Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Acquisition of English and French language proficiency and reading comprehension of multilingual students… Bérubé, Daniel 2013

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

Item Metadata

Download

Media
24-ubc_2013_fall_berube_daniel.pdf [ 1.5MB ]
Metadata
JSON: 24-1.0073901.json
JSON-LD: 24-1.0073901-ld.json
RDF/XML (Pretty): 24-1.0073901-rdf.xml
RDF/JSON: 24-1.0073901-rdf.json
Turtle: 24-1.0073901-turtle.txt
N-Triples: 24-1.0073901-rdf-ntriples.txt
Original Record: 24-1.0073901-source.json
Full Text
24-1.0073901-fulltext.txt
Citation
24-1.0073901.ris

Full Text

 i Acquisition of English and French Language Proficiency and Reading Comprehension of Multilingual Students in French Immersion Programs: A 3-Year Longitudinal Study by Daniel Bérubé  B.A., University of Manitoba, 2003 M.Sc., University of Ottawa, 2005  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Audiology and Speech Sciences)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  June 2013  © Daniel Bérubé, 2013  ii Abstract  This dissertation consists of three articles that explore the oral language proficiency and reading skills in the second language (English) and third language (French) of multilingual students in Canadian French immersion programs. Oral language proficiency included measures of vocabulary knowledge and listening comprehension. In this dissertation, multilingual students are children who are exposed to a home language other than English or French. Chapter 2 shows that multilingual students developed oral language proficiency and reading skills differently from bilingual English-French students in Grades 4 and 6. This emphasizes the importance of studying multilingual learners in greater detail, and several factors such as first language and the interdependence between the L2 and L3 are identified as contributors of English and French proficiency. Chapter 3 examines the relationship between first language typology—defined as the classification of languages according to their structural characteristics (e.g. phonological and writing systems)—and the development of English and French language proficiency and literacy skills in groups of students who are either literate in an alphabetic first language (e.g. Spanish) or literate in a logographic/syllabary first language (e.g. Chinese). Results indicated that students with an alphabetic first language showed advantages in reading comprehension. However, there was no difference in word reading and pseudoword reading. A more accurate picture of what facilitates reading in English and French is enhanced when differences in oral language proficiency are also considered. It was found that vocabulary knowledge had a greater influence on reading comprehension in both English and French in students who were  iii literate in an alphabetic first language rather than a logographic first language. The fourth chapter further explores other factors associated with multilingual students’ oral language proficiency and reading skills in French: socio-linguistic factors (e.g. motivation, socio- economic status, amount of reading in French), metalinguistic awareness factors (e.g. morphological awareness), and language proficiency factors (e.g. vocabulary knowledge and listening comprehension). Controlling for amount of reading in French and morphological awareness in English, regression analyses showed that oral language proficiency and reading comprehension skills in English were the best predictors of oral language proficiency and reading comprehension skills in French.  iv Preface  Chapters 2, 3, and 4 can be considered as stand-alone papers. A version of chapter 3 has been published. The citation for the first publication is: Bérubé, D., & Marinova- Todd, S. H. (2012). The development of language and reading skills in the second and third languages of multilingual children. International Journal of Multilingualism, 9(3), 272-293. A version of chapter 4 is under review. The citation for the second publication is: Bérubé, D., & Marinova-Todd, S. H. (under revision). The effect of socio-linguistic factors and English language proficiency on the development of French as a third language. My contribution was in the formulation of the research questions, the design of the studies, the data collection, the data analyses, the interpretation of the results, and the manuscript preparation. Dr. Stefka H. Marinova-Todd advised me on the design of the studies, interpretation of the results and the manuscript preparation of both chapters 3 and 4. The responsible UBC Research Ethics board for my dissertation was the UBC Behavioral Research Ethics Board and the two project titles were a) Multilingual students in French immersion programs (ethics certificate # H11-01424) and b) ESL students in French immersion programs (ethics certificate # H08-03048).  v Tables of Contents  Abstract .............................................................................................................................. ii Preface ............................................................................................................................... iv Tables of Contents............................................................................................................. v List of Tables .................................................................................................................. viii List of Figures ................................................................................................................... xi Acknowledgements ......................................................................................................... xii Chapter 1: Introduction ................................................................................................... 1 Chapter 2: Word reading and oral language predictors of English and French reading comprehension in French immersion programs: A 3-year longitudinal study ............................................................................................................................................. 9 2.1  Synopsis .............................................................................................................. 9 2.2  Introduction ..................................................................................................... 10 2.3  Method ............................................................................................................. 24 2.3.1  Participants .................................................................................................... 24 2.3.2  Materials ....................................................................................................... 29 2.3.3  Procedure ...................................................................................................... 31 2.4 Results .............................................................................................................. 32 2.4.1 Comparisons between L2 and L3 oral language proficiency and L2 and L3 reading skills ............................................................................................................. 32  vi 2.4.2  The effect of word reading and oral proficiency in the L2 and L3 on reading comprehension in the L2 and L3 .............................................................................. 41 2.5 Discussion......................................................................................................... 47 Chapter 3: First language typology and the development of oral language proficiency and reading skills in the L2 and L3 of multilingual students ................. 56 3.1  Synopsis ............................................................................................................ 56 3.2  Introduction ..................................................................................................... 57 3.3  Method ............................................................................................................. 64 3.3.1  Participants .................................................................................................... 64 3.3.2  Participants from the Grade 6 sample ........................................................... 68 3.3.3  Materials ....................................................................................................... 70 3.3.4 Procedure ...................................................................................................... 73 3.4  Results .............................................................................................................. 74 3.4.1  The effect of L1 typology on oral proficiency in L2 and L3 ........................ 74 3.4.2  The effect of L1 typology on reading in L2 and L3 ..................................... 78 3.4.3  Comparisons between oral proficiency and reading in L2 and L3 ............... 84 3.4.4  The effect of oral proficiency and reading in L2 and L3 .............................. 87 3.5  Results from the Grade 6 sample .................................................................. 96 3.5.1  Comparisons between Grade 4 and Grade 6 samples on oral proficiency and reading in L2 and L3 ............................................................................................... 102 3.6  Discussion....................................................................................................... 107 Chapter 4: The effect of socio-linguistic factors and English language proficiency on the development of French as a third language ......................................................... 117  vii 4.1  Synopsis .......................................................................................................... 117 4.2  Introduction ................................................................................................... 118 4.2.1 Neurophysiological factors ......................................................................... 121 4.2.2  Learner-external factors .............................................................................. 122 4.2.3  Affective factors: Motivation to learn a third language .............................. 123 4.2.4  Metalinguistic factors: Morphological Awareness in L2............................ 124 4.2.5  Linguistic factors: Proficiency in the second language .............................. 126 4.3  Method ........................................................................................................... 128 4.3.1 Participants .................................................................................................. 128 4.3.2 Materials ..................................................................................................... 129 4.3.3  Procedure .................................................................................................... 132 4.4  Results ............................................................................................................ 133 4.5  Discussion....................................................................................................... 144 Chapter 5: Conclusion .................................................................................................. 151 References ...................................................................................................................... 160 Appendices ..................................................................................................................... 177 Appendix A: Parent questionnaire .......................................................................... 177 Appendix B: Motivation to learn French questionnaire ....................................... 185 Appendix C: Derivational morphology test ............................................................ 186    viii List of Tables  Table 2.1 Mean scores (standard deviations) on socio-linguistic measures in English and French ........................................................................................................................ 28 Table 2.2 Test means (standard deviations) on English and French oral language tests and reading tests at T1 (beginning of Grade 4), at T2 (end of Grade 4), and at T3 (end of Grade 6) ........................................................................................................ 36 Table 2.3 Correlations between L2 and L3 reading tasks and L2 and L3 proficiency measures at the beginning of Grade 4 (T1) and the end of Grade 6 (T3) for the bilingual (N = 57) and the multilingual (N = 57) groups .......................................... 38 Table 2.4 Effects of group, time, and group x time interaction on measures of oral language proficiency and reading ability in English and French .............................. 39 Table 2.5 Multiple regression models representing cross-language effects on reading comprehension in English ......................................................................................... 43 Table 2.6 Multiple regression models representing cross-language effects on reading comprehension in French. ......................................................................................... 46 Table 3.1  Demographic information on the two multilingual L1 groups and the bilingual group at the beginning of Grade 4 (T1), the end of Grade 4 (T2), and at the end of Grade 6 (T3) .............................................................................................................. 66 Table 3.2 Mean standard scores and F test comparisons on measures of listening comprehension and vocabulary knowledge at the beginning of Grade 4 (T1), the end of Grade 4 (T2), and the end of Grade 6 (T3) according to L1 typology on L2 (English) and L3 (French) proficiency tests .............................................................. 75  ix Table 3.3 Mean standard scores and F tests on measures of word reading and reading comprehension at the beginning of Grade 4 (T1), the end of Grade 4 (T2), and the end of Grade 6 (T3)  according to L1 typology on L2 (English) and L3 (French) reading tests ............................................................................................................... 79 Table 3.4 Mean standard scores (SD) and t test in each of the L2 and L3 tests of oral proficiency and reading tasks at the beginning of Grade 4 (Time 1) and at the end of Grade 6 (Time 3): Alphabetic and logographic/syllabary L1 groups ....................... 85 Table 3.5 Correlations between L2 and L3 reading tasks and L2 and L3 proficiency measures at the end of Grade 4 (T2) for the alphabetic L1 (N = 20) and the logographic/syllabic L1 (N = 13) groups .................................................................. 88 Table 3.7 Correlations between L2 and L3 reading tasks (standard scores) and L2 and L3 proficiency measures (standard scores) at end of Grade 6 (T3) for the alphabetic L1 (N = 23) and the logographic/syllabic L1 (N = 12) groups ................................. 99 Table 3.8 Language regression equations predicting English broad reading and French broad reading for all participants at end of Grade 6 ................................................ 101 Table 3.9 Mean standard scores from the eight students with a logographic/syllabic L1 on measures of English and French oral proficiency and reading tests at the beginning of Grade 4 (T1), the end of Grade 4 (T2), and the end of Grade 6 (T3) 105 Table 4.1 Mean scores (standard deviations) for all variables in English and French . 135 Table 4.2 Correlations between the socio-linguistic factors, morphological awareness, oral proficiency, and reading comprehension scores in English and French at the end of Grade 6 (N = 55) ................................................................................................. 138  x Table 4.3 Multiple regression models including all socio-linguistic variables predicting oral language proficiency in French and reading comprehension in French .......... 139 Table 4.4 Summary of multiple regression analyses for variables predicting oral language proficiency in French ............................................................................... 141 Table 4.5 Summary of multiple regression analyses for variables predicting reading comprehension in French ........................................................................................ 143         xi List of Figures  Figure 2.1 English reading comprehension scores at T1 (beginning of Grade 4), T2 (end of Grade 4), and T3 (end of Grade 6) for the bilingual students (Group 1, N = 57) and the multilingual students (Group 2, N = 57). Points from the bold lines represent the mean score for each group on the English reading comprehension test; vertical lines depict standard deviations of the means. The thinner lines represent the scores of each participant at T1, T2, and T3 ........................................................................ 34 Figure 2.2 French reading comprehension scores at T1 (beginning of Grade 4), T2 (end of Grade 4), and T3 (end of Grade 6) for the bilingual students (Group 1, N = 57) and the multilingual students (Group 2, N = 57). Points from the bold lines represent the mean score for each group on the French reading comprehension test; vertical lines depict standard deviations of the means. The thinner lines represent the scores of each participant at T1, T2, and T3 ........................................................................ 35 Figure 3.1 Interaction of language group by French vocabulary scores (standard scores) at the beginning of Grade 4 (T1) on English reading comprehension (standard scores) at the end of Grade 4 (T2) ............................................................................. 94 Figure 3.2 Interaction of language group by French vocabulary scores (standard scores) at the beginning of Grade 4 (T1) on French reading comprehension (standard scores) at the end of Grade 4 (T2) ......................................................................................... 95   xii Acknowledgements  I would like to express my deepest gratitude to my mentor and supervisor, Dr. Stefka Marinova-Todd, whose expertise and passion, and unwavering support have been significant factors in my graduate experience at UBC. This dissertation would not have been possible without your guidance, your support, and your encouragement. Thank you for making me feel like I can soar! I would like to sincerely thank Dr. May Bernhardt and Dr. Lee Gunderson for their valuable expertise, mentorship, and friendship. We have shared some really good times and you were there for me when I was trudging through rougher seas. A special thank you to Monique Charest, Paola Collozzo, and Glenda Mason for sharing their PhD experiences and for allowing me to de-stress! I would like to thank all the research assistants, students, teachers, parents, and school principals who have participated in my research. I would like to acknowledge and thank Dr. Theo Koulis and Richard Gagnon for their statistical expertise. A big thank you to my friends in Manitoba and in British Columbia who were there when I needed to unwind and to get away from it all! Most importantly, I would like to thank my wife – you are my confidante, the one I lean on, and the one who inspires me. To my parents and brother: you have shown unwavering support and love. These simple words cannot thank you enough.      1 Chapter 1: Introduction  The educational landscape in Canada changed in 1965 when the first French immersion (FI) program was created in St. Lambert, Québec. Prior to FI programs, Anglophone-speaking students in Quebec were schooled in English only. In this new FI model, students received instruction only in French beginning in Kindergarten before being taught how to read and write in English, usually between Grade 2 and Grade 4. French immersion programs quickly proliferated throughout the province of Québec, then to Ontario, and then across Canada. At about the same time that FI programs became prevalent throughout Canada, the government of former Prime Minister Pierre Elliot Trudeau created the Official Languages Act in 1969, which recognized the need for services, including education, to be offered in both English and in French. In response to the growing number of students receiving this new form of instruction, evaluations of FI programs, many of which were conducted in the 1970s and 1980s, documented the success of these students in developing oral language proficiency and reading comprehension skills in French as a second language (L2) and in English. Although some delays in French reading comprehension and French vocabulary comprehension were noted (e.g. Genesee, 1978; Geva & Clifton, 1994), the French listening comprehension skills of early French immersion students was measured to be equally strong to that of French monolingual students by Grade 6 (Genesee, 1978). Bilingual English-French students continued to demonstrate significant gaps in writing skills when compared with French monolingual students (Harley, Allen, Cummins, & Swain, 1991). Comparisons between bilingual English-French students and French     2 monolingual students were completed primarily in Montreal in early French immersion programs, where students were immersed in French at school and in the community. Immersion students achieved parity with English monolingual students in English oral language proficiency (vocabulary knowledge and listening comprehension), English word reading, and English reading comprehension, within one to two years of formal English instruction (e.g. Genesee, 1983; Lambert & Tucker, 1972; Swain & Lapkin, 1982). In some cases, immersion students even demonstrated advantages in English reading comprehension (e.g. Turnbull, Lapkin, & Hart, 2003). According to these researchers, bilingual students obtained sufficiently high levels (i.e. reached a threshold) of oral language proficiency in both English and in French, which allowed them to exhibit stronger reading comprehension skills in English to that of monolingual Anglophone students. Overall, FI programs promoted the learning of a language (French) through content-based teaching with the central premise of fostering biliteracy skills of students without hindering their development of English, the first/home language (L1). A significant body of empirical research in FI programs (e.g. Genesee, 1983; Lambert & Tucker, 1972; Turnbull et al., 2003) confirmed that in contexts where English and French were highly valued by families and by the broader society, the main goal of FI programs was being achieved, namely students were learning and prospering in French and in English (Auerbach, 1993). In this dissertation, oral language proficiency refers to measures of vocabulary knowledge and listening comprehension. Immigration to Canada increased steadily starting in the latter half of the 1960s and considerable changes in Canadian demographics were felt in all schools, including FI programs. In response, the Trudeau government felt that “biculturalism” inadequately     3 represented the needs of every Canadian. They promoted “multiculturalism,” which was first acknowledged in federal government policies in 1971 and lead to the creation of the Canadian Multiculturalism Act in 1988. To this day, multiculturalism embodies the notion that all Canadians have the right to preserve their multicultural heritage, including language. Historically, the L2 taught in FI programs has been French, because the majority of students were from middle-class Anglophone backgrounds. Today, an estimated 50% of the total student population in Canada comprises multilingual students whose L1 is not English (e.g. Swain & Lapkin, 2005; Vancouver School Board, 2011). Many students who enroll in FI programs, especially in major urban centers, speak a home language other than English or French. In this dissertation, the first/home language (L1) of multilingual students refers to the language, other than English or French, to which the child was exposed at birth and is used regularly in the home. The L2 refers to English, the language some children may have been exposed to at birth, but which is not used regularly in the home. The L3 refers to French, the immersion language first introduced to multilingual children at school. This shift in the educational landscape changes the fundamental feature that defines FI programs: teaching in a student’s home language. Unlike Anglophone-speaking students who are learning French in a school environment that fosters their L1, multilingual students in FI programs are often not receiving direct instruction in their first language. Instead, multilingual students are learning English and French in an educational context that does not directly support their home language. As a result, they may be at an increased risk of reading difficulties. Researchers (e.g. Cummins, 2000) have argued that skills acquired in the L1 provide a solid foundation that     4 facilitates the development of an L2. Alternatively, Cummins predicted that when the L1 is not sufficiently supported in the schools, it remains underdeveloped and can be a limiting factor in the development of subsequent languages. On a practical level, and in light of the increase of multilingual students in FI programs, the overarching question posed in this thesis is—do today’s French immersion programs meet the needs of bilingual and multilingual students in Grade 4 and Grade 6 in a major urban center in Western Canada? This dissertation is organized as follows. In the second chapter, I reported on a longitudinal study that examined the relative success of the oral language proficiency and the reading skills in the L2 (English) and L3 (French) of bilingual and multilingual students enrolled in early FI programs. Numerous theoretical models of English reading comprehension have been proposed (see National Reading Panel, 2000 for a complete review); by contrast, reading models for students acquiring an L2 or an L3 are few (Jessner, 2008; Proctor, August, Carlo, & Snow, 2006). The research questions addressed in the second chapter are framed by the simple view of reading (Hoover and Gough, 1990; Proctor et al., 2006) which states that reading comprehension is best predicted by a combination of decoding skills (e.g. word reading skills) and oral language proficiency (e.g. listening comprehension skills and vocabulary knowledge). Other researchers (Savage, 2001; Storch & Whitehurst, 2002) have shown that over time, when decoding skills are established, oral language proficiency takes on a primary role in explaining reading comprehension. The simple view of reading model was used as the framework for explaining the acquisition of reading comprehension skills in the L2 and the L3 of multilingual students.     5 The study in chapter 2 focused on the cross-language association between oral language proficiency and word reading skills in the L2 and the development of reading comprehension skills in the L3. Likewise, the cross-language transfer of oral language proficiency and word reading skills in the L3 and the acquisition of reading comprehension skills in the L2 was investigated. Transfer involves the influence of the source language on the acquisition of the target language (Odlin, 1989). Previous studies conducted with bilingual and multilingual students suggested that the relative role played by the language of instruction (amount of exposure) may influence the direction of transfer between the oral language proficiency and the reading skills in the L2 and the L3 (Bérubé & Marinova-Todd, 2012; Miller et al., 2006). Recent complementary perspectives on the role of transfer have also recognized that common underlying cognitive processes such as speed of processing and working memory may have an important influence in determining the extent of language transfer (Genesee, Geva, Dressler, & Kamil, 2008). The current study further examined the relative extent of transfer from the L2 to the L3 and from the L3 to the L2 in multilingual students. In FI classrooms such as those described in this study, teaching practices were aimed at content-based learning of oral language proficiency (vocabulary knowledge and listening comprehension) in French and English as a support for fostering the reading skills in both languages. Therefore, the research objectives in chapter 2 reflected the teaching practices implemented in FI classrooms. Researchers have identified that L1 typology (specifically the writing systems) influences the acquisition of oral language proficiency and reading skills in the L2 of bilingual students (Koda, 1989) and in the L3 of multilingual students (Swain, Lapkin,     6 Rowen, & Hart, 1990). Other researchers (Cummins, 2000; Gunderson, 2007) have maintained that there is a common and interdependent association between the language skills of the L1 and L2 which influences the L2 reading achievement of bilingual students. Therefore, in the third chapter, I present an empirical study that was designed to further explore multilingual students in FI programs who are literate in their L1 and to understand the relationship between L1 typology and the development of English (L2) and French (L3) oral language proficiency and literacy skills. Oral language proficiency included measures of vocabulary knowledge and listening comprehension. In addition to language typology, other socio-linguistic factors and meta-linguistic factors have been found to contribute to the development of reading skills in the L3 of multilingual students (Hufeisen & Marx, 2007; Sanz, 2008). To account for differences between bilingual and multilingual students, Hufeisen and Marx (2007) proposed that whereas the L2 student is a novice learner in the process of L2 acquisition, the L3 student has already learned about the process of learning a second language. Therefore, L3 students have language-specific knowledge and abilities at their disposal that L2 students typically do not. In line with Marx and Hufeisen’s (2007) ‘factor model’, chapter 4 of the dissertation examined additional factors that predict French (L3) reading comprehension skills in multilingual students in Grade 6. These include socio-linguistic factors (e.g. motivation, socio-economic status, amount of reading), metalinguistic factors (e.g. morphological awareness), and language factors (e.g. vocabulary knowledge and listening comprehension). The field of third language acquisition has recently focused on the attainment of L3 skills of multilingual students from both theoretical and practical perspectives. A     7 review of the literature has revealed that compared with bilingual students, multilingual students develop stronger L3 skills in vocabulary knowledge, listening comprehension, and reading comprehension when their home language is well supported at school (Brohy, 2001; Cenoz, 2003; Cenoz & Valencia, 1994; Errasti, 2003; Munoz, 2000; Sanz, 2000, 2008). Several studies, by contrast, involving groups of multilingual students whose L1 is not taught in school have shown mixed results; some have shown no advantages for multilingual students (Sanders and Meijers, 1995), and others have shown that multilingual students obtained lower oral language proficiency and literacy skills in the L3 as compared with bilingual students (Schoonen et al., 2002; Van Gerlderen et al., 2003). The conflicting results from the aforementioned studies conducted with multilingual students raise questions as to what additional factors influence L3 acquisition. Moreover, the development of their L2 skills in this context has been largely overlooked. In response, the literature has shown the need to better understand the mechanism by which home language literacy influences the acquisition of reading skills in multilingual students, not only in their L3 but also in their L2 (e.g. Genesee & Jared, 2008). In a Canadian context in particular, the Council of Ministers of Education, Canada (CMEC, 2009) has identified a significant gap in our understanding of literacy achievement of multilingual students and recommended more longitudinal research on the reading outcomes of multilingual students in early FI programs. By examining the oral language and reading outcomes in both English (L2) and French (L3) of multilingual students in Grades 4 and 6, and explaining their literacy achievement, this dissertation aimed to provide important theoretical and practical     8 implications for multilingual students enrolled in FI programs, and more broadly for the field of third language acquisition. More specifically, the purpose of this dissertation was to provide more information to understand which word reading skills and oral language proficiency skills (vocabulary knowledge and listening comprehension) in the L2 and L3 were important in the acquisition of reading comprehension in the L2 and L3. Furthermore, this dissertation investigated whether measures of word reading and oral language proficiency in Grade 4 that predicted reading comprehension in Grade 6 for multilingual students were the same for the bilingual students. The last chapter of this dissertation discusses the results in a more general context, focusing on the major findings of the dissertation, the educational applications in the broader field of third language acquisition, limitation of the dissertation, and future research.         