paraphrase in summary writing: A comparison of L1 and L2 writers. Journal of Second Language Writing, 15, 261–278. Keefe, K., & Shi. L. (2017). An EAP program and students’ success at a Canadian university. TESL Canada Journal, 34, 1-24. Kiely, R. (2006). Evaluation, innovation, and ownership in language programs. Modern Language Journal, 90, 597-601. Knight, J. (2013). The changing landscape of higher education internationalization – for better or worse? Perspectives: Policy and Practice in Higher Education, 17, 84-90. Knoblock, N. & Youngquist, J. (2016). College-level sheltered instruction: Revisiting the issue of effectiveness. Journal of the Scholarship and Teaching of Learning, 16, 49-69.  Kobayashi, M. (2004). A sociocultural study of second language tasks: Activity, agency, and language socialization. Unpublished doctoral dissertation, University of British Columbia, Canada. Kobayashi, M., Zappa-Hollman, S., & Duff, P. A. (2017). Academic discourse socialization. In P. Duff & S. May (Eds.), Language socialization. Encyclopedia of language and education (3rd ed., pp. 239-254). New York: Springer. Kulick, D., & Schieffelin, B. B. (2004). Language socialization. In A. Duranti (Ed.), A companion to linguistic anthropology (pp. 349–368). Oxford, UK: Blackwell. Larsen-Freeman, D. (2019). On language learner agency: A complex dynamic systems theory perspective. The Modern Language Journal, 103 (Supplement), 61-79.  Laupenmühlen, J. (2012). Making the most of L1 in CL (1+2) IL. In D. Marsh & O. Meyer (Eds.), Quality interfaces: Examining evidence & exploring solutions in CLIL (pp. 237–251). Eichstaett: Eichstaett Academic Press. Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge: Cambridge University Press. Lea, M., & Street, B. (1998). Student writing in higher education: An academic literacies approach. Studies in Higher Education, 23, 157–172.  217  Lea, M., & Street, B. (2006). The "academic literacies" model: Theory and applications. Theory into Practice, 45, 368-377. Lee, J. J., Maldonado-Maldonado, A., & Rhoades. G. (2006). The political economy of international student flows: Patterns, ideas, and propositions. In J. C. Smart (Ed.), Higher education: Handbook of theory and research (pp. 545–590). Dordrecht: Springer. Leki, I. (2007). Undergraduates in a second language: Challenges and complexities of academic literacy development. Mahwah, NJ: Erlbaum. Leki, I., & Carson, J. (1994). Students' perceptions of EAP writing instruction and writing needs across the disciplines. TESOL Quarterly, 28, 81-101.  Leki, I., & Carson, J. (1997). "Completely different worlds": EAP and the writing experiences of ESL Students in university courses. TESOL Quarterly, 31, 39-69. Leki, I., Cumming, A., & Silva, T. (2008). A synthesis of research on second language writing in English. New York: Routledge.  Leung, C., Harris, R., & Rampton, B. (1997). The idealised native speaker, reified ethnicities, and classroom realities. TESOL Quarterly, 31, 543-560. Liew, W. M., & Ball, A. F. (2010). Social academic writing: Exploring academic literacies in text-based computer conferencing. In C. Bazerman, R. Krut, K. Lunsford, S. McLead, S. Null, P. Rogers, & A. Stansell (Eds.), Traditions of writing research (pp. 325–338). New York: Routledge. Lillis, T. M. (2001). Student writing: Access, regulation, desire. New York: Routledge. Lillis, T. M. (2003). Student writing as 'Academic Literacies': Drawing on Bakhtin to move from critique to design. Language and Education, 17, 192-207. Lillis, T. M. (2008). Ethnography as method, methodology, and ‘deep theorizing’. Written Communication, 25, 353-388.  Lillis, T. M., & Curry, M.J. (2010). Academic writing in global contexts: The politics and practices of publishing in English. London: Routledge. Lin, A. (2015). Conceptualizing the potential role of L1 in CLIL. Language, Culture and Curriculum, 28, 74-89.   Lincoln, Y., & Guba, E. G. (1985). Naturalistic inquiry. Beverly Hills: Sage. Littlewood, W., & Yu, B. (2009). First language and target language in the foreign language classroom. Language Teacher, 42, 1–14.  218  Lui, P-H., & Tannacito, D. (2013). Resistance by L2 writers: The role of racial and language ideology in imagined community and identity investment. Journal of Second Language Writing, 22, 355-373. Macken-Horarik, M. (2002). “Something to shoot for’’: A systemic functional approach to teaching genre in secondary school science. In Johns, A. (Ed.), Genre in the classroom: Multiple perspectives (pp. 17-42). Mahwah, NJ: Lawrence Erlbaum. Mann, S. (2001). Alternative perspectives on the student experience: Alienation and engagement. Studies in Higher Education, 26, 7-19. Marshall, C., & Rossman, G. (2016). Designing qualitative research (6th ed.). Thousand Oaks, CA: Sage. Marshall, S., & Walsh Marr, J. (2018). Teaching multilingual learners in Canadian writing-intensive classrooms: Pedagogy, binaries, and conflicting identities. Journal of Second Language Writing, 40, 32-43. Martin, J. R. (2000). Design and practice: Enacting functional linguistics. Annual Review of Applied Linguistics, 20, 116-126. Martin, J. R., & Rose, D. (2003). Working with discourse: Meaning beyond the clause. London: Continuum.   Matsuda, P. K. (2003). Process and post-process: A discursive history. Journal of Second Language Writing, 12, 65–83. Matsuda, P. K., Fruit, M., & Burton Lamm, T. (2006). Second language writers and writing program administrators. Writing Program Administration, 30, 11-14. Matsuda, P. K., & Silva, T. (Eds.). (2005). Second language writing research: Perspectives on the process of knowledge construction. New Jersey: Lawrence Erlbaum. McCarthy, L. (1987). A stranger in strange lands: A college student writing across the curriculum. Research in the Teaching of English, 21, 233-265. McKay, S. L. (1993). Examining L2 composition ideology: A look at literacy education. Journal of Second Language Writing, 2, 65-81. McKay, S. L. (2006). Researching second language classrooms. Mahwah, NJ: Lawrence Erlbaum. McMillan, B., & Turnbull, M. (2009). Teachers’ use of the first language in French immersion: Revisiting a core principle. In M. Turnbull & J. Dailey-O’Cain (Eds.), First language use in second and foreign language learning (pp. 15–34). Bristol: Multilingual Matters.  219  Melles, G., Millar. G., Morton, J., & Fegan, S. (2005).