9 Chapter 2: Word reading and oral language predictors of English and French reading comprehension in French immersion programs: A 3-year longitudinal study  2.1  Synopsis  The main goal of the study reported in this chapter was to examine whether multilingual students in Canada, who are exposed to a language other than English or French at home, develop the same oral language proficiency and reading skill, in English and French, as bilingual English-French students in French immersion (FI) programs. The sample included 57 bilingual students and 57 multilingual students enrolled in FI programs. The relationship between English oral language proficiency in Grade 4 and French reading comprehension in Grade 6, as well as the relationship between French oral language proficiency in Grade 4 and English reading comprehension in Grade 6 were investigated. Individual growth modeling revealed that multilingual students developed reading skills that were as strong as those of bilingual students in both English and French. Multilingual students demonstrated weaker vocabulary knowledge and listening comprehension skills both in Grade 4 and 6 in English but equivalent oral language proficiency in French. French vocabulary knowledge was a stronger predictor of English reading comprehension for the multilingual students. These findings suggest that multilingual students in FI programs can develop strong reading skills in English and French that are equivalent to those of bilingual students.       10 2.2  Introduction  The number of multilingual students in French immersion (FI) programs across Canada is growing (Swain & Lapkin, 2005). In this paper, multilingual students’ first/home language (L1) is used to refer to the language, other than English or French, to which the child was exposed at birth and is used regularly in the home. The second language (L2) refers to English, the dominant societal language some children may have been exposed to at birth, but not used regularly in the home, and the third language (L3) refers to French, the language multilingual children were first exposed to at school entry. Anglophone students in this study have been acquiring French as an L2 since school entry (Kindergarten) and I will refer to this group of children as bilingual students. Multilingual students in FI schools in Canada are learning English and French in an educational context that does not support their L1, and may be at an increased risk of reading difficulties. According to Cummins’ Interdependence hypothesis, skills acquired in the L1 provide a solid foundation that facilitates the development of an L2. Alternatively, Cummins predicts that when the L1 is not sufficiently supported in the schools, it remains underdeveloped and can be a limiting factor in the development of subsequent immersion languages (Cummins, 2000). The purpose of FI programs is to support the literacy in English and French of all students; thus, it remains important to identify the oral language skills that facilitate reading acquisition in both languages, especially with students whose L1 are different from English and French. A number of studies have examined the oral language proficiency and reading skills in the L3 of multilingual students (Bild & Swain, 1989; Cenoz, 2003; Cenoz &     11 Valencia, 1994; Sanders & Meijers, 1995; Sanz, 2000, 2008). However, currently only two studies have examined the oral language proficiency and reading skills of multilingual students in both their L2 and their L3 (Bérubé & Marinova-Todd, 2012; Van Gelderen et al., 2003). These two studies involved multilingual students who were not taught in their L1 at school. The literature review will consist of the following two sections: a) An examination of the oral language proficiency and the reading comprehension skills in the L3 of multilingual students in contexts when the L1 was either supported in school or when the L1 was not directly taught in school. The first section also includes a review of two Canadian studies conducted with multilingual students in FI programs (Bérubé & Marinova-Todd, 2012; Bild & Swain, 1989). b) An exploration of the research on cross-language predictors of reading comprehension conducted with bilingual and multilingual students. More specifically, this dissertation will discuss the cross-language association between the oral language proficiency in the L1 and the reading comprehension skills in the L2 with bilingual students. The discussion will provide a context for research on the cross-linguistic relationship between the oral language proficiency in the L2 and L3 of multilingual students and their reading comprehension skills in the L2 and L3. The fact that so few studies have examined multilingual students in the North American context is surprising given the increasing number of immigrant students, and particularly those enrolled in FI programs in Canada (Genesee & Jared, 2008; Taylor, 2006). Until the 1990s, the majority of students in Canadian FI programs were from middle-class Anglophone backgrounds; and numerous studies have documented the     12 success of these students in developing proficiency in English and French as an L2 (Genesee, 1983; Turnbull, Lapkin, & Hart, 2003). Today, an estimated 50% of the total student population in major urban cities in Canada is comprised of multilingual students whose L1 is not English (Swain & Lapkin, 2005; Vancouver School Board, 2011). However, little is known about the development of oral language proficiency and reading skills in both the L2, English, and the L3, French, of multilingual students in FI programs. An increasing number of studies have examined the effect of bilingualism on L3 acquisition in school-aged children in an educational context that supports the L1 of multilingual students (Brohy, 2001; Cenoz, 2003; Cenoz & Valencia, 1994; Errasti, 2003; Munoz, 2000; Sanz, 2000, 2008). The researchers have explored whether multilingual students develop the same level of oral language proficiency and literacy skills as those of bilingual children. All studies were conducted with multilingual students whose L1 and L2 were both official languages, such as Basque-Spanish, Catalan-Spanish, or German-English bilinguals, and directly supported in school. In these contexts, multilingual children between Grades 6 to 10 showed advantages relative to bilingual students in vocabulary knowledge, listening comprehension, reading comprehension and writing when acquiring an L3, such as English or French. Studies in the Spanish, Basque, Catalan, and German settings have led researchers to conclude “that (additive) bilingual learners are at an advantage in L3 acquisition” (Haenni Hotti, Heinzmann, Muller, Oliviera, & Werlen, 2011, p. 100). By contrast, the majority of studies involving groups of multilingual students whose L1 is neither taught nor supported in school (for example, Arabic, Berber,     13 Moroccan, Surinam, and Turkish) have shown that multilingual students have equivalent or inferior oral language proficiency in the L3 when compared with bilingual students (Bérubé & Marinova-Todd, 2012; Bild & Swain, 1989; Bos & Pietsch, 2006; Kleime, 2006; Sanders & Meijers, 1995; Schoonen et al., 2002; Van Gelderen et al., 2003). With the exception of the Van Gelderen et al. (2003) study, only the L3 of multilingual students was examined. In these studies, the students’ L1 was not an official language in the community and, although there may have been a few hours of teaching in the minority language, it was not the principal language of instruction in the schools. Within a North American context, and more specifically in Canada, Bild and Swain (1989) was the first study that examined the effects of bilingualism on the development of the oral language proficiency and reading comprehension skills of multilingual students. These students were tested in French only, their L3. The research included Grade 8 students from three language groups: multilingual Romance L1 (Italian), multilingual non-Romance L1 (all alphabetic languages), and bilingual English- French students. The study involved students in late FI programs acquiring English as an L2 and French as an L3 without formal instruction in the L1 in school. Students enrolled in the late FI programs received English instruction from kindergarten to Grade 4, and then received French instruction from Grade 5 to Grade 8. No differences were found on measures of French academic language proficiency (‘context reduced’ written cloze task); by contrast, the two multilingual groups showed stronger French conversational skills than the bilingual students, relating to grammatical competence, discourse organization, and pronunciation. Bérubé and Marinova-Todd (2012) further examined the relationship between L1 typology (classification of languages according to phonological     14 systems and writing systems) and the development of oral language proficiency and reading skills in the L2 (English) and L3 (French) of multilingual students enrolled in FI programs. The study included Grade 4 students who belonged to one of three language groups: bilingual English-French students, multilingual students who were literate in an alphabetic L1 (for example Spanish), and multilingual students who were literate in a logographic/syllabary L1 (for example Chinese). Findings revealed that multilingual students who were literate in an alphabetic L1 showed equivalent proficiency in vocabulary knowledge, listening comprehension, and reading comprehension in English and French to the bilingual students, but stronger skills than multilingual students who were literate in a logographic/syllabary L1. Those students who were literate in a logographic/syllabary L1 had weaker oral language and reading skills than the bilingual students. These results suggested that oral language proficiency and literacy in the L2 and L3 may be influenced by the orthographic similarities with the L1. The findings from the two studies conducted within FI programs in Canada revealed that multilingual students can develop equally strong French skills (their L3) as those of bilingual students (their L2) when the typology between the L1 and L3 is similar (i.e. both are alphabetic languages). It is possible that multilingual students in the Bild and Swain (1989) study showed advantages in French conversation skills in part because the students were more immersed in French (greater exposure) both in school and in their communities. The multilingual students in the Bérubé and Marinova-Todd (2012) study lived in communities where French exposure was limited mainly to the classroom. Other studies have found bilingual students to have stronger oral language proficiency and reading skills than their multilingual counterparts (Bos & Pietsch, 2006;     15 Schoonen et al., 2002; Van Gerlderen et al., 2003). Van Gelderen et al. (2003) compared the performance of immigrant students in Grade 8 from a variety of L1 backgrounds (North African languages: Berber, Arabic, and Turkish) who were learning Dutch as an L2 and English as an L3 with that of bilingual Dutch students who were learning English as an L2. Findings showed higher English reading comprehension skills for the bilingual group, after 3.5 years of formal English reading instruction. There were no group differences in English vocabulary knowledge and in grammatical knowledge. The bilingual students also showed greater reading comprehension skills in Dutch than the multilingual students, after seven years of formal Dutch reading instruction. Similarly, Bos and Pietsch (2006) showed that multilingual (immigrant) students in Grade 8 whose parents were born outside of Germany had lower listening comprehension skills in English compared with bilingual German students learning English as an L2. Schoonen et al. (2002) also found significant advantages for bilingual Dutch students acquiring English as an L2 with respect to vocabulary and writing ability (global proficiency index) when compared with Arabic or Turkish multilingual students learning English as an L3. All studies showing bilingual students with stronger oral language proficiency and reading skills than their multilingual counterparts were conducted in educational contexts where the L3 was acquired as a foreign language (Bos & Pietsch, 2006; Schoonen et al., 2002; Van Gerlderen et al., 2003). In these settings, the multilingual students were not immersed in the L3 (they received only a few hours of instruction per week) and they received no support in their L1. In the FI programs in Canada, multilingual students are not receiving direct support in their L1, but all students are immersed in the L3. The difference in amount of exposure and immersion in the L3 may have an influence of the     16 ultimate attainment of language proficiency. The cross-language associations between measures of oral language proficiency in the L1 (vocabulary knowledge and listening comprehension) and reading comprehension in the L2 will first be reviewed in bilingual students in the next section. Then, cross-language predictors between measures of oral language proficiency in the L2 and L3 and reading comprehension in the L2 and L3 will be examined in multilingual students.  Cross-language predictors of reading comprehension   Several models of English reading comprehension have been proposed (see National Reading Panel, 2000, and Snow, Burns, & Griffin, 1998 for a complete review); by contrast, reading models for students acquiring an L2 or an L3 are few (Jessner, 2008; Proctor, August, Carlo, & Snow, 2006). Hoover and Gough (1990) proposed a simple model of reading according to which reading comprehension was predicted by a combination of decoding skills and oral language proficiency (listening comprehension skills). Several studies have since provided empirical evidence for this model with English-monolingual students (see the National reading panel, 2000 for review) and for students learning to read in an L2 (Droop & Verhoeven, 2003; Proctor et al., 2006; Verhoeven & Leeuwen, 2012). Additional studies have shown  that other measures of oral language proficiency (e.g. vocabulary knowledge and syntactic knowledge) also predict L2 reading comprehension (Droop & Verhoeven, 2003; Proctor et al., 2006). Moreover, when decoding skills are established in proficient readers, oral language     17 proficiency takes on a primary role in predicting reading comprehension (Savage, 2001; Storch & Whitehurst, 2002). Foundational theory in the field of second language acquisition suggests that well- developed skills in the L1 will likely give rise to increased reading comprehension skills in the L2 (Cummins, 1979, 2000). Moreover, language distance – or the relative similarity between the languages, can promote transfer (see Odlin, 1989, for a review). For example, Spanish was more influential than Chinese in the acquisition of English word reading (Bialystok, Luk, & Kwan, 2005), due to similar phonological units between Spanish and English; Czech was more influential than English in the acquisition of Russian vocabulary (Duskova, 1984), due to a greater proportion of shared vocabulary cognates between Czech and Russian. Transfer involves the influence of the source language (in this case, the L1) on the acquisition of the target language (in this case, the L2) (Odlin, 1989). The association between vocabulary knowledge in the L1 and the reading comprehension skills in the L2 and listening comprehension in the L1 and reading comprehension in the L2 have been constrained to studies involving two alphabetic languages that share a common writing system, including Spanish-speaking students acquiring English as an L2 (Carlisle, Beeman, Davis, & Spharim, 1999; Proctor et al, 2006) and Finnish-speaking students learning English as an L2 (Dufva & Voeten, 1999). In each of these studies, vocabulary knowledge in the L1 and listening comprehension in the L1 were found to be associated positively with the reading comprehension skills in the L2. This line of research has been concerned about whether transfer is attributed to shared linguistic knowledge across the two languages of bilingual students (Geva & Verhoeven, 2000). There has not been a formal investigation to date     18 that has directly tested the transfer of oral language proficiency in a non-alphabetic L1 (e.g. Chinese) to the reading comprehension skills in an alphabetic L2 (e.g. English). However, Gunderson (2007) examined the association between the written composition skills in the L1 of 8- to 17-year-old Chinese-speaking students and their reading comprehension skills in the L2 (English). The Chinese-speaking students were categorized into three groups according to literacy instruction: students from Mainland China who were taught to read through the Latin-based alphabet called Pinyin; students from Taiwan who were introduced to characters/symbols that involve non-alphabetic calligraphic forms derived from Chinese characters (called Bopomofo), and students from Hong Kong who learned to read Cantonese that involved a drill system of learning Chinese characters. The author found a weak but positive correlation between the literacy skills in the L1 and L2 (r varied between .06 and .35) in all three groups and suggested that skills may transfer between two languages that do not share a common orthography. Many studies have examined the extent to which oral language proficiency in the L1 transfers to the child’s reading comprehension skills in the L2 (Carlisle et al., 1999; Dufva & Voeten, 1999; Proctor et al., 2006). Proctor et al. (2006) tested the transfer of word reading and oral language proficiency in the L1 on the development of reading comprehension in the L2 of Grade 4 Spanish-speaking students acquiring English as an L2. They found that word reading and vocabulary knowledge in the L1, Spanish, was a significant predictor of reading comprehension in the L2, English, once vocabulary knowledge in English was controlled. They suggested that L1 skills facilitate L2 reading acquisition. This finding of cross-linguistic transfer of oral language proficiency has been replicated a few times for alphabetic languages (Carlisle, Beeman, Davis, and Spharim,     19 1999; Dufva & Voeten, 1999). Carlisle et al. (1999) found that controlling for English vocabulary knowledge (L2), vocabulary knowledge in Spanish (L1) was a significant predictor of reading comprehension in English (L2) in bilingual Spanish-English students in grade 3. Dufva and Voeten (1999) similarly observed that listening comprehension in Finnish (L1) significantly predicted reading comprehension in English (L2), controlling for English listening comprehension in Grade 3 students. In summary, studies conducted with bilingual students whose two languages shared a common writing system showed a positive association between oral language proficiency in the L1 and reading comprehension in the L2. The results of this line of research on transfer are especially important because of the strong belief among educators that oral language proficiency drives the development of reading comprehension (Geva & Zadeh, 2006). Very little cross-linguistic research is available on the development of reading comprehension in multilingual students. Recently, two studies examined the transfer between the oral language proficiency in the L2 and L3 of multilingual students and their reading comprehension in the L2 and L3 (Bérubé & Marinova-Todd, 2012; Haenni Hoti et al., 2011). Haenni Hoti et al. (2011) examined the extent to which listening comprehension in the L2 influenced the listening comprehension skills in the L3. The authors also tested the influence of the reading comprehension skills in the L2 and the reading comprehension skills in the L3. The study involved multilingual German-English students in Grades 3 and 5, as well as a sub-group of multilingual immigrant students whose L1 was a minority language (e.g. Turkish, Portuguese, Italian, and Dutch), learning English as an L2 and French as an L3. Students were being taught in their L1, L2, and L3 simultaneously; however, the L1 and L2 were the primary languages of     20 instruction. Multiple regression models revealed that both listening comprehension and reading skills in the L2 were significant predictors of listening comprehension and reading skills in the L3 for all students in Grade 3 and in Grade 5. Haenni Hoti et al. (2011) did not examine the transfer of listening comprehension and reading comprehension from the L3 to the L2. Bérubé and Marinova-Todd (2012) tested the influence of English (L2) vocabulary knowledge on the acquisition of French (L3) reading comprehension of multilingual students from various L1 backgrounds. The authors also examined the association between French vocabulary knowledge and English reading comprehension of multilingual students. Findings showed that transfer occurred in only one direction, which was the opposite of that found by Haenni Hotti et al. (2011): French oral language proficiency was a significant predictor of English reading comprehension. The difference in findings between the two studies could be explained with the amount of formal academic instruction provided in each language at school. Students in the Haenni Hoti et al. (2011) study were receiving the majority of their instruction in the L2 (English); whereas students in the Bérubé & Marinova-Todd (2012) study were receiving most of their education in the L3 (French). The intense exposure to the primary language of academic instruction may be more predictive of reading comprehension in the other language. To summarize, studies of bilingual students indicate that the oral language proficiency in the L1 predicts reading comprehension in the L2 (Carlisle et al., 1999; Dufva & Voeten, 1999; Proctor et al., 2006). Studies conducted with multilingual students suggest that the relative role played by the language of instruction (amount of     21 exposure) might influence the direction of transfer between the oral language proficiency and the reading skills in the L2 and the L3 (Bérubé & Marinova-Todd, 2012). In addition to the effects of word reading and oral language proficiency on reading comprehension, the level of literacy skills of students learning an L2 or an L3 has been shown to be significantly associated with SES in students throughout elementary grades (Bild & Swain, 1989; Gunderson, 2007) and secondary grades (Gunderson, D’Silva, Murphy, & Odo, 2012; Murphy, Odo, D,Silva, & Gunderson, 2012). Other studies have revealed that the association between SES and reading systematically declined and was non-significant by Grade 3 (D’Angiulli, Siegel, & Maggi, 2004). Therefore, it is essential to control for the effect of SES, given the somewhat conflicting results; this will lead to a better understanding of the nature of L3 reading acquisition (Bild & Swain, 1989). On the basis of the literature reviewed above, we know that word reading and oral language proficiency in the L1 are important factors in the acquisition of reading comprehension in the L2 for bilingual students (Carlisle et al., 1999; Dufva & Voeten, 1999; Proctor et al., 2006). These findings suggest a potential relationship between the word reading skills and oral language proficiency in the L2 and reading comprehension skills in the L3 for multilingual students. In addition, there may be an association between the decoding skills and oral language proficiency in the L3 and reading comprehension skills in the L2 for multilingual students. Since the multilingual students in the current study are not educated in their L1s, it remains important to identify to what extent word reading and oral language proficiency in the L2 (English) facilitate reading comprehension skills in the L3 (French). It is equally important to investigate to what extent word reading and oral language proficiency in the L3 influence reading     22 comprehension skills in the L2. The two studies that examined the transfer of skills between the L2 and L3 (Bérubé & Marinova-Todd, 2012; Haenni Hoti et al., 2011) included groups of multilingual students only and did not compare the transfer effect between groups of bilingual and multilingual students. What remains to be explored is whether the associations between the word reading, oral language proficiency, and reading comprehension skills in the L2 and L3 are similar, or different, between multilingual students and bilingual students. Educators, researchers, and policy-makers (Council of ministers of education, Canada, 2009; Genesee & Jared, 2008; Swain & Lapkin, 2005; Van Gelderen et al., 2003) have identified that multilingual children may develop language and literacy skills differently from bilingual children enrolled in school contexts that do not directly support the L1 of multilingual students. In classrooms where the L1 is not directly taught, potentially, though not necessarily (as discussed in the review of literature), multilingual students may be at risk of developing lower oral language proficiency and lower reading skills in the L2 and L3 than their bilingual counterparts (Swain & Lapkin, 2005). Within a Canadian context, the Council of Ministers of Education, Canada (CMEC, 2009) has identified a significant gap in our understanding of literacy achievement of multilingual students and has recommended the use of longitudinal studies to examine the ‘‘development of reading skills for these students as compared to native-speaker (L1) students’’ (page 38). Moreover, the European Commission on Multilingualism (European Commission, 2008) has developed a language policy that promotes multilingualism and recognizes the need for more research that aims to maintain and promote language diversity.     23 In line with the simple view of reading (Hoover & Gough, 1990) the role of word reading and listening comprehension as indicators of decoding and oral language proficiency were examined in a group of multilingual students. The present study expanded on the simple view of reading model described previously (Hoover & Gough) by supplementing oral language with a measure of vocabulary knowledge, resulting in a model that fit more precisely with the data gathered with the bilingual and multilingual students. By examining the acquisition of oral language proficiency and reading comprehension in English and French, the study aimed to: 1) provide experimental support for FI programs for bilingual and multilingual students and 2) identify oral language skills that are particularly important for English and French reading comprehension in bilingual children as well as for multilingual students in FI programs. While a number of studies have explored the oral language proficiency and literacy skills in the L3 at a single time point (e.g. Bild & Swain, 1989; Cenoz & Valencia, 1994; Sanz, 2008), this is the first longitudinal study to examine the development of oral language proficiency and reading comprehension for multilingual children at three time points in Grades 4 and 6. The current study addressed the following two research questions: 1) Do multilingual students develop equivalent oral language proficiency and literacy skills in English (L2) and French (L3) to the skills of bilingual students between Grade 4 and Grade 6? In line with the existing evidence on the acquisition of oral language proficiency and reading skills in the L2 (Bérubé & Marinova-Todd, 2012; Van Gelderen et al., 2003) and L3 (Bérubé & Marinova-Todd, 2012; Kleime, 2006; Van Gelderen et al., 2003) of multilingual students who were not educated in their L1, we expected that multilingual     24 students would demonstrate equivalent or inferior proficiency on measures of oral language and reading skills in English and French. 2) Which word reading skills and oral language skills in English and French are important in the acquisition of English (L2) reading comprehension and French (L3) reading comprehension? Furthermore, are the measures of word reading and oral language proficiency in Grade 4 that predicted reading comprehension in Grade 6 for bilingual students the same for multilingual students? In line with the study by Bérubé and Marinova-Todd (2012) it was not expected that word reading, vocabulary knowledge, and listening comprehension in English would be related with reading comprehension in French in bilingual and multilingual students. Instead, it was expected that word reading, vocabulary knowledge, and listening comprehension in French would be related with reading comprehension in English in bilingual and multilingual students. No study has examined systematically whether oral language proficiency predicts reading comprehension differently for bilingual or multilingual students; therefore this part of the analysis was a first exploration.  2.3  Method  2.3.1  Participants  A total of 114 students belonging to one of two language groups, 57 multilingual students and 57 bilingual students, participated in the first two stages of this study (T1 at beginning of Grade 4 and T2 at end of Grade 4). All participants were included in the set     25 of analyses that targeted the first research question. The analyses included a series of individual linear models, using the statistical software R (R Development Core Team, 2011). After the end of Grade 4 (T2), 35 students (19 from the multilingual group and 17 from the bilingual group) did not complete any further testing. Ten of the students moved away, ten did not return consent forms, nine were in English programs, and six declined to participate. The reasons for dropping out of the study after Grade 4 were similar for the two language groups. Thus, 79 of the original 114 students were also tested at the end of Grade 6 (T3). Only the 79 participants who were tested at T1 and T3 were included in the series of analyses that targeted the second research question. These analyses included a series of hierarchical regression models using the statistical software R (R Development Core Team, 2011). No significant differences on the Grade 4 measures were found between the children who dropped out of the study and those who continued their participation at the end of Grade 4 (p > .05). The bilingual group used English exclusively at home and was exposed to French and English at school. All bilingual students were born in Canada. The multilingual students reported using the following home languages: Amharic, Afrikaans, Chinese, Croatian, Czech, Danish, Fanti, German, Greek, Hungarian, Japanese, Korean, Persian, Polish, Punjabi, Romanian, Serbian, Spanish, Tagalog, and Vietnamese. All multilingual students were exposed to their L1 from birth; whereas, 70% were also exposed to English within the first year of life. Moreover, multilingual students stated that they used their home language on a daily basis in a variety of settings (e.g. at home, with grand-parents, or with friends). Forty percent of the multilingual students reported that they were able to read in their L1. All students were enrolled in an early French immersion program, had     26 been receiving French instruction since Kindergarten, and began receiving formal English instruction in Grade 4. The principals of all participating schools with French immersion programs reported that their teachers had a high level of language proficiency in French and the majority of teachers were native speakers of French. Between Grade 4 and Grade 6, students in the French immersion programs received French instruction in math, sciences, social sciences, music, and physical education; and received English instruction only during English language arts classes. Background information on the literacy activities in all languages used at home, amount of use in each language in the home and outside the home, and socio-economic variables was collected to establish that the multilingual group and the bilingual group had similar exposure to English and French (see Table 2.1). The questionnaire was adapted from the Alberta Language Environment Questionnaire (Paradis, 2011) and the Parent Interview of Acquiring Literacy in English Study (Duursma et al., 2007) (see Appendix A). Parents reported on their child’s literacy exposure in the L1, L2, and L3 by indicating the number of times their child read in each language (i.e. on a scale, which ranged from 1 = once or twice per month, 3 = two or three times per week, and 5 = every day). The frequency with which the students read independently in English, t(113) = .23, p = .82, d < .001, and in French, t(113) = 1.50, p = .14, d = .02, was equivalent in the two groups. More than half of the children in the multilingual group were literate in their L1 and read in the home language at least twice per month. Only those multilingual children whose parents reported that their child spoke the L1 at home on a daily basis in a variety     27 of settings (e.g. home, with grand-parents, or with friends) were included, and this assured that they had generally strong oral skills in the L1. On average, mothers and fathers completed at least a bachelor degree and answers varied from high school diploma to PhD degree. The level of education completed by mothers and fathers was significantly correlated in the multilingual group (r = .72, p < .001) and in the bilingual group (r = .74, p < .001), and there was no significant difference between education levels for parents of multilingual students, t(54) = 1.05, p = .43, d = .002, and parents of bilingual students, t(48) = .25, p = .80, d = .001. Therefore, a single composite score of ‘parental education’ was created by averaging the number of years of education completed by mothers and fathers.     28 Table 2.1 Mean scores (standard deviations) on socio-linguistic measures in English and French  Note: The amount the child reads in English and French is measured in number of days. A score of ‘3’ for amount a child reads in English and French indicates that the child reads 1 to 2 days a week; and a score of ‘4’ indicates that the child reads 3 to 5 days a week. For English and French language use, parents estimated the amount of time their child spoke English and French on a weekly basis. The number ‘7’ in degree of maternal education indicates that mothers had achieved at least a bachelor’s degree and ‘6’ indicates that mothers had achieved at least some college education or trade school training.                                                                               Multilingual group                                                Bilingual Group  Mean scores (SD) Minimum Maximum Mean scores (SD) Minimum Maximum Length of residency (in years) 10.96 (.22) 10.25 11.00 10.78 (.61) 8.00 11.00 Socioeconomic status (SES)    Mother highest degree    Father highest degree    Parental highest degree Amount child reads English Amount child reads French English language use French language use  6.98 (1.79) 6.92 (1.78) 6.96 (1.57) 4.71 (.64) 4.22 (.86) 77.40 % (22.15%) 34.82% (11.36%)  4 4 4 3 1 30% 10%  9 9 9 5 5 90% 80%  7.08 (1.49) 7.28 (1.54) 7.12 (1.29) 4.69 (.62) 3.95 (1.12) 84.80% (11.30%) 28.40% (17.24%)  4 4 4 3 1 50% 10%  9 9 9 5 5 100% 90%     29 2.3.2  Materials  The materials consisted of standardized measures of oral language proficiency and reading skill, and included equivalent measures in both L2 (English) and L3 (French).  English Language Proficiency Tests  Vocabulary knowledge: The Peabody Picture Vocabulary Test-III (PPVT-III, Dunn & Dunn, 1997) is a standardized test of English vocabulary knowledge and provided a general measure of one aspect of the children’s English oral language proficiency. The children were asked to identify the picture that correctly corresponds to an orally presented word among four picture alternatives. The test-retest reliability coefficient according to the manual for this age group was .93 (Dunn & Dunn, 1997). Listening comprehension: The Listening Comprehension subtest from the Woodcock Language Battery Proficiency (WLPB-R, Woodcock, 1991) was used as a measure of listening comprehension. In this standardized task, the experimenter read aloud short sentences which were each missing a word.  The task required the children to say a word that is appropriate both in terms of structure (syntax) and meaning (semantics) in the context of the sentence. The test-retest reliability coefficient for this age group in the test manual was .86 (Woodcock, 1991).       30 French Language Proficiency Tests Vocabulary knowledge: The Échelle de vocabulaire en images Peabody (ÉVIP, Dunn, Thériault-Whalen, Dunn, 1993) is a standardized test of French vocabulary knowledge and was used as one measure of children’s French language proficiency. As with the PPVT-III, children were asked to identify from among four pictures the image that correctly corresponds to an orally presented word. The test-retest reliability coefficient for this age group in the manual was .90 (Dunn et al., 1993). Listening comprehension: The Compréhension orale subtest from the Test de rendement individual de Wechsler (WIAT-II, Wechsler, 2007) was used as a measure of listening comprehension in French. This standardized test is similar to the English WLPB-R, although it also uses a picture elicitation task and includes items that measure vocabulary knowledge. The test-retest reliability coefficient in the manual was 0.88 for this age group (Wechsler, 2007).  English Reading Tests  The following subtests of the WLPB-R (Woodcock, 1991), specifically designed to test for reading ability in English, were administered to the two groups of participants at T1, T2, and T3: (1) Letter-Word Identification: children were asked to read aloud words that were increasingly more complex and (2) Passage Comprehension: children were asked to read a short passage with a missing word that required them to produce a word that would be appropriate both in terms of structure (syntax) and meaning (semantics) in the context of the paragraph. The WLPB-R was chosen for this study     31 because of its comprehensive nature, its high degree of reliability (test-retest reliability coefficient of .90 to .94) and validity as reported by the publishers (Woodcock, 1991), and its appropriateness for the age groups under investigation.  French Reading Tests  The following subtests of the WIAT-II (Wechsler, 2007), specifically designed to test for reading ability in French, were administered to all participants at T1 and again at T2: (1) Lecture de mots [word reading]: children were asked to read aloud words that were progressively more complex and (2) Compréhension de lecture [reading comprehension]: participants read short stories (aloud or silently) and answered questions relating to the stories. The WIAT-II was chosen because of its high reliability (test-retest coefficient varied between .88 and .94) and validity, as reported by the publishers (Wechsler, 2007), and is widely used in research and schools.  2.3.3  Procedure  Identical procedures were followed for each group at T1, T2 and T3. Participants were tested twice in Grade 4 (beginning and end of Grade 4) and once at the end of Grade 6. Each child was tested individually in two sessions (one in English and one in French) that lasted approximately 45 minutes each and were separated by one to two weeks. Native English- and French- speaking research assistants who were specifically trained for the purposes of the study administered all tests in the two languages. To     32 counterbalance for order effects, one half of the participants were tested first in English, and the other half were tested first in French.  2.4 Results  2.4.1 Comparisons between L2 and L3 oral language proficiency and L2 and L3 reading skills  The first research objective was to determine whether multilingual and bilingual students developed equivalent skills in listening comprehension, vocabulary knowledge, word reading, and reading comprehension in English and French between Grade 4 and Grade 6. Since the study comprised of data using the same tests at three time points, individual growth modeling (Singer & Willett, 2003) was used to examine acquisition of oral language proficiency and reading comprehension over time. A two-level individual mixed-effects model for each of the oral language and reading measures was used, with level 1 representing measurements in time and level 2 representing measurements on the various tasks within individuals. The advantage of testing a two-level individual mixed- effect model is that the model takes into account the nature of the data as repeated measures. Time 1 (T1) represented the beginning of Grade 4, T2 represented the end of Grade 4 (8 months later) and T3 was collected at the end of Grade 6 (20 months later). Therefore, time was represented in terms of months. Results from the level 1 analysis, representing individual students’ scores over time, indicated that there was considerable variability across measures of reading     33 comprehension in both English and French (see Figure 2.1 and Figure 2.2). On the measure of English reading comprehension, 57% of the bilingual students and 51% of the multilingual students showed an improvement in their reading skills from T1 to T3, while 43% of the bilingual students and 49% of the multilingual students showed a decrease in performance over time. On the measure of French reading comprehension, 74% of the bilingual students and 68% of the multilingual students showed an increase in reading skills from T1 to T3 and 26% of the bilingual students and 32% of the multilingual students showed a decrease in performance over time. When comparing the amount of variance, few differences emerged between English and French, with the exception of vocabulary knowledge where a significantly greater amount of variation was observed in French than in English at all three time points in both the bilingual and multilingual students. Furthermore, variation on the French reading comprehension measure increased over time and this increase was observed to the same extent in the multilingual and bilingual students (see Table 2.2).     34 Figure 2.1 English reading comprehension scores at T1 (beginning of Grade 4), T2 (end of Grade 4), and T3 (end of Grade 6) for the bilingual students (Group 1, N = 57) and the multilingual students (Group 2, N = 57). Points from the bold lines represent the mean score for each group on the English reading comprehension test; vertical lines depict standard deviations of the means. The thinner lines represent the scores of each participant at T1, T2, and T3           35  Figure 2.2 French reading comprehension scores at T1 (beginning of Grade 4), T2 (end of Grade 4), and T3 (end of Grade 6) for the bilingual students (Group 1, N = 57) and the multilingual students (Group 2, N = 57). Points from the bold lines represent the mean score for each group on the French reading comprehension test; vertical lines depict standard deviations of the means. The thinner lines represent the scores of each participant at T1, T2, and T3         36 Table 2.2 Test means (standard deviations) on English and French oral language tests and reading tests at T1 (beginning of Grade 4), at T2 (end of Grade 4), and at T3 (end of Grade 6) Tasks Bilingual Multilingual  Listening comprehension    Time 1    Time 2    Time 3 English 115.20 (19.32) 122.34 (16.62) 121.40 (16.43) French 95.45 (11.67) 95.31 (12.98) 94.89 (12.43) English 104.98 (19.43) 111.78 (19.00) 115.52 (17.98) French 94.16 (14.33) 93.48 (16.20) 93.24 (13.32) Vocabulary knowledge    Time 1    Time 2    Time 3  111.53 (11.97) 115.11 (11.48) 118.28 (14.40)  89.26 (18.62) 88.69 (19.59) 94.53 (17.58)  107.32 (14.35) 110.15 (14.42) 114.10 (18.06)  86.14 (23.13) 85.78 (19.27) 95.57 (16.94) Word reading    Time 1    Time 2    Time 3  119.36 (13.07) 124.43 (12.98) 124.74 (13.14)  87.14 (13.29) 89.02 (14.29) 93.19 (13.50)  112.64 (15.00) 121.35 (14.14) 122.38 (16.27)  85.53 (14.90) 87.85 (12.73) 90.02 (12.86) Reading comprehension    Time 1    Time 2    Time 3  110.50 (10.23) 114.06 (10.84) 113.09 (14.49)  94.80 (7.17) 96.52 (9.17) 105.96 (16.34)  109.32 (12.20) 110.80 (10.97) 109.88 (11.52)  93.51 (8.61) 96.24 (11.80) 100.98 (16.34) Note: The standard score on each test is 100 and the standard deviation is 15.     37 First correlations among the variables representing vocabulary knowledge, listening comprehension, word reading, and reading comprehension in English and French are presented for bilingual students and multilingual students separately (see Table 2.3). Next, findings from the level 2 analyses represent group comparisons on the measures of oral language proficiency and reading in English and French over time as well as the interaction between group and time (see Table 2.4).     38 Table 2.3 Correlations between L2 and L3 reading tasks and L2 and L3 proficiency measures at the beginning of Grade 4 (T1) and the end of Grade 6 (T3) for the bilingual (N = 57) and the multilingual (N = 57) groups   1 2 3 4 5 6 7 Bilingual Group 1. T1 English Listening Comp - 2. T1 English PPVT-III .67** - 3. T1 French Listening Comp .55** .66** - 4. T1 French ÉVIP .66** .71** .60** - 5. T3 English Word Reading  .48** .58** .51** .39* - 6. T3 English Reading Comp .52** .55** .69** .48* .63** - 7. T3 French Word Reading .24 .29 .28 .14 .67** .42** - 8. T3 French Reading Comp .67* .65** .71** .68** .57* .74** .49**  Multilingual Group  1. T1 English Listening Comp  - 2. T1 English PPVT-III .59** - 3. T1 French Listening Comp .51** .58** - 4. T1 French ÉVIP .55** .69** .54** - 5. T3 English Word Reading .54* .54** .58** .52* - 6. T3 English Reading Comp 7. T3 French Word Reading 8. T3 French Reading Comp .47** .22 .55** .61** .30* .63** .58** .32* .59** .68** .32* .70** .57** .55** .51** - .29* .65**   .45**     39 Table 2.4 Effects of group, time, and group x time interaction on measures of oral language proficiency and reading ability in English and French Dependent measure and predictor Coefficient SE T English vocabulary    Group    Time    Group x time  -4.80 .25 .04  2.25 .07 .01  2.13* 3.47** .35 English listening comprehension    Group    Time    Group x time  -11.42 .20 .22  3.31 .11 .16  3.46** 1.85 1.43 English word reading    Group    Time    Group x time  -6.49 .19 .23  2.45 .07 .10  2.65** 2.82** 2.45** English reading comprehension    Group    Time    Group x time  -1.79 .05 -.04  1.93 .06 .09  .93 .84 .45 French vocabulary    Group    Time    Group x time  -3.90 .15 .23  3.38 .09 .14  1.15 1.56 1.69 French listening comprehension    Group    Time    Group x time  -1.72 -0.11 .05  2.38 .06 .08  .72 2.01* .64 French word reading    Group    Time    Group x time  -1.87 .19 .01  2.40 .06 .09  .78 3.14** .14 French reading comprehension    Group    Time    Group x time  -.44 .38 -.12  1.36 .07 .10  .32 5.76** 1.30 Note: *p < .05, **p < .01, time is a measurement that was recorded in months.       40 Results from the level 2 analyses showed that language group and time were significant factors for English vocabulary size, (p < .05), such that bilingual students had stronger vocabulary skills than multilingual students and students from both language groups demonstrated an increase from T1 to T3; however, the interaction effect was not significant (p > .05). On the English listening comprehension task, again, there was a significant group effect (p < .05), showing that bilingual students had higher listening comprehension skills than the multilingual group. The main effect for time was not significant (p = .06) and the interaction was not statistically significant (p > .05). Analyses conducted on the two measures of language proficiency in French showed that there was no group difference and time was not a significant factor (p > .05). The interactions were also not statistically significant (p > .05).  On the English word reading task, there was a significant group effect (p < .05), significant time effect (p < .05), and significant interaction (p = .02). While bilingual students had stronger English word reading skills at the beginning of Grade 4, the multilingual students significantly increased their word reading skills from Grade 4 to Grade 6 and caught up to the bilingual group by T3. On English reading comprehension, the group and time effects were not significant and the interaction was also not significant (p > .05).  On the two French reading tasks, group effect was not a significant predictor for either the word reading or reading comprehension task. On the other hand, time was a significant predictor for the two measures of reading (p < .05), indicating an increase in reading performance from T1 to T3 for the two language groups. The interaction was not statistically significant for either reading task (p > .05).     41 2.4.2  The effect of word reading and oral proficiency in the L2 and L3 on reading comprehension in the L2 and L3  The first set of hierarchical regression analyses investigated whether measures of English word reading and English oral proficiency at the beginning of Grade 4 that predicted French reading comprehension at the end of Grade 6 differed for multilingual and bilingual students. Analyses also examined whether measures of French word reading and French oral language proficiency in Grade 4 that explained English reading comprehension in Grade 6 differed for the two groups. The regression model included each of the group x factors interactions (vocabulary knowledge, listening comprehension, word reading). There was only one significant and positive interaction: French vocabulary was a stronger predictor of English reading comprehension for the multilingual students (p < .001), even though French vocabulary was also significantly associated with English reading comprehension for the bilingual group (r = .65, p < .05).  The significant interaction term (French vocabulary x group) was included as a predictor of English reading comprehension in subsequent hierarchical regression analyses. In order to test the hypothesis that French oral language proficiency in Grade 4 would have an effect on the acquisition English reading comprehension in Grade 6 above and beyond that of English language proficiency and literacy skills, the control variables (SES, language group, English word reading, English vocabulary knowledge, and English listening comprehension) were entered first into the regression analyses, then the main predictors (French word reading, French vocabulary knowledge, French listening     42 comprehension, and French vocabulary x group interaction) were entered into the model. Therefore, the predictors were entered in the following order: in step 1, language group and SES; in step 2, English word reading, English vocabulary knowledge, and English listening comprehension; in step 3, French word reading, French vocabulary knowledge, French listening comprehension, and French vocabulary x group interaction. Socio- economic status was included as a control variable given the strong association between SES and reading comprehension (r value varied between .68 and .73, p < .001). The results are presented in Table 2.5.     43 Table 2.5 Multiple regression models representing cross-language effects on reading comprehension in English  B SE t R 2  R 2  change Model 1    Group    SES   1.87 3.56   2.71 .93   .69 3.85**     .17**  Model 2    Group    SES    English word reading    English vocabulary    English listening comprehension  1.26 2.30 .24 .27 .08  2.30 .80 .09 .11 .07   .55 2.91** 2.71** 2.44** 1.11       .47**       .30** Model 3    Group    SES    English word reading    English vocabulary    English listening comprehension    French word reading    French vocabulary    French listening comprehension    French vocabulary x group interaction  .54 2.08 .06 .02 .11 .09 .23 .29 .31  2.19 .72 .08 .13 .07 .08 .09 .12 .09  .25 .2.88** .18 .08 1.62 .26 2.59* 2.72* 3.32**           .61**           .14* Note: *p < .05, **p < .01     44 The model that included language group and SES was significant and explained 17% of the variance. Only SES was a significant factor (p < .001); however, group effect was not a significant factor (p = .49), indicating that English reading comprehension did not differ in the bilingual and multilingual groups. Next, when English word reading, English listening comprehension, and English vocabulary knowledge were added, these factors contributed an additional 30% of variance and together with SES explained 47% of the variance, and the overall regression was significant (p < .05). However, English listening comprehension was not a significant predictor. An F-test was carried out to compare the amount of variance explained in model 1 and model 2. This procedure verified whether adding the English word reading and the two oral language proficiency variables in English increased the overall regression effect. Results showed that model 2 was a better-fitting model, F(2, 76) = 13.89, p < .001. In model 3, French word reading, French listening comprehension, French vocabulary knowledge, and the interaction French vocabulary x group were included and added an additional 14% of variance and the model explained 61% of the variance. All factors that measured oral language proficiency in French were significant, whereas French word reading was not a significant factor. English word reading, English vocabulary knowledge, and English listening comprehension were no longer significant predictors. This finding does not discount the association between English oral proficiency and English reading comprehension (r values between .52 to .57, p < .05), but rather that the French oral proficiency is simply a stronger predictor of reading ability (r values between .70 to .74, p < .001). Model 2 and model 3 were compared and the F-test confirmed that adding     45 French word reading and the two oral language proficiency measures in French increased the regression effect; thus, model 3 was the best-fitting model, F(2, 73) = 6.49, p < .001. The same procedure, as described above, was used to investigate and isolate the effect of English word reading and English oral language proficiency on French reading comprehension, over and above SES, French word reading, and the two measures of French oral proficiency, and is presented in Table 2.6.     46 Table 2.6 Multiple regression models representing cross-language effects on reading comprehension in French.  B SE t R 2  R 2  change Model 1    Group    SES   -2.44 3.38   3.59 1.23   .68 2.76**     .10**  Model 2    Group    SES    French word reading    French vocabulary    French listening comprehension  -1.08 .87 .26 .36 .28  2.50 .90 .09 .07 .14   .43 .97 2.82** 4.79** 2.02*       .58**       .48** Model 3    Group    SES    French word reading    French vocabulary    French listening comprehension    English word reading    English vocabulary    English listening comprehension  .48 .69 .21 .26 .24 .07 .07 .14  2.56 .89 .09 .09 .15 .10 .16 .08  .19 .76 1.92 2.88** 2.09* .62 .48 1.74           .61**          .03 Note: *p < .05, **p < .01     47 The regression model that included language group and SES was significant (p < .001), and accounted for 10% of the variance. Language group was not a significant predictor (p > .005), indicating that French reading comprehension did not differ in the bilingual and multilingual group. Next, when French word reading, French listening comprehension and French vocabulary knowledge were added to the model, it added 48% of variance and explained 58% of the total variance, and the overall regression effect was significant (p < .001). Socioeconomic status was no longer a significant predictor (p > .05). Then the proportion of variance explained in model 1 and model 2 was compared to verify whether adding the two measures of oral language proficiency in French increased the overall regression effect. Findings indicated that model 2 was a better-fitting model, F(2, 76) = 29.47, p < .001. In model 3, English word reading, English listening comprehension, and English vocabulary knowledge were included and together they added 3% of variance and explained 61% of the total variance; however, English word reading and the two English measures of oral language proficiency were not significant factors (p > .05). When model 2 and model 3 were compared, findings showed that the effects in model 3 were not significant, F(2, 74) = 1.81, p = .15; thus, model 2 was the best-fitting model.  2.5 Discussion  This longitudinal study investigated the oral language proficiency and reading skills of bilingual and multilingual students in English and French. The goals of this study were to examine whether multilingual students developed the same level of oral language     48 proficiency and literacy skills in English and French as that of bilingual students. Furthermore the effects of word reading and oral language proficiency on reading comprehension were examined across languages. Of particular theoretical importance was to understand whether the same oral language proficiency skills that are important to English and French reading comprehension in bilingual children are equally important for multilingual students in FI programs.  Of significance is the finding that multilingual students develop equivalent oral language proficiency, including listening comprehension and vocabulary knowledge, in French, and comparable reading skills in English and French to that of bilingual students in FI programs. The results add to the growing body of research suggesting that acquisition of an L3 in environments that do not support the child’s L1 results in no obvious advantages or disadvantages for multilingual learners (Bérubé & Marinova- Todd, 2012; Kleime, 2006; Sanders & Meijers, 1995). According to the linguistic interdependence hypothesis (Cummins, 2000), when the L1 is not adequately supported in the school context, it remains underdeveloped and can be a limiting factor in the development of subsequent languages. The current findings did not entirely support the linguistic interdependence view, and instead the study showed that multilingual students developed equally strong reading skills in both the L2 and L3, despite not receiving direct instruction in their home language. One possible explanation for the disparity between Cummins’ linguistic interdependence hypothesis and the results of the present investigation may be that the multilingual students in the current study were exposed to a significant amount of literacy in the L2 at home, from an early age, and they received a substantial amount of formal reading instruction in the L2 and L3 at school on a daily     49 basis. The majority of multilingual students read in English and French every day and thus benefited from continued literacy exposure in the L2 and the L3. It is possible that multilingual students were able to transfer their language skills from their L2 to their L3, and vice-versa, due to the fact both languages are similar on both structural and orthographic levels (both languages are alphabetic). An additional explanation may be the level of parental education of students enrolled in French immersion programs. While there was a range in terms of parental education, it is possible that in this Canadian study, because of immigration policies that favor an educated workforce, parents were overall more educated than in similar studies conducted in the Netherlands (e.g. Schoonen et al., 2002) or studies with Spanish-speakers in the United States (e.g. Carlo et al., 2004). Therefore, in the current study, it is likely that both bilingual and multilingual students developed their reading skills in English and French to a similar high level as a consequence that parental education was equivalently high in the two language groups. It seems that, as Bild & Swain (1989) showed, being immersed in French reading instruction leads to strong reading skills in the L3 for multilingual students. Multilingual students demonstrated lower oral language proficiency in English than bilingual students. However, multilingual students developed equally strong reading comprehension skills in English. The differences in English oral language proficiency between multilingual and bilingual students must be interpreted carefully, however, since multilingual students’ scores on the two measures of English oral language proficiency remained at or above the mean on test norms between Grades 4 through 6. While multilingual students clearly demonstrated lower English oral proficiency than the bilingual students, it is possible that the multilingual students obtained sufficiently high     50 levels (i.e. reached a threshold) of oral language proficiency both in English and in French, which allowed them to exhibit similar reading comprehension skills in English to that of bilingual students. According to the threshold hypothesis, when students attain a sufficient level of language proficiency (a threshold) in their L1 and L2, they can display strong reading comprehension skills in the L2 (Cummins, 2000).  However, to date the language threshold has not been clearly defined (Bernhardt & Kamil, 1995; Genesee et al., 2008). In other words, it is not clear what the exact levels of language proficiency in the L1 and L2 is required for the acquisition of strong reading comprehension skills in the L2 to emerge, and thus we can only speculate about its effect on reading. In this study, most multilingual students had achieved standard scores that were at or above the mean on test norms that measured vocabulary knowledge and listening comprehension in English and in French and they also demonstrated strong reading comprehension skills in English and French. Although the threshold was not directly tested,  the results from the current study appear to be supported by Cummins’ threshold hypothesis (1976), according to which a threshold of language competence attained in two languages can allow students to exhibit high levels of cognitive ability (e.g. reading comprehension). The current study has provided evidence that French listening comprehension and French vocabulary knowledge in Grade 4 are unique predictors of English reading comprehension in Grade 6 in FI students in Canada. English oral language proficiency in Grade 4, by contrast, did not play a significant role in French reading comprehension skills in Grade 6. Transfer in the current study was observed in one direction only, from French to English. These results extend existing knowledge on transfer between oral language proficiency in the L3 and reading comprehension skills in the L2 reported by     51 Bérubé & Marinova-Todd (2012) for Grade 4 students to older students in Grade 6 from diverse linguistic backgrounds. A possible explanation for the unique contribution of French oral language proficiency to English reading comprehension is that the more intense and formal French academic instruction in the first four years of French immersion may have helped students to acquire French academic vocabulary and listening skills consisting of words that are more common in academic discourse. Those are also words (e.g. ‘élipsis’ and ‘convexe’) that help with comprehension of academic texts in both English and French in later elementary grades. Miller et al. (2006) also found that the cross-language measures of English oral language proficiency (vocabulary diversity and narrative structure) explained a larger proportion of variance in Spanish reading comprehension (6%) as compared to measures of Spanish oral language proficiency which accounted for 2% of English reading comprehension. Students in the Miller et al. (2006) study were bilingual Spanish-English children between Kindergarten and Grade 3 and were receiving mostly English instruction at school. Similar to the Miller et al. (2006) study, multilingual students in the current study were receiving the majority of their instruction in French. These results attest to the importance of exposure (immersion) to academic language in the classroom in determining the direction of transfer; therefore, oral skills in the L3 were more indicative of later ability to read in the L2. The findings from the current study are in line with the simple view of reading model, revealing that oral language proficiency  (i.e. vocabulary knowledge and listening comprehension) is important in the development of reading comprehension skills. These results are similar to those found in previous studies with bilingual Spanish-English     52 students (Proctor et al., 2006) and bilingual Dutch-English students learning to read in an L2 (Verhoeven & Leeuwen, 2012), which also found that the listening comprehension skills and vocabulary knowledge in the bilinguals students’ L1 were important predictors of the reading comprehension skills in the L2. Word reading, by contrast, did not predict reading comprehension. Several researchers have suggested that when decoding skills are well established, oral language proficiency, especially vocabulary knowledge, takes on a primary role in explaining reading comprehension (e.g. Savage, 2001; Storch & Whitehurst, 2002).  Since the students in the current study had developed word reading skills expected for their age it is possible that their language proficiency has become increasingly more important in reading comprehension. The significant interaction between group and French vocabulary knowledge demonstrated that the association between French vocabulary knowledge and English reading comprehension was stronger in multilingual students than in bilingual students. Hufeisen and Marx (2007) identified a number of factors that explain differences in the acquisition of an L2 and an L3. Where the L2 student is a novice learner in the process of L2 acquisition the L3 student has already learned about the process of acquiring a second language. In comparison with bilingual students learning French as an L2, multilingual students in the current study had already learned about the process of developing English as a second language, and may have benefited from language-specific abilities in acquiring French as an L3. Similarly, Bérubé & Marinova-Todd (2012) reported that acquisition of oral language proficiency in the L3 of Grade 4 students, particularly vocabulary knowledge, influenced reading comprehension in the L2 to a greater extent when the multilingual students’ L1 was typologically similar to the L2 and L3. These     53 results indicate that multilingual students have language-specific knowledge and abilities at their disposal, which L2 students typically do not. Although multilingual students obtained lower English word reading skills in Grade 4, they ultimately caught up to the bilingual group by the end of Grade 6. Much like the transfer between French vocabulary and English reading comprehension observed in the multilingual students, it is possible that French vocabulary knowledge also contributed to an increase in English word reading ability as a result of more intense exposure to French vocabulary in FI programs. Metsala (1999) proposed that as more word forms are added to the lexicon, students become increasingly sensitive to sublexical components (e.g. phoneme awareness and morphological awareness), and thus become more skilled readers. While this explanation was not directly tested here, a number of studies conducted with monolingual students have indicated that vocabulary knowledge is an important factor in determining word reading skills (see Snow et al., 1998, for a comprehensive review). A similar relationship may exist with multilingual students. The findings from the current study suggest a cross-language association between French vocabulary and English word reading. English and French share many cognates and it is possible that multilingual students transferred their vocabulary knowledge from French to English, which, in turn, helped increase their English word reading. Cognates are vocabulary words in two languages that are similar both orthographically and semantically (Holmes & Guerra Ramos, 1995).  Future studies should explore the unique contribution of vocabulary on word reading in the L2 and L3 of multilingual students. Overall, the success of bilingual students in FI programs has been well- documented (Genesee, 1983; Turnbull et al., 2003) and the current findings demonstrate     54 that multilingual students also develop literacy skills that are equally strong as those of bilingual students in Canada’s two official languages. However, the results need to be interpreted with caution. One issue is whether the findings would apply to multilingual students in other parts of Canada. The current study involved students from culturally and linguistically diverse communities in Vancouver, in which English is the dominant language outside the school. In other urban areas, such as Montreal, students are continually immersed in French, both in school and in the community. It would be important to examine whether the development of French and English of multilingual students differs in a variety of language settings, such as multilingual students who are enrolled in English immersion programs in Québec, learning English as an L3. Bilingual French-English students in Québec have the option to enroll in intensive English programs beginning in Grade 5. These programs are similar to the late FI programs across Canada and offer instruction in English. Germain, Lightbown, Netten, and Spada (2004) reported that by Grade 9, after receiving five years of English instruction in immersion programs, bilingual French-English students in Québec demonstrated equivalent oral language proficiency and similar reading comprehension skills to English monolingual students. These findings suggest that multilingual students in Québec enrolled in intensive English programs may also demonstrate strong English language skills (their L3) by Grade 9. Furthermore, measures of listening comprehension and vocabulary knowledge were used as the only measures of oral language proficiency in English and French. Future research will need to investigate other predictors of reading comprehension, such as morphological awareness and grammatical knowledge, both of which have been found     55 to be associated to reading ability within and across languages (Deacon, Wade-Woolley, & Kirby, 2007; Manis, Lindsey, & Bailey 2004). The current study involved regression analyses on a somewhat limited sample size. Future studies should aim to include a larger sample of students, which would allow for enough power to utilize alternative statistical approaches, such as factor analyses or multivariate analyses that would allow researchers to investigate simultaneously a greater number of factors that influence the acquisition of reading skills of multilingual students (Gunderson, 2007). As more and more immigrant students learn English and French in Canada, it will be important to investigate their reading success. The current study has identified the oral language variables associated with reading achievement in Canada’s two official languages in the attempt to inform best educational practices, which will help all children attain high levels of literacy in English and French. Two instructional implications emerged from this study: 1) French vocabulary and French listening comprehension skills were strongly associated with English and French reading comprehension. These findings indicate that FI programs are working: the biliteracy skills of bilingual and multilingual students are strong, and teachers should continue to support the French language skills of all students which will help foster later reading ability, and 2) multilingual students demonstrated lower oral language proficiency in English, the L2, but equivalent oral language proficiency in French, and equivalent literacy skills in both English and French. Why multilingual students have lower oral skills in English needs to be further examined. Once multilingual students are directly taught in French and English, fostering both French and English vocabulary and listening skills would appear paramount.     56 Chapter 3: First language typology and the development of oral language proficiency and reading skills in the L2 and L3 of multilingual students  3.1  Synopsis  The relationship between first language (L1) typology, defined as the classification of languages according to their structural characteristics (e.g. phonological systems and writing systems), and the development of second (L2) and third (L3) language skills and literacy proficiency in multilingual students was investigated in this study. The sample included 90 students in Grade 4, tested once at the beginning of Grade 4 (T1) and again at the end of Grade 4 (T2). The students belonged to one of three language groups: bilingual students, multilingual students who were literate in an alphabetic L1, and multilingual students who were literate in a logographic/syllabary L1. In addition, a separate sample of 82 students belonging to the same three language groups was tested at the end of Grade 6. The study examined the extent to which the development of L2 and L3 literacy skills varied primarily as a function of orthographic similarities with the L1. Results revealed that students who were literate in an alphabetic L1 showed advantages in L2 and L3 reading comprehension. However, there were no differences on tasks that measured word reading and pseudoword reading. A more accurate picture of what facilitates L2 and L3 reading development is enhanced when differences in L2 and L3 proficiency were also considered.     57 3.2  Introduction  Few studies have explored the relationship between first (L1) language typology and the acquisition of a second (L2) and third (L3) language (Bild & Swain, 1989; Cenoz & Valencia, 1994; Gunderson, 2007; Haenni Hoti, Heinzmann, Muller, Oliveira, Wicki, & Werlen, 2011; Munoz, 2000; Sanz, 2000; Swain, Lapkin, Rowen, & Hart, 1990). From a linguistic perspective (Foote, 2007), language typology is defined as the classification of languages according to their structural characteristics, such as phonological systems, writing systems, and word order (Proctor, August, Snow, & Barr, 2010). Of those studies that examined the effects of L1 typology on L3 development, the majority included bilinguals whose L1 had official status in the community, was the language of instruction, and was typologically similar to the target L3 (Cenoz & Valencia, 1994; Munoz, 2000; Sanz, 2000). By contrast, relatively few studies have examined whether differences in L1 typology influenced the development of oral language proficiency and literacy skills in the L2 and L3 (Gunderson, 2007; Swain et al., 1990). Moreover, no studies have examined the relationship between L2 and L3 oral skills and L2 and L3 literacy acquisition. The study of L1 typological differences on L2 and L3 oral skills and reading development will increase our understanding of the effects of different L1s when acquiring a second and third language, especially in environments where the L1 is not directly taught in the classroom and when the L1 is typologically dissimilar to the target L2 and L3 (Gunderson, 2007).  The question of language typology has been addressed by the script-dependent hypothesis. This hypothesis suggests that differences in L2 reading development are     58 related to how closely writing systems from different languages share a common orthography (Dressler & Kamil, 2006; Geva & Siegel, 2000; Kats & Frost, 1992). Therefore, L2 reading in languages with common orthographies (e.g. languages that share an alphabetic writing system) is assumed to develop more quickly than in languages with dissimilar writing systems (e.g. alphabetic and logographic writing systems such as in Cantonese and Mandarin).  Within the field of L2 acquisition, research has examined the extent to which L1 typology influences L2 reading acquisition (Gunderson, 2007; Jarvis & Pavlenko, 2008 for a review of cross-language influence). Gunderson (2007) investigated the English reading outcomes of immigrant students between the ages of 8 and 17 years learning English as a second language and enrolled in English programs in Canada. Immigrant students were categorized according to L1 orthography (e.g. alphabetic Roman, alphabetic Greek, logographic, and mixed languages such as Japanese). Findings provided evidence that English reading comprehension was not influenced by L1 orthography. Instead, the results suggested that immigrant students shared a common and interdependent proficiency in the L1 and L2, whereby the number of years of exposure to reading in the L1 was associated with reading skills in the L2. Other studies in the field of L2 acquisition have shown that reading comprehension in the L2 is related to L1 typology, whereby the L1 influenced the L2 to a greater extent in languages which share a common orthographic system (see Jarvis & Pavlenko, 2008 for a survey of cross-language influence). The development of L1 and L2 literacy in terms of word reading, pseudoword reading, and reading comprehension was compared in alphabetic languages (Portuguese, Spanish) with that of orthographically     59 dissimilar languages (Cantonese). Generally in cases when the two languages were orthographically similar (Durgunoglu, Nagy, & Hacin Bhatt, 1993; Royer & Carlo, 1991) L1 reading ability significantly predicted L2 reading ability. By contrast, studies that examined reading skills between typologically-different languages have reported weak or no correlations across languages (Biaystok, Luk, & Kwan, 2005; Gottardo, Yan, Siegel, & Wade-Woolley, 2001; Pasquerella, Chen, Lam, Luo & Ramirez, 2011). The difference in findings between the Gunderson (2007) study and the research conducted by Bialystok et al. (2005) and Gottardo et al. (2001) could be explained with the age at which the students were first exposed to English as an L2. The majority of students in the Gunderson (2007) study began receiving English literacy instruction after Grade 3, after being instructed in the L1; whereas students in the Bialystok et al. (2005) and Gottardo et al. (2001) studies had been exposed to literacy in the L1 and L2 simultaneously and from a young age. The age of exposure and the sequence of exposure to literacy in the L1 and L2 may be important factors in developing reading comprehension in the L2.   The role of language typology has also been examined in understanding how L1 oral skills, such as vocabulary, influence the acquisition of L2 oral skills and L2 reading proficiency. The research include language pairs from Indo-European languages that share a common root as well as other languages that do not share a common language root. Studies of bilingual students with orthographically similar languages, such as Spanish, German, and Portuguese, have demonstrated limited evidence of positive transfer in vocabulary knowledge (Durgunoglu et al., 1993; Garcia, 1991; Ordonez, Carlo, Snow, & McLaughlin, 2002). Moreover, L1 and L2 vocabulary knowledge has been positively associated with L2 reading comprehension when the two languages are     60 alphabetic, e.g., Spanish (L1) and English (L2) (August, Carlo, Dressler, & Snow, 2005; Dressler & Kamil, 2006; Jimenez, Garcia, & Pearson, 1996; Miller, Heilmann, & Nockerts, 2006; Proctor, August, Carlo, & Snow, 2006). Only one study to date has examined this relationship in languages that are typologically distant. Koda (1989) examined the acquisition of L2 Japanese in two groups of young adults (mean age of 21 years) at the beginning and end of university term: English and Spanish monolinguals (alphabetic L1) and Chinese monolinguals (non-alphabetic L1 that uses logographic characters). The Japanese writing system uses both Kanji (logographic characters similar to Chinese) and Kana (syllabary sound based writing system). The author hypothesized that a stronger and positive relationship between L2 vocabulary and L2 reading comprehension would be observed in the Chinese L1 group because of the common orthographies in the writing systems. The results confirmed the hypothesis; stronger L2 vocabulary was observed in the Chinese L1 group and, in turn, L2 vocabulary was positively associated with L2 reading comprehension in the Chinese L1 group only.  What is less known in multilingual development, is how L1 typology relates to the development of L2 and L3 oral language proficiency and reading development. Bild and Swain (1989), Cenoz, (2003) and Swain et al. (1990) have shown that when the L1 was typologically related to the L3, multilingual students acquired high levels of vocabulary and writing skills in the L3.  Yet, only one study (Swain et al., 1990) focused on the L3 reading development of individuals from orthographically dissimilar L1s. The authors found that L1 typology influenced the development of L3 (French) writing, listening and reading comprehension skills of two different groups of school-age multilingual students; in one group the students had alphabetic L1s, and in the other group, the students had     61 non-alphabetic L1s. All students had received five years of English instruction (Kindergarten to Grade 4) and three years of bilingual English-French instruction (Grades 5 through 7). Therefore, all students shared a common alphabetic L2, English. The students whose L1s were Romance languages (i.e., also alphabetic) scored higher on a written vocabulary task, a global understanding index of listening comprehension, and on a fluency measure of oral French. There was a trend for students with a Romance L1 to perform better on the reading comprehension measure than the non-Romance group, but the difference was not statistically significant. Taken together the results from this study suggested that in these groups of multilingual students L1 typology may have influenced the acquisition of certain L3 skills; by contrast, language typology did not appear to affect reading comprehension. However, in the Swain et al. (1990) study, within the non-Romance group, many languages (e.g. German, Tagalog, Greek, Polish) were alphabetic and orthographically similar to those in the L1 Romance group and related more or less closely to French. Furthermore, it was not clear whether the multilingual students had developed both oral and literacy skills in their L1 or whether the students had acquired only oral skills. Language typology would be expected to influence reading development to a greater extent in students who were literate in their L1 because these students had acquired knowledge of the orthographic system (Cenoz, 2003; Errasti, 2003; Munoz, 2000).  Research on cross-linguistic influence in the acquisition of L3 oral language proficiency has consistently shown that students who have an alphabetic and orthographically similar L1 to the target L3 demonstrated increased positive influence in phonology and grammatical ability (Bild & Swain, 1989; Cenoz, 2003; De Angelis &     62 Selinker, 2003; Foote, 2007; Hammarberg, 2003; Leung, 2005; Ringbom, 2003; Swain et al., 1990). Fewer studies, by contrast, have demonstrated evidence of positive influence in L1 and L3 oral skills when the languages were orthographically dissimilar (Leung, 2005; Swain et al., 1990). Leung (2005) compared the acquisition of French (L3) determiner phrases between two groups of multilingual Chinese L1 and Vietnamese L1 young adults, whose L1 typology differed from that of the target language. Results showed that L3 French students with orthographically dissimilar L1 (Chinese L1 – English L2) demonstrated a significantly greater number of grammatical errors, whereas, the L3 students with common alphabetic L1 (Vietnamese L1 – English L2) performed better on French determiner phrases only.  Of the few studies that have examined cross-linguistic influence in L3 development, most have focused on the identification of three common factors that can explain the acquisition of oral language proficiency and literacy: a) typological distance, b) proficiency in the different languages, such that students who demonstrate greater L1 and L2 reading proficiency acquired greater L3 reading skills (Cenoz & Valencia, 1994), and c) factors associated with L1 use (Cenoz, 2003; Cenoz & Valencia, 1994; Sanz, 2000, 2008). The focal point of the current study is typology; nonetheless, factors associated with language use (e.g. age of arrival, literacy practice) and SES are important in the acquisition of language and literacy (D’Angiulli, Siegel, & Maggi, 2004; Gunderson, 2007; Leseaux & Siegel, 2003) and, thus, will also be considered. The present study set out to examine the effects of L1 typology on the development of L2 and L3 oral proficiency and L2 and L3 reading skills in two groups of multilingual students in Grade 4 who were exposed to English and French as additional     63 languages in an early French immersion program. The study also explored the effects of L1 typology on the acquisition of L2 and L3 oral proficiency and L2 and L3 reading skills in a separate sample of multilingual students two years later in Grade 6. In particular, the study investigated the hypothesis that exposure to L1 literacy in orthographically similar alphabetic languages, such as Spanish and German, would increase reading skills in additional languages such as English and French. The following three research questions were addressed: 1) How does the performance of two groups of multilingual students, one group with an alphabetic L1 and another with a non-alphabetic L1 compare with that of a group of bilingual students learning French as a second language (FSL) on measures of L2 (English) and L3 (French) language proficiency? In addition, how does the performance between the two multilingual groups compare on measures of L2 (English) and L3 (French) language proficiency? 2) How does the performance of two groups of multilingual students compare with that of a group of bilingual students learning FSL on measures of L2 English and L3 French reading skills? In addition, how does the performance between the two multilingual groups compare on measures of L2 English and L3 French reading skills? 3) Do measures of oral language proficiency in L2 and L3 predict L2 and L3 reading scores differently in students from typologically similar and different L1?     64 3.3  Method  3.3.1  Participants  The participants included 90 students in Grade 4 who belonged to one of three groups: 1) 57 bilingual students studying French as a second language, 2) 20 multilingual students who were literate in an alphabetic L1 (representing 15 different languages), and 3) 13 multilingual students who were literate in a logographic/syllabary L1 (representing 4 different languages). The age of the students in the three language groups at T1 and T2 was equivalent (see Table 3.1 for participant demographics). All students were enrolled in an early French immersion program and had received French instruction from Kindergarten to Grade 4 and began receiving formal English instruction in Grade 4. Instruction in early French immersion programs consisted of 100% French from Kindergarten to Grade 3, and then 80% French and 20% English from Grades 4 to 7. The age of the students in the three language groups at T1 and T2 was equivalent (see Table 3.1 for participant demographics). All students were enrolled in an early French immersion program and had received French instruction from Kindergarten to Grade 4 and began receiving formal English instruction in Grade 4 through Grade 6. Instruction in early French immersion programs consisted of 100% French from Kindergarten to Grade 3, and then 80% French and 20% English from Grades 4 to 6. Participants were recruited from schools in culturally and linguistically diverse communities in a major metropolitan city in Canada. Background information on SES variables, amount of use in each language in the home and outside the home, as well as     65 the literacy activities in all languages used at home was collected, to establish that the multilingual students in both alphabetic and non-alphabetic groups and the English bilingual students had similar exposure to English and French. There were no significant differences across the three groups in the level of education attained by parents, F(2, 89) = .09, p = .91 who on average had achieved at least a bachelor’s degree across all three groups. The level of education attained by parents varied between having graduated from high school to having completed a graduate degree.     66 Table 3.1  Demographic information on the two multilingual L1 groups and the bilingual group at the beginning of Grade 4 (T1), the end of Grade 4 (T2), and at the end of Grade 6 (T3)  Note: The number ‘7’ in degree of maternal education indicates that mothers had achieved at least a bachelor’s degree and ‘6’ indicates that mothers had achieved at least some college education or trade school training. Group Multilingual Alphabetic L1 Multilingual Logographic/Syllabary L1 Bilingual N   Grade 4   Grade 6  20 23  13 12  57 47 L1 Afrikaans, Amharic, Croatian, Czech, Danish, Fanti, German, Greek, Hungarian, Korean, Polish, Serbian, Spanish, Tagalog or Vietnamese Cantonese, Japanese, Mandarin, Shanghainese English Age at T1, T2, Grade 6 9;3 / 9;9 /11;8 9;4 / 9;10 / 11;9 9;4 / 9;9 / 11;8 Number of student attrition from Grade 4 to Grade 6 5 5 21 Number of new students from Grade 4 to Grade 6 8 4 11 Degree of parental education (SD)   Grade 4   Grade 6   7.03 (1.53) 6.98 (1.63)   6.81 (1.71) 6.88 (1.93)   7.86 (1.37) 7.22 (1.37)     67 Data on home language literacy exposure was obtained through a comprehensive written parent questionnaire with three particular questions: a) how often parents read to their child in each language, b) how often their child read independently in each language, and, finally, c) who else read in each language with their child. Only those students whose parents reported that their child spoke the L1 at home on a daily basis and that the child read in the L1 at least twice per month were included and this assured that they had generally strong exposure to the L1 and that they were all literate in their L1. The age at which students were first exposed to English, their L2, was equivalent in both the group of students with an alphabetic L1 (M = 0;1, SD = 0;5) and the group of students with a logographic/syllabary L1 (M = 0;2, SD = 0;7), t(32) = .61, p = .54, d = .01. The average length of residence (in years) in Canada for the group of students with an alphabetic L1 (M = 8.75, SD = 0.75) was also equivalent to the group of students with a logographic/syllabary L1 (M = 8.67, SD = .82), t(32) = .27, p = .79, d = .002. The multilingual students lived in English-dominant communities and they were exposed to English before French in the home; therefore, English was considered as their L2 and French as their L3. Parents reported on their child’s literacy exposure in the L2 and L3 by indicating the number of times their child read in each language (i.e. on a scale, which ranged from 1 = once or twice per month, 3 = two or three times per week, and 5 = every day). Parents of the group of students with an alphabetic L1 read to their children more often in the L2 (M = 3.80; SD = 1.36) than in the L3 (M = 1.75, SD = 1.48), t(19) = 4.74, p = .001, d = .54. Parents’ responses ranged from never reading to their children in English and French to reading to their children in both languages every day. Similar patterns were observed in the students with a logographic/syllabary L1,     68 where parents read more often in the L2 (M = 3.54, SD = 1.76) than in the L3 (M = 1.38, SD = 2.02), t(12) = 4.17, p = .001, d = .59. Again, parents’ responses ranged from never reading to their children in English and French to reading to their children in both languages every day. However, there were no group differences in the amount of exposure to reading in L2 and L3 between the two multilingual groups, F(1, 31) = .45, p = .51, 2 =  .01. Therefore this variable was not controlled for in subsequent analyses.  3.3.2  Participants from the Grade 6 sample  The participants included 82 students in Grade 6 who belonged to one of three language groups: 47 bilingual students, 23 multilingual students who were literate in an alphabetic L1, and 12 multilingual students who were literate in a logographic/syllabary L1. Fifty-nine students were tested in both Grade 4 and Grade 6. At the end of Grade 4, 31 students did not complete any further testing. Ten of the students moved away, nine were in English programs, six did not return consent forms, and six declined to participate. Twenty-three additional students who did not participate in the study in Grade 4 were tested in Grade 6 (see Table 3.1 for the number of student attrition and the number of new participants). The background information observed for the students in Grade 6 was very similar to that for the students in Grade 4. There were no significant differences across the three language groups in the level of education attained by parents, F(2, 73) = .39, p = .68, who on average had achieved at least a bachelor’s degree. Education attained by parents varied from having graduated from high school to having completed a graduate     69 degree. Moreover, the average length of residency (in years) in Canada for the group of students with an alphabetic L1 (M = 10.74, SD = .54) was equivalent to the group of students with a logographic/syllabary L1 (M = 10.42, SD = 1.28), t(33) = .95, p = .33, d = .03. On measures of literacy practice, students with an alphabetic L1 read more often in the L2 (M = 4.77, SD = .53) than in the L3 (M = 4.00, SD = 1.15), t(21) = 3.55, p = .001, d = .39. Moreover, similar patterns were observed in the group of students with a logographic/syllabary L1, where students read more often in the L2 (M = 4.82, SD = .41) than in the L3 (M = 3.64, SD = 1.21), t(10) = 3.63, p = .001, d = .60. Students’ responses to amount of literacy practice ranged from reading in English and French once or twice per week to reading in both languages every day. Chinese-speakers comprised the largest group of multilingual students (N = 10) and in order to confirm that students had generally strong L1 skills, they were tested on measures of oral language proficiency, using the Chinese Peabody Picture Vocabulary Test (standard scores from Chinese PPVT-R) (Lu & Liu, 1997) and the Chinese Reading Comprehension Test (raw scores from the Chinese Reading Comprehension Test) (Pasquerella et al., 2011). Findings indicated a broad range of oral language proficiency in Chinese. Two students had low vocabulary skills (standard score below 60) and the other eight students had a larger vocabulary (standard score between 75 and 98). Only five of the ten students completed the Chinese Reading Comprehension Test and scores varied between 15 and 34 (out of 35).        70 3.3.3  Materials  The materials consisted of language proficiency and reading measures, and included equivalent standardized measures in both L2 (English) and L3 (French). The same measures were used for the Grade 4 and Grade 6 samples. In addition, Chinese vocabulary knowledge and Chinese reading comprehension measures were included in the Grade 6 sample.  English Proficiency Tests  The Peabody Picture Vocabulary Test-III (PPVT-III, Dunn & Dunn, 1997) is a standardized test of English vocabulary knowledge and provided a general measure of one aspect of the student’s English oral language proficiency. The students were asked to identify the picture that correctly corresponds to an orally presented word among four picture alternatives. The test-retest reliability coefficients for this age group as reported by the publisher were .93 (Dunn & Dunn, 1997). The Listening Comprehension subtest from the Woodcock Language Battery Proficiency (WLPB-R, Woodcock, 1991) was used as a measure of listening comprehension in English. In this standardized task, the experimenter read aloud short sentences which were each missing a word. The task required the students to say a word that is appropriate both in terms of structure (syntax) and meaning (semantics) in the context of the sentence. The test-retest reliability coefficient for this age group as reported by the publisher was .86 (Woodcock, 1991).     71 French Proficiency Tests  The Échelle de vocabulaire en images Peabody (ÉVIP, Dunn, Thériault-Whalen, & Dunn, 1993) is a standardized test of French vocabulary knowledge and was used as one measure of student’s French language proficiency. As with the PPVT-III, students were asked to identify from among four pictures the image that correctly corresponds to an orally presented word. The test-retest reliability coefficient for this age group as reported by the publisher was .90 (Dunn et al., 1993). The Compréhension orale subtest from the Test de rendement individual de Wechsler (WIAT-II, Wechsler, 2007) was used as a measure of listening comprehension in French. This standardized test is similar to the English WLPB-R, although it also uses a picture elicitation task and includes items that measure vocabulary knowledge and production. The test-retest reliability coefficient as reported by the publisher was 0.88 for this age group (Wechsler, 2007).  Chinese Proficiency Test  The Peabody Picture Vocabulary Test-Revised (PPVT-R, Lu & Liu, 1997), a standardized test of Chinese vocabulary knowledge, similar to the English PPVT and the French ÉVIP provides a general measure of one aspect of the student’s Chinese oral language proficiency. Students were asked to identify from among four pictures the image that correctly corresponds to an orally presented word. The test-retest reliability coefficient for this age group as reported by the publisher was .86 (Lu & Liu, 1997).     72 English Reading Tests  The following subtests of the WLPB-R (Woodcock, 1991), specifically designed to test for reading ability in English, were administered to the three groups of participants at T1 and again at T2: (1) Letter-Word Identification, students were asked to read aloud words that were increasingly more complex; (2) Word Attack, students were asked to read pseudowords, for example yosh and thrept; and (3) Passage Comprehension, students were asked to read a short passage with a missing word that required them to produce a word that would be appropriate both in terms of structure (syntax) and meaning (semantics) in the context of the paragraph. The WLPB-R was chosen for this study because of its comprehensive nature, its high degree of reliability (test-retest reliability coefficient of .90 to .94) and validity as reported by the publishers, and its appropriateness for the age groups under investigation.  French Reading Tests  The following subtests of the WIAT-II (Wechsler, 2007), specifically designed to test for reading ability in French, were administered to all participants at T1 and again at T2: (1) Lecture de mots [word reading]: students were asked to read aloud words that were progressively more complex; (2) Décodage de pseudomots [pseudoword reading]; and (3) Compréhension de lecture [reading comprehension]: participants read short stories (aloud or silently) and answered questions relating to the stories. The WIAT-II was chosen because of its high reliability (test-retest coefficient varied between .88 and     73 .94) and validity, as reported by the publishers (Wechsler, 2007), and is widely used in research and schools.  Chinese Reading Test  The Chinese reading comprehension test (Pasquerella et al., 2011), an experimenter-designed task, was used as a measure of reading ability. In this task, students had to identify the picture that correctly corresponded to a sentence or short passage that they read among four picture alternatives. There were 40 test items.  3.3.4 Procedure  Identical procedures were followed for each group at T1 and T2. Thus, participants were tested twice in Grade 4 (once at the beginning and once at the end of Grade 4). Each child was tested individually in two sessions (one in English and one in French) that lasted approximately 45 minutes each and were separated by 1 to 2 weeks. To counterbalance for order effects, one half of the participants were tested first in English, and the other half were tested first in French. The same procedure was followed for each group in Grade 6. In addition, Chinese-speaking students were tested in their home language in an additional 15-minute session. Native English-, French-, and Chinese-speakers administered all tests in the three languages.       74 3.4  Results  Grade 4 results will be presented first, followed by the Grade 6 data. The same set of analyses was completed in Grade 4 and in Grade 6.  3.4.1  The effect of L1 typology on oral proficiency in L2 and L3  The F tests and descriptive statistics on the English and French measures of language proficiency are provided in Table 3.2.     75 Table 3.2 Mean standard scores and F test comparisons on measures of listening comprehension and vocabulary knowledge at the beginning of Grade 4 (T1), the end of Grade 4 (T2), and the end of Grade 6 (T3) according to L1 typology on L2 (English) and L3 (French) proficiency tests L2 tests Alphabetic Logographic/ syllabary     Bilingual df F p ᵨ2 Listening Comprehension    Time 1    Time 2    Grade 6 At T1 & T2    Within subjects    Between subjects Grade 6    Between subjects   108.70 (20.52) 113.85 (17.57) 114.43 (16.86)    94.57 (16.08) 104.21 (15.58) 116.00 (15.81)   114. 37 (17.54) 121.47 (15.46) 119.74 (16.00)        1, 89 2, 89  2, 81       9.06** 10.33**  .79       .001 .001  .45       .09 .19  .02  Vocabulary knowledge    Time 1    Time 2    Grade 6 At T1 & T2    Within subjects    Between subjects Grade 6    Between subjects  107.85 (15.88) 114.10 (16.47) 114.04 (15.84)   98.07 (11.57) 102.85 (10.83) 110.17 (15.89)  110.76 (10.48) 112.39 (16.21) 116.85 (14.38)      1, 89 2, 89  2, 81      5.67* 5.01*  1.13      .01 .01  .33      .06 .10  .02      76 Continue Table 3.2 L3 tests Alphabetic Logographic/ Syllabary Bilingual df F p ᵨ2 Listening comprehension    Time 1    Time 2    Grade 6 At T1 & T2    Within subjects    Between subjects Grade 6    Between subjects   98.70 (12.88) 99.95 (12.87) 95.96 (13.43)   86.00 (14.63) 85.26 (11.40) 90.67 (11.31)   93.51 (15.03) 94.43 (11.17) 92.69 (13.15)      1, 81 2, 81  2, 81      .08 5.36*  .79      .79 .007  .46      .002 .18  .02 Vocabulary knowledge    Time 1    Time 2    Grade 6 At T1 & T2    Within subjects    Between subjects Grade 6    Between subjects  83.25 (21.97) 89.90 (16.72) 97.22 (15.33)  69.64 (30.39) 78.27 (16.63) 90.50 (15.38)    86.96 (17.49) 87.64 (18.33) 93.89 (17.42)      1, 81 2, 81  2, 81      4.11* 3.03  .68      .04 .06  .51      .04 .05  .01 Note: *p < .05, **p < .01.     77 The relationship between L1 typology and L2 language proficiency was examined for multilingual students from all L1 backgrounds and for the bilingual students at T1 and T2. A series of two-way mixed ANOVA for group (3) and time (2) were conducted on the standard scores of the listening comprehension task of the WLPB-R and on the standard scores of the English PPVT-III. There was a group effect on the listening comprehension task with post-hoc (Bonferonni) analyses demonstrating that bilingual students and students with an alphabetic L1 had higher scores than the students with a logographic/syllabary L1, with no difference between the bilingual students and students with an alphabetic L1. There was also a significant time effect, indicating that students at the end of Grade 4 had higher scores than at the beginning of Grade 4. There was no significant interaction on the listening comprehension task, F(2, 89) = .22, p = .80, ᵨ2 =.005. On the English PPVT-III measure, there was a significant group effect, with follow-up analyses (Bonferroni post-hoc test) showing that the bilingual students and students with an alphabetic L1 had higher scores than students with a logographic/syllabary L1, and no difference between the bilingual group and the group of students with an alphabetic L1. There was a significant time effect showing that students had greater vocabulary knowledge at the end of Grade 4. The interaction was not significant, F(2, 89) = .90, p = .41,ᵨ2 =.02.  A series of two-way mixed ANOVAs for group (3) and time (2) comparing the performance of oral language proficiency between T1 and T2 on the French oral measures revealed different patterns in the two tasks: on the listening comprehension task, there were significant group effects; however, there was no time effect. A     78 Bonferroni post-hoc test showed that students with an alphabetic L1 and the bilingual students had higher scores on the listening comprehension measure than students with a logographic/syllabary L1, and there was no difference between the bilingual students and the students with an alphabetic L1. On the vocabulary knowledge task, the main group effect nearly reached statistical significance (p = .054). The students with an alphabetic L1 and the bilingual students tended to have greater vocabulary knowledge than students with a logographic/syllabary L1. There was a time effect, where vocabulary was greater at T2 than at T1. Results revealed no significant interactions on any of the tasks that measured French proficiency (p > .05).  3.4.2  The effect of L1 typology on reading in L2 and L3  Table 3.3 shows the F-tests and mean scores on all English and French reading measures for the three groups. For these analyses, the performance of the group of students with a logographic/syllabary L1 was compared to that of the group of students with an alphabetic L1 with that of bilingual students at the beginning and end of Grade 4.     79 Table 3.3 Mean standard scores and F tests on measures of word reading and reading comprehension at the beginning of Grade 4 (T1), the end of Grade 4 (T2), and the end of Grade 6 (T3)  according to L1 typology on L2 (English) and L3 (French) reading tests              L2 Tests Alphabetic Logographic/ Syllabary      Bilingual df F p ᵨ2 Word Reading    Time 1    Time 2    Grade 6 At T1 & T2    Within subjects    Between subjects Grade 6    Between subjects  115.85 (15.29) 127.10 (15.59) 121.74 (16.41)  109.71 (14.21) 114.21 (11.30) 126.58 (13.83)  118.16 (13.03) 121.64 (19.17) 125.15 (12.53)       1, 89 2, 89  2, 81      11.84** 2.26  .68      .001 .11  .51      .12 .05  .01  Pseudoword    Time 1    Time 2    Grade 6 At T1 & T2    Within subjects    Between subjects Grade 6    Between subjects  108.35 (14.36) 119.35 (13.33) 117.74 (15.46)  104.85 (16.97) 111.07 (14.80) 125.67 (16.11)  107.98 (16.04) 115.85 (17.03) 119.53 (15.11)      1, 89 2, 89  2, 81      27.40** .69  1.10      .001 .51  .34      .24 .02  .02      80 Continued Table 3.3 L2 Tests Alphabetic Logographic/ Syllabary      Bilingual df F p ᵨ2 Reading Comp    Time 1    Time 3    Grade 6 At T1 & T2    Within subjects    Between subjects Grade 6    Between subjects  113.90 (13.18) 113.85 (9.77) 110.91 (10.95)      104.93 (14.66) 104.85 (9.93) 109.00 (11.43)  108.85 (16.14) 113.34 (10.28) 111.82 (14.35)      1, 89 2, 89  2, 81      .78 2.75  .24      .38 .07  .70      .01 .06  .005 Broad Reading    Time 1    Time 2    Grade 6 At T1 & T2    Within subject    Between subjects Grade 6    Between subjects  116.40 (12.70) 123.50 (12.15) 119.26 (14.21)   107.71 (15.14) 111.14 (10.43) 120.83 (13.45)  115.50 (11.20) 121.24 (12.15) 121.19 (14.17)      1, 89 2, 89  2, 81      25.43** 4.15*  .16      .001 .01  .85      .22 .09  .003 Basic Reading    Time 1    Time 2    Grade 6 At T1 & T2    Within subjects    Between subjects Grade 6    Between subjects  114.30 (14.06) 125.90 (14.90) 121.48 (16.35)  108.43 (16.32) 114.50 (13.61) 130.25 (14.88)   113.83 (15.06) 120.53 (19.11) 124.92 (14.02)      1, 89 2, 89  2, 81      21.35** 1.43  1.41      .001 .24  .25      .19 .03  .03     81 Continued Table 3.3               L3 Tests Alphabetic Logographic/ Syllabary      Bilingual df F p ᵨ2 Word Reading    Time 1    Time 2    Grade 6 At T1 & T2    Within subjects    Between subjects Grade 6    Between subjects  90.11 (11.26) 91.84 (13.42) 92.09 (10.88)       82.64 (15.36) 83.27 (12.84) 94.33 (13.36)  86.15 (13.74) 88.34 (15.05) 93.11 (12.28)       2, 80 2, 80  2, 81      .73 1.53  .14      .39 .22  .86      .01 .04  .003  Pseudoword    Time 1    Time 2    Grade 6 At T1 & T2    Within subjects    Between subjects Grade 6    Between subjects  90.55 (11.58) 94.80 (11.24) 95.57 (11.86)  89.64 (10.15) 90.09 (12.36) 96.50 (13.65)  89.09 (13.21) 89.87 (13.25) 97.98 (12.89)      1, 81 2, 81  2, 81      3.46 .52  .32      .07 .60  .72      .04 .01  .007  Reading Comp    Time 1    Time 2    Grade 6 At T2 & T3    Within subjects    Between subjects Grade 6    Between subjects  94.79 (8.36) 99.73 (12.76) 103.00 (12.54)   87.82 (6.71) 91.09 (8.38) 97.17 (15.05)  94.00 (6.96) 95.79 (9.35) 103.84 (15.65)      1, 80 2, 80  2, 81      12.33** 3.31*  1.01      .001 .04  .37      .13 .08  .02      82 Continued Table 3.3       Note: *p < .05, **p < .01 L3 Tests Alphabetic Logographic/ Syllabary      Bilingual df F p ᵨ2 Broad Reading    Time 1    Time 2    Time 3 At T1 & T2    Within subjects    Between subjects Grade 6    Between subjects  89.84 (10.55) 94.11 (12.89) 95.87 (10.46)  83.37 (10.80) 85.91 (11.38) 95.00 (12.26)  87.85 (11.07) 89.15 (11.30) 97.89 (13.74)      1, 80 2, 80  2, 81      9.26* 1.65  .38      .003 .20  .69      .10 .04  .008      83 A series of two-way mixed ANOVAs for group (3) and time (2) were carried out on the standard scores of the Letter-Word Identification (word reading), Word Attack (pseudoword reading) and Passage Comprehension (reading comprehension) subtests of the WLBP-R as well as two composite scores: broad reading (letter-word and passage comprehension) and basic reading skills (letter-word and word attack) from the WLPB- R. Findings revealed no main group effect on the letter-word identification, word attack, or the basic reading skills tasks. Conversely, there was a significant main group effect on the broad reading task, with post-hoc (Bonferroni) tests confirming that bilingual students and students with an alphabetic L1 were more proficient in a composite score of word-reading and reading comprehension than the group of students with a logographic/syllabary L1. The main group effect on the passage comprehension task approached, but did not reach, statistical significance (p = .06); however, a comparison of means between the two multilingual groups shows that students with an alphabetic L1 tended to have higher reading comprehension scores than the students with a logographic/syllabary L1. Among all tasks measuring different reading skills in English except reading comprehension, the patterns of results were consistent across the multilingual and bilingual groups: students increased their reading skills in English from the beginning to the end of Grade 4. There was no significant interaction on any of the measures (p > .05). A series of two-way mixed-design ANOVAs for group (3) and time (2) was conducted separately for each French reading task from the WIAT-II. There was a significant effect of group on the reading comprehension task, where Bonferroni post-hoc tests confirmed that the students with an alphabetic L1 and the bilingual students scored higher than the     84 student with a logographic/syllabary L1; however, there was no difference between the alphabetic L1 and bilingual groups. In the word reading, pseudoword reading, and broad reading tasks no group difference was found. There was also a main effect for time: L3 reading comprehension increased from the beginning to the end of Grade 4. There were no main effects for time in word reading and pseudoword reading (p > .05). Findings showed no significant interactions on any of the tasks that measured L3 reading (p > .05).  3.4.3  Comparisons between oral proficiency and reading in L2 and L3  In order to compare the performance on the language proficiency and reading tasks in each of the L2 and L3 of the multilingual students, the scores from the two groups of multilingual students were examined separately. A series of 12 paired sample t tests, with Bonferroni correction for multiple tests (p < .002) were conducted for each measure of language proficiency and reading outcomes in order to examine the student’s performance after one year (T2) of formal reading instruction in both English and French. The mean scores and t test comparisons are presented in Table 3.4. The patterns of results were consistent across the students from both alphabetic and logographic/syllabary L1 backgrounds: their L2 (English) skills and reading proficiency scores were higher than their L3 (French) skills and reading proficiency on all tasks related to language and reading ability. Thus, all students, regardless of L1 typology, had stronger L2 than L3 skills in domains related to language and literacy.     85 Table 3.4 Mean standard scores (SD) and t test in each of the L2 and L3 tests of oral proficiency and reading tasks at the beginning of Grade 4 (Time 1) and at the end of Grade 6 (Time 3): Alphabetic and logographic/syllabary L1 groups Alphabetic L1 Grade 4 tasks L2 L3     df t p d   Listening Comp 113.86 (17.57) 99.95 (12.87) 19 3.86** .001 .43   Vocabulary 114.10 (16.47) 89.80 (16.72) 19 6.63** .001 .70   Word Reading 127.10 (15.59) 91.84 (13.42) 19 14.87** .001 .92   Pseudoword 119.35 (13.33) 94.80 (11.24) 19 9.48** .001 .83   Reading Comp 113.85 (9.77) 99.73 (12.76) 19 6.15** .001 .67   Broad Reading 123.50 (12.15) 94.11 (12.89) 19 11.38** .001 .87  Grade 6 tasks    Listening Comp 114.43 (16.86) 95.96 (13.43) 22 5.22** .001 .59   Vocabulary 114.04 (15.84) 97.22 (15.33) 22 6.39** .001 .65   Word Reading 121.74 (16.41) 92.09 (10.88) 22 9.63** .001 .82   Pseudoword 117.74 (15.46) 95.57 (11.86) 22 10.48** .001 .83   Reading Comp 110.91 (10.95) 103.00 (12.53) 22 3.68** .001 .38   Broad Reading 119.26 (14.22) 95.87 (10.46) 22 10.12** .001 .82 Logographic/syllabary L1  Grade 4 tasks    Listening Comp 104.21 (15.58) 85.26 (11.40) 12 4.61** .001 .64   Vocabulary 102.85 (10.83) 78.27 (16.63) 12 5.00** .001 .68   Word Reading 114.21 (11.30) 83.27 (12.84) 12 9.91** .001 .89   Pseudoword 111.87 (14.80) 90.09 (12.36) 12 5.03** .001 .68   Reading Comp 104.85 (9.93) 91.09 (8.38) 12 4.75** .001 .65   Broad Reading 111.14 (10.43) 85.91 (11.38) 12 8.30** .001 .86  Grade 6 tasks    Listening Comp 116.00 (15.81) 90.67 (11.31) 11 6.40** .001 .65   Vocabulary 110.17 (15.89) 90.50 (15.38) 11 7.97** .001 .74   Word Reading 126.58 (13.83) 94.33 (13.36) 11 9.70** .001 .81   Pseudoword 125.67 (16.11) 96.50 (13.65) 11 8.32** .001 .76     86 Continued Table 3.4  Note: *p < .05; **p < .001. Logographic/syllabary L1 Grade 6 tasks L2 L3 df t p d   Reading Comp 109.00 (11.43) 97.17 (15.05) 11 4.81** .001 .51   Broad Reading 120.83 (13.45) 95.00 (12.26) 11 10.87** .001 .84     87 3.4.4  The effect of oral proficiency and reading in L2 and L3  Correlations between L2 and L3 oral proficiency scores (standard scores) and L2 and L3 reading scores (standard scores) were examined for each of the groups of alphabetic and non-alphabetic L1s separately (see Table 3.5). The findings showed that the L2 and L3 listening comprehension scores were consistently correlated with L2 and L3 reading comprehension and L2 and L3 broad reading tasks within- and across- languages in both L1 literacy groups (r values varied between .50 and .91, p < .05) . Correlations between L2 and L3 listening comprehension scores as well as L2 and L3 vocabulary knowledge and L2 and L3 word reading and L2 and L3 pseudoword reading were generally low in both L1 literacy groups (r values varied between .26 and .54, p > .05), with the exception of L2 vocabulary knowledge being more highly correlated with L2 word reading (r = .78, p < .05) and L2 pseudoword (r = .72, p < .05) reading in the group of students with an alphabetic L1. By contrast, the L2 vocabulary knowledge scores were more strongly correlated with L2 reading comprehension (r = .91, p < .05; r = .52, p > .05) and L3 reading comprehension (r = .64, p < .05; r = .66, p < .05) in the group of students with an alphabetic L1 than the group of students with a logographic/syllabary L1. This same tendency was found between L3 vocabulary knowledge scores with L2 reading comprehension (r = .61, p < .05; r = .31, p > .05) and L3 reading comprehension (r = .72, p < .05; r = .48, p > .05).     88 Table 3.5 Correlations between L2 and L3 reading tasks and L2 and L3 proficiency measures at the end of Grade 4 (T2) for the alphabetic L1 (N = 20) and the logographic/syllabic L1 (N = 13) groups   1 2 3 4 5 6 7 8 9 10 11 Alphabetic L1 1. L2 Listening Comp - 2. L2 PPVT-III .62** - 3. L2 Word Reading .57** .78** - 4. L2 Pseudoword .37 .72** .63** - 5. L2 Reading Comp  .61** .91** .78** .66** - 6. L2 Broad Reading .58** .82** .94** .56* .84** - 7. L3 Listening Comp  .47* .72** .56** .68** .66** .60** - 8. L3 ÉVIP .54* .51* .59** .43 .52* .61** .58** - 9. L3 Word Reading .45* .55* .75** .57** .62** .65** .47* .65** - 10. L3 Pseudoword .26 .48* .45* .57** .51* .35 .28 .33 .76** - 11. L3 Reading Comp .54* .64** .53* .56** .60** .56* .91** .72** .56** .36 - 12. L3 Broad Reading .50* .66** .67** .69** .68** .61** .68** .68** .91** .82** .79**  Logographic/Syllabic L1  1. L2 Listening Comp - 2. L2 PPVT-III .54* - 3. L2 Word Reading .57* .49 - 4. L2 Pseudoword .69** .52 .70** - 5. L2 Reading Comp .63* .52 .50 .65* - 6. L2 Broad Reading .69** .59* .89** .81** .83** -     89 Continued Table 3.5  1 2 3 4 5 6 7 8 9 10 11 Logographic/Syllabic L1 7. L3 Listening Comp  .43 .88** .64* .68* .63* .75** - 8. L3 PPVT-III .52 .56* .58* .65* .31 .57* .45 - 9. L3 Word Reading .73** .53 .70** .76** .34 .66* .53 .59* - 10. L3 Pseudoword .54 .38 .63* .57* .18 .53 .35 .56* .91** - 11. L3 Reading Comp .66* .66* .62* .74** .68* .76** .69** .48 .79** .62* - 12. L3 Broad Reading .70** .59* .70* .75** .44 .71** .57* .62* .97** .91** .87**  Note: *p < .05; **p < .001.     90 Based on the correlation analysis, a third research question was investigated, i.e., whether L2 and L3 oral skills (and vocabulary in particular, since only the correlations with the vocabulary knowledge measures were significant in the two multilingual groups) predicted L2 and L3 reading comprehension differently with respect to L1 typology. A series of hierarchical multiple regression models were tested in order to examine whether L1 typology (alphabetic L1 and logographic/syllabary L1 groups) significantly moderated the predictive relationship between L2 and L3 vocabulary (at T1 and T2) and L2 and L3 reading comprehension (also at T1 and T2). The standard scores of vocabulary in L2 or L3 (predictor) was centered, than entered as the predictor in the first step of the regression, then the interaction term between vocabulary and L1 typology (moderator) was entered in the second regression step. This procedure identified whether the interaction contributed unique variance to the overall model, determined by the statistical significant change in the R 2  between the first and second model (Frazier, Tix, & Barron, 2004).  The results of all the regression models (see Table 3.6) showed a significant moderator effect for L3 French vocabulary on both English (R 2  change = .23, F = 11.24, p = .001) and French reading comprehension (R 2  change = .21, F = 12.93, p = .001) at T1. Similarly, at T2, French vocabulary measured at T1 moderated both L3 French reading comprehension at T2 (R 2  change = .29, F = 16.34, p = .001) and L2 English reading comprehension at T2 (R 2  change = .18, F = 8.92, p = .006). In follow-up analyses, I computed predicted values for the outcome variables (English and French reading comprehension) at T2 for the two multilingual groups based on their French vocabulary scores at T1 (at the mean and 1 standard deviation above and below the mean     91 standard score of French vocabulary). The predicted values obtained from these analyses were then used to create the figures summarizing the moderating effect of French vocabulary scores (see Figure 3.1 and Figure 3.2, respectively). Findings revealed that the association between vocabulary skills and reading comprehension was different in the different language groups. French vocabulary at the beginning of Grade 4 did not significantly predict English and French reading comprehension respectively at the end of Grade 4 in the students with a logographic/syllabary L1. However, in the group of students with an alphabetic L1, increase in French vocabulary was associated with an increase in French reading comprehension.     92 Table 3.6 Language regression equations predicting English and French reading comprehension at the beginning of Grade 4 (T1) and at the end of Grade 4 (T2) Language of measure (reading comprehension)  Predictors R Adjusted R 2  R 2  change F change p  T1 English  T1 ÉVIP, group T1 ÉVIP, group, interaction  .41  .11   .64 .34 .24 11.24** .001   T1 PPVT, group T1 PPVT, group, interaction .81 .63  .81 .63 .008 .71 .41  T1 French T1 ÉVIP, group T1 ÉVIP, group, interaction .55 .26  .73 .47 .22 12.93** .001   T1 PPVT, group T1 PPVT, group, interaction .67 .42  .68 .40 .006 .30 .59  T2 English T1 ÉVIP, group T1 ÉVIP, group, interaction .52 .22  .67 .39 .018 .8.87* .006   T1 PPVT, group .76 .54  T1 PPVT, group, interaction .76 .53 .001 .001 .97   T2 ÉVIP, group .59 .30  T2 ÉVIP, group, interaction .60 .29 .01 .70 .42   T2 PPVT, group .82 .65  T2 PPVT, group, interaction .82 .65 .007 .65 .43  T2 French T1 ÉVIP, group T1 ÉVIP, group, interaction .48 .18  .72 .47 .29 16.34** .001   T1 PPVT, group T1 PPVT, group, interaction .72 .49  .72 .47 .001 .001 .97   T2 ÉVIP, group T2 ÉVIP, group, interaction .69 .44  .72 .47 .04 2.63 .12   T2 PPVT, group T2 PPVT, group, interaction .67 .41  .67 .39 .001 .002 .96     93 Note: *p < .05; **p < .001.                          94 Figure 3.1 Interaction of language group by French vocabulary scores (standard scores) at the beginning of Grade 4 (T1) on English reading comprehension (standard scores) at the end of Grade 4 (T2)  0 20 40 60 80 100 120 140 160 low Fr_PPVT mid Fr_PPVT high Fr_PPVT E n g li sh  r e a d in g  c o m p re h en si o n  T 2  Non-alphabetic Alphabetic     95 Figure 3.2 Interaction of language group by French vocabulary scores (standard scores) at the beginning of Grade 4 (T1) on French reading comprehension (standard scores) at the end of Grade 4 (T2)  0 20 40 60 80 100 120 140 160 low Fr_PPVT mid Fr_PPVT high Fr_PPVT F re n ch  r ea d in g  c o m p re h en si o n  T 2  Non-alphabetic Alphabetic     96 3.5  Results from the Grade 6 sample  The L2 and L3 language proficiency was examined for multilingual students from all L1 backgrounds and for the bilingual students in Grade 6. A series of one-way between-groups ANOVAs for group (3) were conducted on the standard scores using the same measures of English listening comprehension from the WLPB-R, English vocabulary knowledge from the PPVT-III, the French listening comprehension from the WIAT, and French vocabulary knowledge from the ÉVIP. Findings revealed no main group effect on any of the measures of listening comprehension or vocabulary knowledge in either English or French (p > .05) (see Table 3.2). Next, groups were compared on their English and French reading ability. For these analyses, the performance of the students with a logographic/syllabary L1 was compared with that of the students with an alphabetic L1 and that of bilingual students at the end of Grade 6. A series of one-way ANOVAs for group (3) were tested on the same measures used in Grade 4 of English reading from the WLPB-R and French reading from the WIAT-II.  Results showed no main group effect on any of the tasks of word reading, pseudoword reading, reading comprehension, basic reading, or broad reading skills in either English or French (p > .05) (see Table 3.3). In order to compare the performance on the language proficiency and reading tasks in each of the L2 and L3 of the multilingual students, the scores from the two groups of multilingual students were examined separately. Identical analyses to the Grade 4 data were completed: a series of 12 paired sample t tests, with Bonferroni correction for multiple tests (p < .004) were conducted for each measure of language proficiency and     97 reading outcomes in order to examine the student’s performance after three years of formal English reading instruction and after seven years of formal French reading instruction. The mean scores and t test comparisons are presented in Table 3.4. The patterns of results were consistent across the students from both alphabetic and logographic/syllabary L1 backgrounds: their L2 (English) skills and reading proficiency scores were higher than their L3 (French) skills and reading proficiency on all tasks related to language and reading ability. Next, the association between oral proficiency in the L2 and L3 and reading skills in the L2 and L3 was investigated for multilingual students. Correlations between L2 and L3 oral proficiency scores (standard scores) and L2 and L3 reading scores (standard scores) were examined for each of the groups of alphabetic and logographic/syllabary L1s separately in Grade 6 (see Table 3.7). Correlations between L2 and L3 listening comprehension scores, L2 and L3 vocabulary scores, L2 and L3 word reading, and L2 and L3 pseudoword reading were generally low in both multilingual groups. Moreover, associations within and across vocabulary scores and reading comprehension scores were moderate in the two multilingual groups. By contrast, correlations between L2 listening comprehension and L2 and L3 broad reading were stronger in the group of students with an alphabetic L1 than in the group of students with a logographic/syllabary L1. Based on the correlation analyses, the question was whether L2 listening comprehension predicted L2 and L3 broad reading differently based to L1 typology. Subsequently, a series of hierarchical multiple regression models were tested in order to explore whether L1 typology influenced the association between L2 listening comprehension and L2 an L3 broad reading at T3. When testing the moderated     98 regressions, listening comprehension in L2 (the standard scores) was centered, than entered as the predictor in the first step of the regression; then, the interaction term between listening comprehension and L1 typology was entered in the second regression step. Findings for the regression models (see Table 3.8) showed that language typology was not a significant moderator for L2 listening comprehension and L2 and L3 broad reading (p > .05) at T3.     99 Table 3.7 Correlations between L2 and L3 reading tasks (standard scores) and L2 and L3 proficiency measures (standard scores) at end of Grade 6 (T3) for the alphabetic L1 (N = 23) and the logographic/syllabic L1 (N = 12) groups  1 2 3 4 5 6 7 8 9 10 11 Alphabetic L1 1. L2 Listening Comp - 2. L2 PPVT-III .74** - 3. L2 Word Reading .84** .57** - 4. L2 Pseudoword .46 .39 .65** - 5. L2 Reading Comp  .36 .50* .46* .27 - 6. L2 Broad Reading .70** .63** .94** .62** .72** - 7. L3 Listening Comp  .53** .46* .30 .29 .73** .51* - 8. L3 ÉVIP .49* .67** .37 .05 .54* .48* .59** - 9. L3 Word Reading .52* .32 .47* .58** .70* .39 .23* .22 - 10. L3 Pseudoword .17 .37 .61** .75** .34 .69** .24 .24 .76** - 11. L3 Reading Comp .74** .43* .30 .22 .63** .46* .73** .57** .18 .19 -  12. L3 Broad Reading  .65**  .49*  .60**  .68**  .47*  .64**  .55**  .46*  .83**  .83**  .63**     100 Continued Table 3.7   1 2 3 4 5 6 7 8 9 10 11 Logographic/syllabic L1 1. L2 Listening Comp - 2. L2 PPVT-III .48* - 3. L2 Word Reading .07 .74* - 4. L2 Pseudoword .22 .13 .64* - 5. L2 Reading Comp .24 .55* .75* .52* - 6. L2 Broad Reading .25 .69* .95** .62* .88** - 7. L3 Listening Comp  .39 .54 .54 .31 .44 .50 - 8. L3 PPVT-III .36 .85** .58* .02 .57* .62* .63* - 9. L3 Word Reading .14 .30 .64** .67* .46 .65* .28 .32 - 10. L3 Pseudoword .21 .07 .35 .68* .20 .34 .08 .11 .76** - 11. L3 Reading Comp .35 .78** .82** .47 .70* .81** .81** .76** .38 .01 - 12. L3 Broad Reading .21 .46* .81** .78** .60* .70** .47 .46 .92** .75** .63*  Note: *p < .05; **p < .001.     101 Table 3.8 Language regression equations predicting English broad reading and French broad reading for all participants at end of Grade 6 Language of measure (broad reading) Predictors R Adjusted R 2  R 2  change F change p   English  Grade 6 listening comp, group Grade 6 listening comp, group, interaction  .32  .10   .41 .16 .06 2.43 .18   French  Grade 6 listening comp, group Grade 6 listening comp, group, interaction  .34  .02   .41 .09 .07 2.78 .12   Note: *p < .05; **p < .001.       102 3.5.1  Comparisons between Grade 4 and Grade 6 samples on oral proficiency and reading in L2 and L3  As reported above in the results section, students with an alphabetic L1 had stronger L2 and L3 oral proficiency and stronger L2 and L3 reading comprehension skills than the group of students with a logographic/syllabary L1 in Grade 4. However, in the Grade 6 sample, there were no group differences on any of the measures in English and French. There are a number of possible explanations for these findings: first, the oral language proficiency and reading skills of the students with logographic/syllabary L1 increased to a greater degree over the two-year period than those of the students with alphabetic L1s. A second possibility is that the students with a logographic/syllabary L1 who may have had generally low scores on the measures of oral language and reading ability were also the students who left the FI program at the end of Grade 4; thus, only students with strong oral language and reading skills in English and French remained in FI programs in Grade 6. In order to investigate more formally the first explanation, the scores on all measures of oral language proficiency and reading in English and French for the eight students with a logographic/syllabary L1 who were tested in Grade 4 and in Grade 6 were examined. On the measure of English listening comprehension, three students demonstrated a large increase (more than one standard deviation above the group average) from Grade 4 to Grade 6, whereas four different students showed an increase on the measure of English vocabulary knowledge. Four students (including two of the same students on scores of vocabulary knowledge) showed an increase in English reading     103 comprehension and five students (including same three students on reading comprehension) had significant increases in English broad reading. On the measure of French listening comprehension, four different students demonstrated an increase, while four students (including the same three on the listening comprehension test) showed an increase on the French vocabulary task. Three different students demonstrated an increase on the French reading comprehension measure and only two students had an increase on the French broad reading task (see Table 3.9). Thus, there is evidence that students with a logographic/syllabary L1 who remained in FI programs showed a general increase in their scores on measures of oral language and reading ability in English and French. Furthermore, no students showed a significant decrease (more than one standard deviation below the group average) from Grade 4 to Grade 6.  In conclusion, the observations from the current study suggest that at least some students with non- alphabetic L1s require additional exposure to formal academic instruction to attain equivalent English and French oral proficiency and French reading skills in FI programs. To explore the second possible explanation, the 10 students who did not participate in Grade 6 were identified and their performance across the different measures of oral language proficiency and reading ability in English and French at T1 and T2 were examined. In the group of five students with a logographic/syllabary L1, there were two students who showed patterns of low performance (more than one standard deviation below the group average) on the oral language and reading measures in English and French. In particular, these two students had low scores in listening comprehension and vocabulary in English and French, as well as lower skills in reading comprehension and broad reading in both English and French. One of the students from the     104 logographic/syllabic L1 group showed stronger skills on the measures of vocabulary knowledge and reading comprehension in English and French. In the group of five students with an alphabetic L1, one student demonstrated a consistent pattern of low scores on all measures of oral language proficiency and reading in English as well as low performance in listening comprehension and vocabulary in French. One other student in the alphabetic L1 group showed stronger (more than one standard deviation above the group) skills on all measures in English and higher scores in French reading comprehension. Therefore, there was no strong indication that only those students with a logographic/syllabary L1 and low oral language proficiency and reading skills in English and French tended to drop out from the FI programs. The current findings suggest that the first explanation – that students with a logographic/syllabary L1 showed a general increase in oral language and reading ability in English and French from Grade 4 to Grade 6 – is the more feasible of the two alternatives.     105 Table 3.9 Mean standard scores from the eight students with a logographic/syllabic L1 on measures of English and French oral proficiency and reading tests at the beginning of Grade 4 (T1), the end of Grade 4 (T2), and the end of Grade 6 (T3)   L2 tests Student ID 1 2 3 4 5 6 7 8 Listen comprehension    Time 1    Time 2    Grade 6   90 95 104   124 119 115   109 122 124   78 84 97   81 104 110   82 110 116   81 95 97   103 112 128 PPVT-III    Time 1    Time 2    Grade 6  93 93 99  113 113 117  110 109 121  92 93 92  84 88 101  78 92 118  94 98 108  106 112 127 Reading Comp    Time 1    Time 2    Grade 6  91 102 106  125 121 138  108 119 120  97 110 114  99 106 109  93 91 103  99 93 115  98 99 116 Broad  Reading    Time 1    Time 2    Grade 6  96 106 102  125 131 140  125 122 138  99 104 118  110 114 111  95 97 112  107 110 123  108 102 116 L3 tests Listen comprehension    Time 1    Time 2    Grade 6   77 86 92   104 108 108   99 105 104   69 79 91   83 73 96   73 86 99   70 79 95   99 94 99      106 Continued Table 3.9   L3 tests Student ID 1 2 3 4 5 6 7 8 ÉVIP    Time 1    Time 2    Grade 6  67 66 78  100 110 111  106 110 118  75 61 89  59 68 78  61 64 93  63 77 84  114 110 111 Reading comp    Time 1    Time 2    Grade 6  78 87 87  95 102 107  100 102 115  89 93 93  87 81 94  84 87 85  81 88 90  89 93 110 Broad Reading    Time 1    Time 2    Grade 6  67 67 76  90 101 100  106 108 118  88 91 95  86 89 92  77 86 85  91 87 93  93 90 104       107 3.6  Discussion  The present study set out to explore the extent to which the acquisition of oral language proficiency and literacy skills in the second and third languages of multilingual students can be represented in terms of orthographic similarities to the L1. In addressing the first research question, similar to previous research in L2 (Koda, 1989) and L3 (Leung, 2005; Swain et al., 1990) development, results from the current study demonstrated that listening comprehension and vocabulary skills in L2 and L3 are at least somewhat influenced by the first language typology. After one full year of English instruction and five years of French instruction (i.e. at end of Grade 4), the bilingual students and the group of students with an alphabetic L1 had stronger English and French oral skills than the group of students with a logographic/syllabary L1. Moreover, the bilingual students and the students with an alphabetic L1 also both had better English and French reading comprehension skills. A large body of research has revealed that the relationship between vocabulary and reading comprehension is a reciprocal one, where vocabulary growth takes place through reading, and that students with larger vocabularies tend to better comprehend written text (McKeown, Beck, Onanson, & Perfetti, 1983; Stanovich, 1986). Reading is a significant contributor to the development of vocabulary (Nagy & Anderson, 1984) and, by extension, higher proficiency of L2 and L3 reading comprehension in the group of students with an alphabetic L1 may have also led to an increase in their L2 and L3 vocabulary size. Therefore, L1 typology per se may not have explained group differences in L2 and L3 oral skills; rather, increased proficiency of L2 and L3 reading comprehension may have resulted in higher L2 and L3 vocabulary skills     108 in the group of students with an alphabetic L1. Alternatively, it is possible that the students with a logographic/syllabary L1 had a greater amount of home language exposure and a smaller degree of English and French exposure when compared with the students with an alphabetic L1. This increase in L1 support could have contributed to weaker L2 and L3 oral proficiency because the students with a logographic/syllabary L1 used English and French less frequently than the students with an alphabetic L1. However, the students in the current study had equal exposure to English (L2) and French (L3) in terms of the number of people around them who spoke to them in the L2 and L3. The second objective of this study was to explore how L1 typology influenced the development of L2 and L3 literacy skills. In line with previous research on L2 literacy acquisition (e.g., Bialystok et al., 2005; Dugunoglu et al., 1993; Gottardo et al., 2001), it was expected that L1 typology would be associated with literacy skills in the L2 and in the L3 of multilingual students. The effects of L1 typology on the development of reading proficiency in L2 and L3 can be influenced by factors such as language of literacy instruction, home literacy practice, length of residency in the country where L2 and L3 are acquired, age of acquisition, the degree of similarity of orthography between languages, and SES. Because participants in the present study were acquiring their L2 and L3 at equivalent ages, had the same language of literacy instruction (French from Kindergarten to Grade 3; then English in Grade 4), were raised in households with similar SES, and received equivalent L2 and L3 literacy instruction in school, it seems plausible that the one factor that could explain their performance was the degree of typological similarity in terms of orthography between their L1 and the target L2 and L3.     109 The present findings provided further support to the script-dependent hypothesis showing consistent effects of L1 typology on the development of literacy skills in the L2 and L3. When the multilingual students were compared based on how closely their L1 writing system matched the orthography of the L2 and L3, the group of students with an alphabetic L1 performed just as well as the bilingual students on all measures of L2 and L3 reading; and it was only the students with a logographic/syllabic L1 who had much lower proficiency in L2 reading comprehension. Moreover, it was found that the lower performance of the group of students with a logographic/syllabic L1 was present not only in L2 reading comprehension, but also in the L3. These findings suggest that multilingual students may process differently French and English reading, depending on how closely their L1 writing system corresponds to the orthography of the L2 and L3. This hypothesis presents a complementary theory to the script-dependent hypothesis that acknowledges both positive and negative transfer in terms of specific linguistic features in the L1, L2, and L3. The students with an alphabetic L1 may have benefited from sharing a common writing system with the target L2 and L3 based on a shared knowledge of the alphabetic principle which leads to greater efficiency when decoding text (Adams, 1990; Akamatsu, 1999; Bialystok et al., 2005). The results from the current study contrasted with those of Gunderson (2007) who found that orthography in the L1 did not influence the development of reading comprehension in the L2 (English) of bilingual students between the ages of 8 and 17 years. The extent and the type of exposure to L1 and L2 literacy instruction in the Gunderson (2007) study were different from the current study. However, immigrant students in Gunderson (2007) were initially exposed to literacy in their L1 in their home     110 country and subsequently received formal instruction in English at school. The majority of multilingual students in the current study were exposed to literacy in the L1 and the L2 simultaneously before the age of 5 years, and then were exposed to French literacy at the age of 5 years in school. Similarly to the findings in Bialystok et al. (2005) and Gottardo et al. (2001), students in the current study were exposed to the L1 and L2 from an early age and the L1 typology may have had an influence on L2 and L3 reading acquisition. Gunderson (2007) also suggested that the L1 instructional practices of Chinese- speaking students may have had an effect on L2 reading comprehension. In two separate longitudinal studies carried out with Chinese-speakers learning English in Canadian high schools, Gunderson et al. (2012) and Murphy et al. (2012) found that Mandarin-speaking girls were four times more likely to be qualified for university than Cantonese-speaking boys. Moreover, two-thirds of Cantonese speakers who went to school in Vancouver did not have high enough grades at the end of Grade 12 to enter university. The current study did not categorize Chinese-speaking students into specific groups according to L1 instruction given the small number of participants in the logographic/syllabary L1 group. In summary, the L1 instructional practice may have an important influence on the acquisition of reading skills in the L2 and academic readiness. Future studies should consider the effect of L1 instruction in Chinese-speaking students on the acquisition of reading comprehension in the L2 and L3. According to the script-dependent hypothesis it was expected to find group differences in all aspects of literacy, including word reading and pseudoword reading. The results failed to show any group differences in L2 or L3 word-level reading skills. One possible explanation could be that the students were old enough and were already     111 fluent in decoding skills (decoding scores on L2 word reading and L2 pseudoword reading were well within and above the expected age range for the groups in this study); and they were now developing their reading comprehension skills. The final objective was to examine whether L1 typology influenced the relationship between L2 and L3 vocabulary and L2 and L3 reading comprehension. Findings showed that for the group of students with an alphabetic L1 an increase in French vocabulary was associated with an increase in English and French reading comprehension; whereas, in the group of students with a logographic/syllabic L1, an increase in the French vocabulary skills was not associated with an increase in their English and French reading comprehension. These results are in line with previous research by Koda (1989) and August et al. (2005) whose findings showed that L2 vocabulary knowledge was positively associated with L2 reading comprehension in languages that share a common alphabetic system. Conversely, when the target language is typologically dissimilar from the L1, transfer of oral skills is less likely and students show less advanced proficiency in reading comprehension (Koda, 1989). However, in the present study, only French vocabulary skills were associated with reading comprehension in both English and French, even though the student’s French vocabulary skills were lower than their English skills. It is possible that as a result of the more intense formal academic instruction in French in the first four years of French immersion students had acquired vocabulary in French that consisted of words that are more commonly used in school, i.e., in the context of an academic discourse. Those were also perhaps words that are helpful with comprehension of academic texts. On the other hand, even though their English vocabulary was larger, it is possible that it consisted of words that are not     112 essential for classroom discourse, and thus less likely to assist in the comprehension of academic texts (e.g. textbooks used in social science or math classes). Bournot-Trites & Séror (2003) explained that as bilingual English-French students progress through intermediate grades (Grades 5 through 8), they must develop sufficiently strong reading and writing skills in French that allow them to comprehend textbooks in social studies, mathematics, and science. Thus, as bilingual and multilingual students progress through the intermediate grades, they must be able to read and comprehend increasingly complex content in French. Findings from the current study suggest that the more intense and direct academic instruction in French could help immersion students develop stronger reading skills. An alternative explanation for group differences in the development of French vocabulary skills related to L2 and L3 reading comprehension is that the group of students with an alphabetic L1 may have benefited from shared vocabulary cognates between their L1 and French; the students with a logographic/syllabic L1 on the other hand may not have had the opportunity to develop cognate awareness because their L1 and French did not share any cognates (i.e. languages that shared alphabetic knowledge also benefitted from cognate awareness) (Carlo et al., 2004; Chen, Ramirez, Luo, Geva, & Ku, 2012; Jarvis & Pavlenko, 2008). Chen et al. (2012) have demonstrated that cognate awareness only develops in an L2 when the first language is orthographically related to the target second language. By Grade 6, students with a logographic/syllabary L1 had oral language proficiency and reading skills in English and French equivalent to the skills of the students with an alphabetic L1 and the bilingual students. Because the participants in     113 Grade 4 and Grade 6 differed, it is not possible to determine with certainty whether the students with a logographic/syllabary L1 substantially increased their oral language proficiency and reading skills and caught up with the other two language groups. Nonetheless, a few students with a logographic/syllabary L1 who remained in the FI program from Grade 4 through Grade 6 increased their listening comprehension, vocabulary knowledge, and reading skills. Of the students who left the FI program after Grade 4, there was no clear pattern that they had generally lower language knowledge. Combined, these two explanations suggest the possibility that although students from orthographically dissimilar L1 backgrounds initially have lower skills, they may eventually catch up to their alphabetic L1 and bilingual peers. The students in the present study were enrolled in FI programs in culturally diverse communities in a major urban city in Western Canada. It is possible that these results may not be readily generalizable to similar populations in other parts of Canada, or across the world.  Therefore, more research is needed to investigate the development of oral language proficiency and reading skills of groups of students from alphabetic and non-alphabetic L1s in a variety of FI programs in earlier and later grades. It is also possible that students with a logographic/syllabary L1 who had been taught for seven years in French have received sufficient academic instruction to catch up to the students with an alphabetic L1. Cummins (2000) has indicated that bilingual students may require up to seven years of formal academic instruction to develop strong academic skills (cognitive academic language proficiency) in another language. The current findings suggest that it may take students whose L1 is orthographically different than the L3 up to seven years to reach parity with their bilingual and alphabetic L1 peers,     114 with regard to oral language proficiency and reading comprehension skills. To confirm whether the students from orthographically dissimilar backgrounds significantly increase their English and French skills, future studies will need to include larger samples and measure the performance of the same students across several grades. The data presented here supported the research by Leung (2005) and Swain et al. (1990) who also found that L1 typology influenced the development of skills in the L3 to a greater extent when languages were orthographically similar. In the present study unlike the Swain et al. (1990) study, multilingual students with distinct L1 typologies (alphabetic and non-alphabetic) and who were literate in their L1 were included, thus allowing a greater confidence about the effect of L1 typology, particularly in terms of orthography, on L2 and L3 reading proficiency. Based on the alphabetic similarities between the L1 and the target L2 and L3, it seems that L1 typology influences the development of reading comprehension in both the second and third language. An effort was made to assure that all students were literate in their L1 because this was essential for the research design; however, future studies should also gather more detailed information regarding the student’s use of the different languages in their homes. Due to the diversity of L1s represented in the sample (representative of the L1s in the school demographics in this Western Canadian city), it was not possible to formally measure L1 oral skills and literacy proficiency in all the languages. As an initial study, information was gathered on oral language and literacy proficiency of the Chinese- speakers in the L1, L2 and L3 and, ideally, future studies should aim to include students with the same L1 background and directly measure their L1 oral skills and L1 literacy     115 ability. Based on the data from the study, it was revealed that some of the students did not have strong oral language proficiency and reading skills in Chinese. The current study involved numerous analyses on a limited sample, which can give rise to statistical type I errors (i.e. rejecting the null hypothesis when it is actually true). In cases of group comparisons, an effort was made to counter the effects of statistical type I errors by applying Bonferroni correction for multiple tests (Holm, 1979). Future studies should include a larger sample of students, and include other statistical analyses such as factor analyses that would allow researchers to investigate a greater number of factors that influence the acquisition of reading skills of multilingual students (Gunderson, 2007). Furthermore, given the relatively limited sample of multilingual students from the logographic/syllabary L1 group and from the alphabetic L1 group and the considerable attrition of participants from Grade 4 to Grade 6, the conclusions drawn from this study remain conditional until additional research explores a larger number of participants. Findings from the current study nevertheless suggest important associations between L1 typology and the acquisition of oral language proficiency and reading skills in the L2 and L3 of multilingual students.   In light of these main findings, specifically to address the gap in English and French reading proficiency of multilingual students with typologically dissimilar L1s, many educational strategies could be adopted by teachers to increase the English and French reading skills of all multilingual students. First, there appears to be a need to strengthen the English and French oral language proficiency, particularly vocabulary, which, in turn, could lead to increased English and French reading comprehension skills. Secondly, listening comprehension skills in English and French were strongly correlated     116 within and across English and French reading comprehension in both multilingual groups; therefore, it will be important to foster these skills in both languages in schools. Educators, researchers, and policy-makers have identified the need to explore whether multilingual students in French immersion programs develop the same level of language skills and literacy proficiency as bilingual students in French immersion programs (Genesee & Jared, 2008; Swain & Lapkin, 2005). In contrast to the well- documented success of students from English monolingual backgrounds in French immersion programs (Genesee, Holobrow, Lambert, Cleghorn, & Wallig, 1985; Turnbull, Lapkin, & Hart, 2003), the present findings suggest that multilingual students with an orthographically dissimilar L1 may experience difficulties in L2 and L3 reading after 5 years of French instruction and 1 year of formal English instruction. By Grade 6, however, students with an orthographically different L1 appeared to have caught up with students with an orthographically similar L1. Therefore, it is important to foster the development of oral language proficiency in both English and French of multilingual students to ensure continued academic success of all learners.     117 Chapter 4: The effect of socio-linguistic factors and English language proficiency on the development of French as a third language  4.1  Synopsis  The classroom demographics in French immersion (FI) programs across Canada are changing: there are a growing number of multilingual students who are learning English as a second language (L2) and French as a third language (L3). However, little is known about the development of French language proficiency and reading skills of multilingual students in the FI programs. The association between socio-linguistic factors and metalinguistic awareness, in addition to language proficiency, and literacy skills in the L2 and language proficiency and literacy skills in the L3 were investigated in this study. The sample included 55 students in early FI programs who were tested at the end of Grade 6. Multiple regression analyses revealed that oral language proficiency and reading comprehension in English predicted oral language proficiency and reading comprehension skills in French, controlling for amount of reading in French and morphological awareness in English. This study demonstrates that although the majority of instruction in FI programs is in French, multilingual students continue to develop strong oral language proficiency and reading skills in English and French.         118 4.2  Introduction  Since the 1960s, the vast majority of students enrolled in French immersion (FI) programs in Canada have been from English monolingual backgrounds, and numerous studies have documented the success of these students in developing proficiency in English and French as a second language (L2) (Genesee, 1983; Genesee & Jared, 2008; Turnbull, Lapkin, & Hart, 2003). Within one to two years of formal instruction, the English monolingual students in FI reach parity in English vocabulary skills and reading comprehension when compared with their monolingual peers in English mainstream programs (Genesee, 1983; Genesee & Jared, 2008; Swain & Lapkin, 1982). Moreover, many bilingual students experience advantages in terms of academic success, metalinguistic skills, and cognitive flexibility (Bialystok, 2007). Today, however, close to 50% of the total student population in FI programs comprises multilingual students from homes where English is not the dominant language (Vancouver School Board, 2012). An ever-increasing number of multilingual students are entering FI programs (Genesee & Jared, 2008; Taylor, 2006), and the majority of them are from language backgrounds for which the first language is not supported or taught at school. In this paper, the term first/home language (L1) is used to refer to the language, other than English or French, that all the students were exposed to at birth and is used regularly in the home, second language (L2) refers to English, the language some students may have been exposed to at birth, but use less regularly in the home, and third language (L3) refers to French, the language students were first exposed to at school entry. Students were not exposed to French prior to starting school in Kindergarten. The     119 order of the languages does not reflect the competency in each achieved by individual learners. Given the large number of home languages present in FI programs, it is not possible to foster academic skills in all the L1s. When students’ oral language and literacy in the L1 are not supported, their level of bilingualism may not be sufficient to benefit their L3 acquisition (Cenoz, 2003; Cummins, 2000; Jessner, 2008). As reported in study 1, when compared to bilingual students in FI programs, the multilingual students demonstrated equivalent French oral language skills and comparable French reading skills. Little is known about other factors that influence the development of oral language proficiency and reading skills in the L3, French, of multilingual students in FI programs. Because the L1 skills of multilingual students are not supported in school, it remains important to identify the socio-linguistic factors and oral language skills in the L2 (English) that facilitates oral language and reading skills in the L3 (French). According to Cummins’ Interdependence hypothesis, skills acquired in the L1 and L2 facilitate the development of an L3 when the L1 is supported in the school (Cummins, 2000). Alternatively, when the L1 is not sufficiently supported in the school context, it remains underdeveloped and can be a limiting factor in the development of subsequent languages (Cummins, 2000). A number of studies involving groups of immigrant students whose L1 (Arabic, Berber, Moroccan, Surinam, Turkish) is neither taught nor supported in school have shown mixed results; some have shown no bilingual advantages (Sanders & Meijers, 1995), and others have shown that monolingual students even obtained higher oral language proficiency or literacy skills in the target language (Schoonen et al., 2002; Van Gerlderen et al., 2003).     120 On the other hand, two studies conducted with multilingual students learning English as an L3 (Klieme, 2006) or French as an L3 (Haenni Hoti et al., 2011) have shown that multilingual students in an education setting that did not support the L1 obtained equivalent or even higher proficiency than monolingual students. Klieme (2006) studied 10,632 multilingual students (13% of the students were immigrants) learning German as an L2 and English as an L3 and found that the immigrant students in Grade 9 who were not born in Germany had equivalent proficiency in German and English grammatical awareness, vocabulary knowledge, and reading comprehension to the monolingual German students learning English as an L2, given similar conditions in terms of socioeconomic status and gender. However, it was unclear what languages the immigrant students spoke at home and how long they had been exposed to English as an L3. In the second study by Haenni Hoti et al. (2011), 928 multilingual students in Grades 3 and 5 in Switzerland (13% of the students were immigrants) learning English as an L2 and French as an L3 were found to have higher scores in French listening comprehension than monolingual German students learning French as an L2. Moreover, home literacy practice (number of books at home) and motivation to learn a foreign language, were also found to be important predictors in the acquisition of L3. The conflicting results from the aforementioned studies conducted with multilingual students developing an L3 in a bilingual setting that does not support the L1 raise questions as to what additional factors other than bilingualism – such as socio- linguistic factors and proficiency in the L2 – influence L3 acquisition. Hufeisen and Marx’s (2007) ‘factor model’ addresses this question and identifies a number of factors that explain differences in the acquisition of an L2 and an L3. Whereas the L2 student is a     121 novice learner in the process of L2 acquisition, the L3 student has already learned about the process of learning a second language. Therefore, L3 students have language-specific knowledge and abilities at their disposal that L2 students typically do not. According to the ‘factor model’ by Hufeisen and Marx (2007) six different variables influence L3 acquisition: (1) neurophysiological factors (e.g. age), (2) learner external factors (e.g. socio-economic status, type and amount of language input), (3) affective factors (e.g. motivation), (4) cognitive factors (e.g. metalinguistic awareness), (5) foreign-language specific factors (e.g. transfer: the influence of the L1 on the acquisition of the L3, and (6) linguistic factors (e.g. vocabulary knowledge and listening comprehension in L2). Moreover, the individual trajectories of L3 development could be different and some factors may exert greater influence on the acquisition of L3; whereas, others may become less important. Each of these factors will be reviewed in the next sections.  4.2.1 Neurophysiological factors  The age at which learners are exposed to the L3 can influence the acquisition oral language proficiency and reading skills (Jessner, 2008). Multilingual learners may include children who grow up with three languages simultaneously from birth, bilingual children with simultaneous acquisition of the L1 and L2 before learning the L3 at school, or children with simultaneous acquisition of the L2 and L3 after learning the L1. In the current study, age of L3 acquisition was constant across all participants and therefore was not considered in determining French language acquisition. The students were first     122 exposed to French as an L3 when they were five years old in Kindergarten and the majority of students were first exposed to English (L2) from birth.  4.2.2  Learner-external factors  While the level of literacy skills of students learning an L2 has been shown to be significantly associated with socio-economic status (SES) in students throughout elementary grades (Gunderson, 2007) and secondary grades (Gunderson, D’Silva, Murphy, & Odo, 2012; Murphy, Odo, D,Silva, & Gunderson, 2012), others have revealed that the association between SES and literacy skills systematically declined and was non- significant by Grade 3 (D’Angiulli, Siegel, & Maggi, 2004). Moreover, in several studies that have examined the association between SES and the acquisition of an L3, for example French (Swain, Lapkin, Rowen, & Hart, 1990) and English (Cenoz & Valencia, 1994; Sanz, 2008), SES was not a significant predictor of reading outcomes. Therefore, it is essential to control for the effect of SES, given the somewhat conflicting results, thus allowing a better understanding of the nature of L3 reading acquisition (Swain et al., 1990).  The effects of L3 language use and literacy practice on the development of oral proficiency and literacy in the L3 have been established, showing that the amount of L3 used at home (Cenoz & Valencia, 1994; Sanz, 2000), and home literacy practice in the L3 (such as the number of books at home) (Haenni Hoti et al., 2011) are important predictors of oral language and reading achievement in the L3.  Moreover, Baker (2011) highlights the importance of including scales that measure language use and not just language     123 knowledge into research dealing with multilingual development, in order to fully describe the patterns of L3 acquisition.  4.2.3  Affective factors: Motivation to learn a third language  The bulk of research on the effect of motivation on L2 acquisition has been guided by Gardner’s (1985) Socio-Educational Model. Gardner (1985) defines motivation as “the extent to which the individual works or strives to learn the language because of a desire to do so and the satisfaction experienced in this activity" (p. 10). In the Socio-Educational Model, motivation is a construct characterized by two orientations: a) instrumental and b) integrative. Instrumental orientation refers to the overall regard for other groups of people who speak a language (e.g. attitudes towards a foreign group of people), whereas integrative orientation includes measures of effort and desire to learn a second language and the satisfaction of learning the target language. Instrumental and integrative motivation have been examined in the context of L2 acquisition (e.g., Gardner & Lasynchuck, 1990); and likewise, research in L3 acquisition has examined the association between instrumental and integrative motivation and the acquisition of oral language proficiency and reading skills in the L3 (Cenoz & Valencia; 1994; Sanz, 2000; Haenni Hoti et al., 2011). In all studies, only the integrative motivation factor was a strong and significant predictor of L3 oral language proficiency and reading comprehension skills. On the other hand, the measure of instrumental motivation was not found to be strongly associated to oral language proficiency or reading in the L3.      124 4.2.4  Metalinguistic factors: Morphological Awareness in L2  There is a substantial body of evidence showing the importance of the role of metalinguistic awareness in reading comprehension (see Nagy, 2007, for a comprehensive review). Metalinguistic awareness is referred to as the ability to reflect and manipulate the structural features of language (Nagy, 2007). Two of its component skills, phonological awareness and morphological awareness, have been studied in relation to reading development (Nagy, 2007). While phonological awareness has been demonstrated to be a crucial skill for the acquisition of decoding skills in early elementary grades (see National reading panel, 2000), morphological awareness has been shown to be a stronger predictor of reading comprehension than phonological awareness for students in the later elementary grades (Grade 5) (Deacon & Kirby, 2004). While there is evidence that inflectional morphology, or the awareness of inflected forms of words (e.g. jump/jumped), is associated with word reading skills in younger students in Grade 1 and Grade 2 (Nagy, Berninger, & Abbott, 2006), derivational morphology or the ability to manipulate the structure of morphologically complex words and reflect on its meaning (for example, product/produce/production) has proved to be essential for the development of reading comprehension in later elementary grades (Carlisle, 2000). More than fifty percent of words in English and French are morphologically complex (Casalis & Louis-Alexandre, 2000; Nagy, 2007) and with each grade level students encounter more morphologically complex words. The ability to quickly recognize the morphological structure of these complex words allows students to infer meaning from written text (Nagy et al., 2006). Therefore, morphological awareness, and more     125 specifically derivational morphology, was used as the measure of metalinguistic awareness in the present study. Many studies have shown that morphological awareness is an important skill for reading development in monolingual students across a variety of languages and orthographies, such as Arabic (Abu-Rabia, 2007), Chinese (Ku & Anderson, 2003), English (Deacon & Kirby, 2004), and French (Casalis & Louis-Alexandre, 2000). Morphological awareness is defined as the ability to segment words into their morphological units (Carlisle, 2000). By contrast, only four studies have examined the cross-language association between morphological awareness and reading comprehension in bilingual students; namely English-speaking students acquiring French as an L2 (Deacon, Wade-Woolley, & Kirby, 2007), Chinese-speaking students acquiring English as an L2 (Pasquerella, Chen, Lam, Luo, & Ramirez, 2011; Wang, Cheng, & Chen, 2006), and Korean-speaking students learning English as an L2 (Wang, Ko, & Choi, 2009). Based on their findings it appears that the morphological structure of the languages involved determines the presence and direction of the transfer effects. When the two languages had similar morphological structure (Deacon et al., 2007) the transfer was bidirectional. By contrast, when the two languages had different morphological structure, (Pasquerella et al., 2001; Wang et al., 2006) the transfer occurred only in one direction, usually from the L2 to the L1. While morphological awareness appears to be an important predictor of reading in the L2 across alphabetic (French and Korean) and non- alphabetic languages (Chinese), to my knowledge, no study to date has explored this association in the context of L3 acquisition.      126 4.2.5  Linguistic factors: Proficiency in the second language  Studies in L3 acquisition have shown a positive association between L1 and L2, on one hand, and L3 acquisition, on the other hand, such that learners can draw upon the skills from their home language and their L2 as resources in developing oral language proficiency and reading skills in the L3 (Brohy, 2001; Haenni Hoti et al., 2011; Sanz, 2008). It is well-documented that under favourable conditions, such as bilingual Spanish and Catalan or Spanish and Basque students acquiring English as an L3, where the L1, the L2, and L3 are supported in school and in society, multilingual students with stronger and more balanced bilingual skills also have stronger oral language proficiency and literacy skills in the L3 (Cenoz & Valencia, 1994; Errasti, 2003; Sanz, 2000, 2008). More recently, Haenni Hoti et al. (2011) was the first study that examined the extent to which oral language proficiency and reading comprehension skills in the L2 predicted the acquisition of the same oral language proficiency and literacy skills in the L3. Associations between the L2 and L3 on the measures of oral language proficiency and literacy were found for all multilingual students, including a sub-section of immigrant students from numerous language backgrounds. In addition, the level of proficiency in the L2 was a much stronger predictor than the skills attained in the L1 in explaining the acquisition of listening comprehension and literacy in the L3; however, in their study, L1 proficiency was assessed only in the bilingual group, and not with the immigrant students. On the basis of the literature reviewed above, it appears that neurophysiological, affective, socio-linguistic, metalinguistic, and linguistic factors influence the acquisition     127 of an L3, with the majority of studies conducted when students’ L1 was supported in school. In line with Hufeisen and Marx’s (2007) factor model the role of several predictors was considered for the acquisition of oral skills and literacy in the L3, French in this study: (1) Age, as a neurophysiological factor, was constant across all of the participants (all participants were first exposed to French as an L3 when they were five years old in Kindergarten; therefore, it was not included as a predictor, (2) as learner- external factors, the study included the families’ SES, the child’s amount of French language use, the child’s amount of reading in French, and the total number of books in the home, (3) as an affective factor, motivation to learn French as an L3 was included, (4) as a metalinguistic factor, morphological awareness in English was included, and (6) as linguistic factors, oral language and reading comprehension in English were included. The large number of home languages present in the sample precluded the examination of the fifth group of factors, namely foreign-language specific factors. It was hypothesized that language proficiency and reading comprehension in English, the L2, would have a positive association with the development of oral language and reading comprehension skills in the L3, in addition to socio-linguistic factors and metalinguistic awareness. The following research question was addressed: What are the learner-external, affective, metalinguistic, and linguistic factors in English, the L2, associated with the acquisition or oral language proficiency and reading comprehension in French, the L3?        128 4.3  Method  4.3.1 Participants  The participants were 55 multilingual students in Grade 6 recruited from schools in culturally and linguistically diverse communities in a major metropolitan city in Canada. The students reported using the following home languages: Amharic, Cantonese, Croatian, German, Greek, Hungarian, Italian, Korean, Mandarin, Polish, Romanian, Spanish, Serbian, and Tagalog. The average age of the participants was 11 years 7 months. The participants in the current study were the same students from study 1. All students were exposed to their L1 from birth, whereas 69% (38 out of 55 students) were also exposed to English within the first year of life. Only those students whose parents reported that their child spoke the L1 at home on a daily basis were included, and this assured that they had generally strong oral skills in the L1. On average, students stated that they currently used their home language on a daily basis in a variety of settings (e.g. home, with grand-parents, or with friends). All students, with the exception of four, were born in Canada. These four students have been living in Canada for over eight years. Students were first exposed to French in Kindergarten at the age of five years. All students were enrolled in an early French immersion program, had been receiving French instruction since Kindergarten, and began receiving formal English instruction in Grade 4.       129 4.3.2 Materials  Learner-external factors  Parents of each participant completed a questionnaire that asked about the SES, the French language use, and the literacy practice in the household (see Appendix A). The questionnaire was adapted from the Alberta Language Environment Questionnaire (Paradis, 2011) and the Parent Interview of Acquiring Literacy in English Study (Duursma et al., 2007). The most common and accepted methods of controlling for SES when examining reading achievement is to control for parental income, occupation, or education level (Jeynes, 2002). The level of parental education was used as a proxy for SES in this study (education attained by parents varied from having graduated from high school to having completed a graduate degree). Parents indicated the percentage of time their child spoke in French on a weekly basis (percentage varied from 10% to 80%); how often their child read independently in French (answers varied from never (score of 1) to almost every day (score of 5); and the total number of books in the two languages (the majority of parents, 96%, reported that they had both English and French books in the home) (see Table 4.1).  Motivation to learn French  Students completed a Likert-type questionnaire in English, adapted and translated from well-established surveys by Gardner (1985) and Sanz (2000), which comprised five     130 choices (totally disagree, disagree, neutral, agree, totally agree) for each of 11 statements. The questionnaire included positive (e.g., ‘I enjoy learning French’) and negative statements (e.g., ‘I never quite feel sure of myself when I am speaking French’). The ‘motivation’ construct was calculated for each participant by adding the scores from the 11 items in the questionnaire. The answer ‘totally agree’ corresponded to five points, whereas the answer ‘totally disagree’ received one point.  Negative statements were reversed for scoring purposes. Raw scores were used in the analyses and the maximum score was 55 points (see Appendix B). This identical procedure was used in other studies conducted with multilingual students (Haenni Hoti et al., 2011; Sanz, 2000).  English morphological awareness  Morphological awareness, as it contributes to reading, reflects the ability to parse words and analyze each morpheme component for the purpose of constructing meaning (Nagy, 2007). The derivational morphology test, an experimental task adapted from the Test of Morphological Structure (Carlisle, 2000) was used to measure morphological awareness, for the reasons discussed above. The derivational morphology task consisted of 27 items and required students to modify a word (the derivative), for example magic, to fit into a sentence that is appropriate in terms of the morphology and meaning, he was a very good ____ [magician] (see Appendix C).  The internal consistency reliability measured with a Chronbach’s alpha for this measure was .62, as reported in the current study. The skeweness value (.59) was within the acceptable range for morphological awareness task.     131 English oral language proficiency and reading skills  Vocabulary knowledge: The standardized Peabody Picture Vocabulary Test-III (PPVT-III, Dunn & Dunn, 1997) required students to identify the picture that correctly corresponded to an orally presented word among four picture alternatives. The publisher- reported test-retest reliability coefficient for this age group was .93 (Dunn and Dunn, 1997). Listening comprehension: In the Listening Comprehension subtest from the standardized Woodcock Language Battery Proficiency (WLPB-R, Woodcock, 1991) the experimenter read aloud short sentences which were each missing a word.  The task required the students to say a word that is appropriate both in terms of structure (syntax) and meaning (semantics) in the context of the sentence. The publisher-reported test-retest reliability coefficient for this age group was .86 (Woodcock, 1991). The Passage Comprehension subtest from the Woodcock Language Proficiency Battery (WLPB-R) (Woodcock, 1991) asked students to read a short passage with a missing word that required them to produce a word that would be appropriate both in terms of structure (syntax) and meaning (semantics) in the context of the paragraph. The test-retest reliability coefficient of the WLPB was .90 for this age group (Woodcock, 1991).         132 French oral language proficiency and reading skills  Vocabulary knowledge: The Échelle de vocabulaire en images Peabody (ÉVIP, Dunn, Thériault-Whalen, & Dunn, 1993), a standardized test of French vocabulary knowledge was used, which is equivalent to the PPVT-III. The publisher-reported test- retest reliability coefficient for this age group was .90 (Dunn et al., 1993). Listening comprehension: The Compréhension orale subtest from the Test de rendement individual de Wechsler (WIAT-II, Wechsler, 2007) is similar to the English WLPB-R, although it also uses a picture elicitation task and includes items that measure vocabulary knowledge and production. The publisher-reported test-retest reliability coefficient was .88 for this age group (Wechsler, 2007). In the Compréhension de lecture [reading comprehension] subtest from the WIAT-II (Wechsler, 2007) participants were asked to read short stories (aloud or silently) and answer questions relating to the stories. The publisher-reported test-retest coefficient was .94 for the age group (Wechsler, 2007).  4.3.3  Procedure  Each participant was tested individually in two sessions (one in English and one in French) that lasted approximately 45 minutes each and were separated by 1 to 2 weeks. Native English- and French- speaking research assistants who were trained for the purposes of the study administered all tests in the two languages. In addition, to     133 counterbalance for order effects, one half of the participants were tested first in English, and the other half were tested first in French.  4.4  Results  The primary purpose of this study was to investigate the association between English oral proficiency and reading comprehension, in addition to socio-linguistic and metalinguistic factors, and the French oral language proficiency and reading comprehension skills of multilingual students in an education setting that does not teach in the students’ L1. To achieve this goal, two major hierarchical regression analyses were completed: (1) regression analyses to identify the socio-linguistic factors related to French oral language proficiency and reading comprehension skills and (2) regression analyses to identify the English metalinguistic factors and linguistic factors, over and above the socio-linguistic factors, associated with French oral language proficiency and reading comprehension skills. Residual analyses were first carried out to identify possible outliers, deviations from the normal distribution, multicollinearity, and variance heterogeneity. First, several composite scores were created where appropriate. The degrees of education between the two parents were significantly correlated (r = .71, p < .001) and there was no difference between their level of education, t(51) = .08, p = .03, d = .001; therefore, a single composite score of ‘parental education’ was created by averaging the number of years of education completed by mothers and fathers. Student’s performance on the English tests of listening comprehension and vocabulary knowledge were     134 significantly correlated (r = .67, p < .001) (see Table 4.2) and there was no difference between the two tasks, t(54) = .99, p = .33, d = .02; therefore, a single composite variable for English oral proficiency was created by averaging the standard scores from the two tasks. Similarly, the scores on the French tests of listening comprehension and vocabulary knowledge were significantly correlated (r = .72, p < .001) and there was no difference between the two tasks, t(54) = .54, p = .58, d = .005; therefore, a single composite variable for French oral proficiency was created by averaging the scores from the two tasks.     135 Table 4.1 Mean scores (standard deviations) for all variables in English and French  Mean scores (SD) Minimum Maximum Length of residency (in years) 10.69 (.78) 7.0 11.0 Socio-linguistic variables    Socioeconomic status    Language use in French    Amount of reading in French    Number of books    Motivation to learn French  6.96 (1.61) 39.60% (10.86%) 3.87 (1.03) 4.75 (1.12) 39.44 (4.40)  3 10% 2 1 29  9 90% 5 6 50 Morphological awareness .77 (.15) .31 1.00 Proficiency in English (L2)    Oral language proficiency    Reading comprehension  115.94 (16.40) 110.57 (11.61)  83.50 91.00  159.50 143.00 Proficiency in French (L3)    Oral language proficiency    Reading comprehension  94.38 (13.81) 100.52 (15.66)  70.50 76.00  131.50 160.00  Note: A score of ‘6’ for socioeconomic status indicates that parents have achieved at least some post-secondary education; and a sore of ‘7’ indicates that parents have completed a bachelor degree. A score of ‘3’ for amount a child reads in French indicates that the child reads in French one or two days per week; and a score of ‘4’ indicates that the child reads three to five days a week. For French language use, parents estimated the amount of time their child spoke French on a weekly basis. For morphological awareness,     136 scores are represented by percentages. For English (L2) proficiency and French (L3) proficiency, scores are represented by standard scores     137 An examination of the correlations between the socio-linguistic factors and oral proficiency and reading comprehension scores (standard scores) in French revealed that SES and amount of reading in French significantly correlated with French oral proficiency and French reading comprehension (Pearson r’s varied between .39 and .53, p < .05). Moreover, the number of books in the home was significantly associated with oral proficiency in French (r = .29, p < .05), but not with reading comprehension (r = .21, p > .05). Moreover, morphological awareness in English, oral language proficiency in English, and reading comprehension in English were significantly correlated with oral language proficiency in French (Pearson correlation coefficients varied between r = .61 and .74, p < .001), and with reading comprehension in French (Pearson correlation coefficients varied between r = .55 and .69, p < .001) (see Table 4.2 for the correlations). Next, two sets of regression analyses were conducted on oral proficiency in French and reading comprehension in French separately (see Table 4.3). Both regression models, with all five socio-linguistic factors included, were significant and explained 35% of the total variance in oral language proficiency in French, F(5, 50) = 4.48, p = .001, and 33% of the total variance in reading comprehension in French, F(5, 50) = 4.17, p = .001, respectively. Amount of reading in French was the strongest, and the only statistically significant socio-linguistic predictor of oral language proficiency in French; whereas, amount of reading in French and SES were significant predictors of reading comprehension in French, and only these factors were included in the subsequent analyses.     138 Table 4.2 Correlations between the socio-linguistic factors, morphological awareness, oral proficiency, and reading comprehension scores in English and French at the end of Grade 6 (N = 55)   1 2 3 4 5 6 7 8 9 10 11  1. French motivation scale - 2. Time spend in English .12 - 3. Time spend in French .21 .21 - 4. Parental SES .07 .21 .12 - 5. Frequency child reads English  .28 .07 .17 .46** - 6. Frequency Child Reads French .03 .16 .14 .28 .26 - 7. Number of Books .04 .06 .13 .51** .68** .24 - 8. Morphological Awareness .03 .17 .05 .09 .01 .37** .06 - 9. English Oral Language .01 .11 .11 .42** .29* .36** .33** .56** - 10. French Oral Language .04 .10 .27 .39** .26 .53** .29* .51** .74** - 11. English reading Comp .19 .07 .01 .47** .31* .36** .28* .39** .61** .67** - 12. French reading Comp .19 .13 .09 .44** .27 .53** .21 .52** .65** .76** .70**     139 Table 4.3 Multiple regression models including all socio-linguistic variables predicting oral language proficiency in French and reading comprehension in French  Socio-linguistic factors           Oral language proficiency in French                         Reading comprehension in French B SE t B SE t SES 1.31 1.27 1.03 3.18 1.48 2.14* French language use .48 .93 .52 -.09 1.08 .08 Amount of reading in French 6.69 1.83 3.66** 7.03 2.14 3.28** Number of books 1.52 1.76 .66 .84 2.06 .41 Motivation to learn French .10 .40 .25 .07 .46 ..15 R 2   .35**  .33** F                            4.48                             4.17 Note: *p < .05, **p < .01       140 In order to test the hypothesis that oral language proficiency and reading comprehension in English would have a facilitative effect on the acquisition of oral language proficiency and reading, over and above socio-linguistic factors and metalinguistic awareness, control variables (amount of reading in French and English morphological awareness) were first entered into regression analyses, then the main predictors (standard scores of English oral language proficiency and reading comprehension) were entered into the model. Therefore, the predictors were entered in the following order: in step 1, amount of reading in French; in step 2, morphological awareness in English; in step 3, oral language proficiency in English; and in step 4, reading comprehension in English, controlling for all other variables in the model. The results are presented in Table 4.4.     141 Table 4.4 Summary of multiple regression analyses for variables predicting oral language proficiency in French  B t R 2  R 2 change F change Model 1 Step 1    Amount of reading in French  7.36   4.43**   .28**  Step 2    Amount of reading in French    Morphological awareness in English   4.08 3.41  2.39* 3.80**    .44**    .16**    14.46** Step 3    Amount of reading in French    Morphological awareness in English    Oral language proficiency in English   3.65 .75 .48   2.62* 1.04 5.06**      .63**     .19**     25.65** Step 4    Amount of reading in French    Oral language proficiency in English    Reading comprehension in English  3.50 .40 .37  2.98* 4.59** 3.12**     .69**     .07**     9.12**  Note: *p < .05, **p < .01     142 The regression model that included amount of reading in French was significant, F(1, 54) = 19.00, p = .001, and accounted for 28% of the variance. Next, when morphological awareness in English was added to the model, this factor contributed an additional 16%, and the overall regression effect was significant, F(2, 53) = 14.46, p = .001, and accounted for 44% of the variance. The inclusion of English oral language proficiency in step 3 explained an additional 18% of the variance, and together with amount of reading in French and morphological awareness, it explained 62% of the variance in oral language proficiency in French, F(3, 52) = 25.65, p < .001. However, in step 3, morphological awareness was no longer a significant predictor of French oral proficiency and it was removed from the model. The amount of reading in French and oral language proficiency in English explained 63% of the variance in French oral language proficiency, F(2, 53) = 46.20, p < .001. Finally, in step 4, reading comprehension in English explained an additional 6% of the variance, and together with amount of reading in French and oral language proficiency in English, it explained 69% of the variance, F(3, 52) = 9.12, p = .01. Thus, French oral language proficiency was best explained by amount of reading in French, oral language proficiency in English, and reading comprehension in English. The same procedure, as described above, was used to investigate and isolate the effect of each of the morphological awareness, oral language proficiency, and reading comprehension factors in English on reading comprehension in French, over and above socio-linguistic factors, and is presented in Table 4.5.     143 Table 4.5 Summary of multiple regression analyses for variables predicting reading comprehension in French  B t R 2  R 2 change F change Model 1 Step 1    SES    Amount of reading in French  3.11 6.84  2.60* 3.52**    .35**  Step 2    SES    Amount of reading in French    Morphological awareness in English   2.73 4.21 7.71  2.44* 2.02* 2.86*     .45**     .10**     8.20** Step 3    SES    Amount of reading in French    Morphological awareness in English    Oral language proficiency in English   1.59 4.21 .69 .39   1.45 2.22* .89 3.01*       .54**      .10**      9.07** Step 4    Amount of reading in French    Oral language proficiency in English    Reading comprehension in English   4.21 .30 .58   2.80* 2.77* 3.80**     .64**     .10**     14.47**  Note: *p < .05, **p < .01     144 The regression model that included the socio-linguistic factors SES and amount of reading in French was significant, F(2, 53) = 13.09, p = .001, and accounted for 35% of the variance in French reading comprehension. Next, when morphological awareness in English was added to the model, it explained an additional 10% of the variance, and the overall regression effect was significant, F(3, 52) = 8.20, p = .006. The addition of oral language proficiency in English in step 3 significantly improved the overall regression model (added 10% of variance) after controlling for amount of reading in French and morphological awareness, and together explained 54% of the variance in reading comprehension in French, F(4, 51) = 9.07, p < .001. However, in step 3, SES and morphological awareness were no longer significant predictors and were removed from the model. Next, amount of reading in French and oral language proficiency in English were entered into the regression equation and these two predictors explained 52% of the variance in French oral language proficiency, F(2, 52) = 25.34, p < .001. Finally, in step 4, reading comprehension in English together with amount of reading in French and oral language proficiency in English, F(3, 52) = 14.47, p < .01 explained 64% of the variance. Therefore, amount of reading in French, oral language proficiency in English, and reading comprehension in English were the best predictors of French reading comprehension.  4.5  Discussion  The main goal of this study was to examine the association between oral language proficiency and reading comprehension in the L2, English, and the oral language     145 proficiency and reading comprehension in the L3, French, in addition to socio-linguistic factors, in multilingual students studying in an education setting that did not support the L1. To better understand the effect of each of these variables, I used Hufeisen and Marx’s (2007) factor model as a framework for the development of oral language proficiency and reading comprehension skills in the L3. The most important contribution of this study comes from the evidence that oral language proficiency and reading comprehension skills in the L2 are both important factors in the acquisition of oral language proficiency and reading comprehension in the L3, even after controlling for socio-linguistic factors and metalinguistic awareness. This finding is consistent with Haenni Hoti et al.’s (2011) work in which they found a significant association between oral language proficiency and reading comprehension skill in the L2 (French) and the same oral language and literacy skills in the L3 (English). Moreover, the findings are in line with Cummins’ interdependence hypothesis according to which language competence in one language is related to language proficiency in the other. Similarly to the research by Cenoz and Valencia (1994), showing that bilingual Spanish-Basque students obtained high levels of listening comprehension and reading comprehension in English (L3) in environments that supported the L1, the current study revealed that oral language proficiency and reading skills in English are similarly important sources of language acquisition in French, even in bilingual school contexts that do not teach in the students’ L1. Moreover, the findings from this study are in line with Stanovich (1986), who postulated that the relationship between vocabulary and reading comprehension is a reciprocal one, where vocabulary growth takes place through reading, and that students     146 with larger vocabularies tend to better comprehend written text. Indeed, the results extended Stanovich’s hypothesis to the context of L2 and L3 acquisition, and revealed a reciprocal association between the L2 and L3 in multilingual students, where students with higher L2 oral language proficiency (which includes vocabulary knowledge) have stronger L3 reading comprehension (and L3 oral language proficiency), and similarly, students who demonstrate greater L2 reading comprehension have stronger L3 oral language proficiency (and L3 reading comprehension). According to Hufeisen and Marx’s (2007) factor model, acquisition of an L3 is dependent on a number of learner-external, affective, metalinguistic, and linguistic factors which may exert a varying amount of influence. Similarly to Aarts and Verhoeven, (1999), whose research was conducted with immigrant students in the Netherlands, this current study found that amount of reading in the L3 (French) was the only socio-linguistic factor that was important in determining oral language proficiency and reading comprehension skills in the L3. The number of books at home, by contrast, was not important in determining reading comprehension skills. If bilingual and multilingual students show important gains in acquisition of reading comprehension in the L2 and L3, the studies clearly point out that there is at least one variable that is important to the growth of reading comprehension in the L3: reported amount of reading in the L3. Findings from the current study revealed that SES was not significantly associated with French oral language proficiency. Moreover, SES was not significantly related with French reading comprehension. Parental education has been shown to be a strong predictor of language proficiency, especially in studies conducted with samples where a wide range of parental education is present (Gunderson, 2007; Gunderson,     147 D’Silva, Murphy, & Odo, 2012). By contrast, in studies with samples where the average parental education is a university degree, as observed in the current study, it was more difficult to find an effect for SES on language proficiency (Cenoz & Valencia, 1994; Sanz, 2008; Swain, Lapkin, Rowen, & Hart, 1990). Hufeisen and Marx’s factor model has also identified the importance of motivation in the acquisition of L3. Surprisingly, motivation was not a significant predictor of L3 acquisition, as would have been expected based on the previous studies conducted with students acquiring an L3 (Cenoz & Valencia, 1994; Haenni Hoti et al., 2011; Sanz, 2000, 2008). Overall, the students’ motivation to learn French was average. The current study involved multilingual students from culturally and linguistically diverse communities where English is the dominant language outside the school. French is limited to the classroom and is not necessary for everyday interactions with parents and friends. Therefore, it is possible that the students’ degree of motivation (measured with a student questionnaire adapted from Gardner’s Socio-Educational Model) may have been overshadowed by their limited amount of exposure to French at home and in the community, which is much less than in other urban areas, such as Montreal, where students are continually immersed in French, both in school and in their communities. Another possible explanation for the lack of a strong association between motivation to learn French and reading comprehension in French is the lack of variation in the scores obtained on the measure of motivation. There was little variance observed on the measure of French motivation, and the degree of motivation was generally average. Moreover it is possible that the measure was not suitable for the sample of students, who came from a different socio-linguistic and demographic areas that the students in the previous studies     148 (Cenoz & Valencia; 1994; Sanz, 2000). Future studies should examine whether the degree of motivation to learn French differs in a variety of socio-linguistic settings..   In line with research on the cross-language associations between L1 morphological awareness and L2 reading comprehension when languages share a similar morphological structure (Deacon et al., 2007), the findings from this study showed that this association extends to the acquisition of reading comprehension in the L3. English (L2) morphological awareness was a significant predictor of French (L3) reading comprehension in Grade 6, after controlling for learner-external factors, suggesting that the benefits of being able to segment words into their morphological units in English is also important when learning to read in French. The effect of morphological awareness disappeared once English oral language proficiency was entered into the model, however. The English oral language proficiency was more strongly associated with French reading comprehension (stronger correlation), and is thus masking the association between English morphological awareness and French reading comprehension. This may be a result of the fact that the multilingual students in this sample were still developing their oral language proficiency in English, the L2; therefore, oral skills in L2 were more indicative of the ability to read in the L3. Another possible explanation is that the morphological awareness task may not have been sensitive enough for the students in the current study. Previous studies have shown that tests of derivational morphological awareness using items with phonological shifts (where the pronunciation of a stem is phonologically altered, such as strong/strength) may be more strongly associated with reading comprehension than those that include items that are phonologically transparent (where the pronunciation of a stem does not change by the addition of a morpheme, such     149 as assist/assistance), especially with students who are mastering their phonological skills between Grade 3 and Grade 5 (Carlisle, Stone, & Katz, 2001). The derivational morphology task in the current study comprised mostly of items that were phonologically transparent; thus it is possible that the task was not sensitive enough for the Grade 4 and Grade 6 students in the sample. Future studies should examine associations between morphological awareness in the L3 and reading comprehension in the L2 and ensure that the morphological awareness task is better suited (for example, more of the items should involve phonological shifts) for older students in Grade 6 in to order to confirm that transfer is bidirectional. As more and more multilingual students learn English and French in Canada, it will be necessary to investigate their oral language proficiency and reading success. Findings from this study have important educational implications for developing the oral language proficiency and reading skills in French for multilingual students. First, while educators in early FI programs focus their instruction mostly in French, multilingual students continue to develop strong oral language proficiency and reading skills in English. Although Grade 6 students receive only 30% of their education in English, their vocabulary knowledge, listening comprehension, and reading comprehension skills in English are strongly associated with oral language proficiency and reading comprehension skills in French. In particular, French and English share many vocabulary cognates – words that have similar spelling and meaning in the two languages – and teachers could focus on identifying the similarities between the two languages (e.g. education and éducation) which leads to increased reading comprehension in French (Meara, 1993). Secondly, teachers in FI programs may need to help children become     150 more aware of morphology for supporting French (L3) reading comprehension. The derivational morphological structure in English and French share many similarities; incorporating lessons that teach students the common derivational morphemes in the two languages could facilitate the acquisition of French reading comprehension. Finally, the school system should encourage students to read in French which could produce more fully biliterate students with stronger oral language proficiency and reading skills. Given the growing linguistic heterogeneity of Canadian classrooms, it is important to continue to monitor the French reading skills of multilingual students enrolled in FI programs.      151 Chapter 5: Conclusion  Given that the driving question behind this thesis is whether early French immersion programs meet the needs of students in Grades 4 and 6 in a major city in Western Canada, including increasing numbers of multilingual learners, the conclusion is a decisive YES. To date, there has been a significant gap in our understanding of literacy achievement in multilingual students, particularly due to a lack of longitudinal research on the reading outcomes of multilingual students in early French immersion (FI) programs in Canada. In addition, researchers have called for a need to examine multilingual students’ home language literacy and how this influences the acquisition of reading skills in their L2 and L3 (Jared & Genesee, 2008; Swain & Lapkin, 2005). The understanding is that this would build on previous knowledge in L2 acquisition in students attaining a second language at school. Chapter 2 in this thesis consisted of a longitudinal study that examined the relative success of multilingual and bilingual students enrolled in early FI programs. It was specifically designed to investigate the oral language proficiency and reading skills, in English and French, of multilingual students in Grade 4 and Grade 6. The participants in the study included a group of multilingual students who were exposed to a language other than English or French at home and were developing English (L2) and French (L3) skills in school. Their scores were compared with those of a group of Anglophone speakers who were learning French as a second language in school. The exact same tasks, under the exact same conditions, were presented to all students at the beginning of Grade 4, at     152 the end of Grade 4, and at the end of Grade 6. The tasks used in the study were measures of oral language proficiency across two major domains of language including vocabulary knowledge and listening comprehension. Reading skills were also assessed using measures of word reading and reading comprehension. Findings revealed that multilingual students had the same level of oral language proficiency in French and the same level of reading comprehension skills in English and in French as bilingual students. However, multilingual students had relatively lower vocabulary knowledge and listening comprehension skills in English compared with the bilingual group. It is possible that the multilingual students had reached a threshold in oral language proficiency in English and French to exhibit strong reading comprehension skills in English (Cummins, 2000). According to the threshold hypothesis, when students achieve a sufficient level of language proficiency (a threshold) in their L1 and L2, they can exhibit strong reading comprehension skills in the L2 (Cummins, 2000). However, to date the language threshold has not been clearly defined (Bernhardt & Kamil, 1995; Genesee et al., 2008). In other words, it is not clear what the exact levels of language proficiency in the L1 and L2 is required for the acquisition of strong reading comprehension skills in the L2 to emerge, and thus we can only speculate about its effect on reading. The results also revealed that the oral language factors influencing reading comprehension were generally similar between multilingual and bilingual students. In other words, the vocabulary knowledge and the listening comprehension skills of all students in FI programs were equally important to the acquisition of reading comprehension in the immersion language, regardless of whether French was the second     153 language or the third language. These findings are supported in part by the simple view of reading model (Hoover & Gough, 1990; Poctor et al., 2006). Vocabulary knowledge and listening comprehension in French (L3) were significant predictors of reading comprehension in English (L2). Transfer did not occur between oral language proficiency in English and reading comprehension in French. These findings provide further support that the language of instruction (amount of exposure) might influence the direction of transfer between the oral language proficiency and the reading skills in the L2 and the L3 (Bérubé & Marinova-Todd, 2012; Miller et al., 2006). In FI classrooms such as those described in chapter 2, where teaching practices were aimed at content-based learning of oral language proficiency (vocabulary knowledge and listening comprehension) in French and English, the findings reported in the current study provide additional support that the teaching paradigm is working. Teachers are encouraged to continue to foster oral language skills and reading skills in both languages. The third chapter of this dissertation was designed to better understand the reasons why multilingual students developed lower oral language proficiency in English than the bilingual students. Researchers have identified that L1 typology (specifically the writing systems) influences the acquisition of oral language proficiency and reading skills in the L2 of bilingual students (Koda, 1989) and in the L3 of multilingual students (Swain, Lapkin, Rowen, & Hart, 1990). Chapter 3 investigated the role of language typology (e.g. phonological and writing systems) in the first language of multilingual students along with the development of oral language proficiency and literacy skills in English and French in FI programs. In this study, only multilingual students who were literate in their L1 were included. Students were given a background questionnaire asking     154 them to report on their home language and literacy experience. Based on the responses from participants, two groups of multilingual students were identified: one group was literate in an alphabetic first language (e.g. Spanish) and the other group was literate in a logographic/syllabary first language (e.g. Chinese). The performance of the two multilingual groups on tasks that measured vocabulary knowledge, listening comprehension, word reading, pseudoword reading, and reading comprehension was compared with skills of a control group of Anglophone-speaking students who were learning French in FI programs. The following main findings were discussed: 1) first language typology had a relatively strong influence at least in Grade 4 on the acquisition of oral language proficiency and reading comprehension in the L2 and L3, 2) first language typology did not influence the acquisition of decoding skills of multilingual children who were exposed to immersion languages for over three years, and 3) first language typology influenced the association between the vocabulary knowledge and the reading comprehension skills in the L3. For example, multilingual students with a logographic/syllabary L1 who were learning an alphabetic L2 and L3 initially obtained lower scores on measures of oral language proficiency and reading comprehension. When the same group of students was tested again at the end of this study, in Grade 6, however, there was no difference in their scores compared with those of other students. These preliminary data suggested that students whose home language is orthographically different than the immersion language may take longer to develop L2 and L3 literacy than students whose home language is orthographically similar to the immersion language. Moreover, these findings suggest that multilingual students may process     155 differently French and English reading, depending on how closely their L1 writing system corresponds to the orthography of the L2 and L3. The conclusions drawn from these results need to be interpreted with caution given the relatively small number of participants included in this study. In addition to language typology, other socio-linguistic factors and meta-linguistic factors have been found to contribute to the development of reading skills in the L3 of multilingual students (Hufeisen & Marx, 2007; Sanz, 2008). Chapter 4 of this thesis was an empirical study of the factors that best predicted oral language proficiency and reading comprehension skills in the L3 (French) of multilingual students in FI programs. The study focused exclusively on multilingual students in Grade 6 because information on meta-linguistic awareness and motivation to learn an L3 was not collected in Grade 4. All multilingual students in Grade 6, including those that were not literate in the L1, participated in the study. Four categories of predictors were identified: 1) learner-external factors (e.g. socio-economic status, French language use, French literacy practice, the total number of books in the home), 2) affective factors (e.g. degree of motivation to learn French), 3) metalinguistic factors (e.g. morphological awareness), and 4) linguistic factors (e.g. vocabulary knowledge, listening comprehension, and reading comprehension in English). The study first explored the correlations between each of the factors and each of the two outcome variables (i.e. French oral language proficiency and French reading comprehension). Next, results from hierarchical regression models were considered. The findings from chapter 4 were in line with Hufeisen and Marx’s factor model and resulted in three important conclusions: 1) the amount of time multilingual students were reported to read in French was an important factor in determining the growth of oral language     156 proficiency and reading comprehension skills in French, 2) morphological awareness skills in English were strongly associated to reading comprehension skills in French, and 3) oral language proficiency and reading comprehension skills in English were the strongest predictors of oral language proficiency and reading comprehension skills in French, even in school contexts that do not support the student’s L1. Beyond these theoretical research findings, there are potential educational implications in the broader field of third language acquisition. Findings lead us to understand that the oral language proficiency acquired in an immersion language (in this case, French) play an important role in supporting the reading comprehension skills in both the L2 (in this case, English) and the L3 (French). In short, FI programs are working. French immersion teachers (in middle grades) are encouraged to continue to support French vocabulary knowledge and French listening comprehension of all students in order to help foster later reading ability. Moreover, educators are encouraged to continue to foster the reading skills in the immersion language (e.g. French) which could produce more fully biliterate students with stronger oral language proficiency and reading skills. Findings in this dissertation further revealed that the vocabulary knowledge and listening comprehension skills in French and English were strongly correlated. In particular, French and English share many cognates – words that have similar spelling and meaning in the two languages – and teachers are encouraged to focus on identifying the similarities between the two languages (e.g. education and éducation) which could help multilingual students increase their English oral language proficiency. In addition to possible future research directions identified in the three parts of this thesis, this longitudinal study also makes way for two further major studies. First,     157 regarding the gap in literacy outcomes identified in multilingual students whose home language is orthographically different than the immersion language, the precise nature of why this is occurring needs to be further examined. If researchers can identify other factors that best contribute to literacy success in the L2 and L3 of these students, teachers and educators may develop more appropriate instruction. In the meantime, teachers and educators should continue to foster and support vocabulary knowledge, morphological awareness skills, and listening skills in the immersion language, which in turn could increase multilingual students’ English oral language proficiency. Second, for multilingual students who may be at risk of language difficulties beyond the typical stages of language acquisition, future studies could examine the oral language proficiency and reading skills of multilingual students in early elementary grades such as Kindergarten to Grade 2. This will help teachers to identify students who may need additional support before they reach Grade 4. The present dissertation has several limitations. First, the longitudinal analyses were based on multilingual students from culturally and linguistically diverse communities in a large Western Canadian city, in which English was the dominant language outside school. An important issue is whether the oral language proficiency and reading skills in English and French of multilingual students would develop in the same fashion in other parts of Canada, such as Montreal, where students are continually immersed in French, both in school and in the community. Since amount of exposure plays an important role in acquisition of an immersion language, it would be important to examine whether the development of French and English of multilingual students differs in a variety of language settings, such as for multilingual students who are enrolled in     158 English immersion programs in Québec, learning English as an L3. Until additional research is conducted across a variety of language settings, the findings from the current study are limited to multilingual students who are exposed primarily to English outside of school. Measures of listening comprehension and vocabulary knowledge were used as the two measures of oral language proficiency in English and French. It is not possible to comment about the role of other measures of oral language proficiency in the L2 or L3, such as grammatical knowledge and the use of language structures in oral narratives, on the development of reading comprehension in the L3. These other measures of oral language proficiency have been shown to be important factors in the development of L2 reading comprehension in bilingual students (Manis et al., 2004; Miller et al., 2006). Recent theoretical perspectives on language transfer have identified that in addition to oral language proficiency, other factors may have an important influence in determining the extent of language transfer (Genesee et al., 2008). The current study has identified that learner-external factors (e.g. French literacy practice) and metalinguistic factors (e.g. morphological awareness) affect the amount of transfer from the L2 to the L3. Other factors, such as underlying cognitive processes (working memory and speech of processing) may influence it as well (Genesee et al., 2008). Future research will need to examine the role of multiple aspects of oral language proficiency, in addition to the role of underlying cognitive processes on the acquisition of reading comprehension in the L3. Due to the diversity of L1s represented in the sample (representative of the L1s in the school demographics in this Western Canadian city), it was not possible to formally measure L1 oral language proficiency and reading skills in all the languages. As an initial     159 study, information was gathered on vocabulary knowledge and reading comprehension of the Chinese-speakers in the L1, L2 and L3. Not all Chinese-speaking students completed the Chinese reading comprehension test. There are many possible reasons why the Chinese-speaking students chose not to complete testing in their L1. Anecdotally, two students reported that they felt ‘embarrassed’ by their ability to read in Chinese. Ideally, future studies should aim to include measures of word reading, reading comprehension, and discourse writing in the L1 to get a better sense of their literacy abilities (Gunderson, 2007). Since the inception of FI programs in Canada, the goal has been to develop excellent literacy skills in Canada’s two official languages. As more people immigrate to Canada and choose to educate their children in both of Canada’s official languages, it is necessary to investigate their reading success. Through this dissertation, we have gained a clearer understanding of multilingual students’ L2 and L3 proficiency, which can play a role in convincing parents, teachers, and school administrators that multilingual students can thrive in FI programs and become successful learners of English and French.     160 References  Aarts, R., and L. Verhoeven. (1999). Literacy attainment in a second language submersion context.  Applied Psycholinguistics, 20, 377-393. Abu-Rabia, S. (2007). The role of morphology and short vowelization in reading Arabic among normal and dyslexic readers in grades 3, 6, 9, and 12. Journal of Psycholinguistic Research, 36, 89-106. Adams, M. J. (1990). Beginning to Read: Thinking and Learning about Print. Cambridge, MA: MIT Press. Akamatsu, N. (1999). The effects of first language orthographic features on word recognition processing in English as a second language. Reading and Writing: An Interdisciplinary Journal, 11, 381-403. August, D., Carlo, M., Dressler, C., & Snow, C. (2005). The critical role of vocabulary development for English language learners. Learning Disabilities Research & Practice, 20(1), 50-57. August, D., Genesee, F., Geva, E., Dressler, C., & Kamil, M. L. (2008). Cross-linguistic relationship in second language learners. In D. August & T. Shanahan. (Eds.), Developing Reading and writing in second-language learners: Lessons from the report of National Literacy Panel on language-minority children and youth (pp. 61-94). New York, NY: Routledge. Baker, C. (2011). Foundations of bilingual education and bilingualism (5 th  edition). Tonawanda, NY: Multilingual Matters.     161 Bérubé, D., & Marinova-Todd, S. H. (2012). The development of language and reading skills in the second and third languages of multilingual children in French immersion. International Journal of Multilingualism, 9(3), 272-293. Bialystok, E. (2007). Cognitive effects of bilingualism: How linguistic experience leads to cognitive change. International Journal of Bilingual Education and Bilingualism 10(3), 210-223. Bialystok, E., Luk, G., & Kwan, E. (2005). Bilingualism, biliteracy, and learning to read: Interactions among languages and writing systems. Scientific studies of reading, 9(1), 43-61. Bild, E. R., Swain, M. (1989). Minority language students in a French immersion program. Journal of Multilingual and Multicultural Development, 10, 255-174. Bos, W., & Pietsch, M. (2006). KESS 4 Kompetenzen und Einstellungen von Schulerinnen und Schulern am Ende der Jahrgangsstufe 4 in Hamburger Grundschulen. Hamburger Schriften zur Qualitatim Bildungswesen, Band 1 [KESS 4 competencies and attitudes of students at the end of 4th grade in primary schools in Hamburg. Hamburg studies on quality in the educational system (vol. 1)]. Munster: Waxmann. Bournot-Trites, M., & Séror, J. (2003). Students’ and teachers’ perceptions about strategies which promote proficiency in second language writing. Canadian Journal of Applied Linguistics, 6(2), 129-157.     162 Brohy, C. (2001). Generic and/or specific advantages of bilingualism in a dynamic plurilingual situation: The case of French as official L3 in the school of Samedan (Switzerland). International Journal of Bilingual Education and Bilingualism, 4(1), 38-49. Carlisle, J. F. (2000). Awareness of the structure and meaning of morphologically complex words: Impact on reading. Reading and Writing, 12, 169-190. Carlisle, J. F., Beeman, M., Davis, L. H., & Spharim, G. (1999). Relationship of metalinguistic capabilities and reading achievement for children who are becoming bilingual. Applied Psycholinguistics, 20, 459–478. Carlisle, J., Stone, C. A., & Katz, L. (2001). The effects of phonological transparency on reading derived words. Annals of Dyslexia, 51, 249–274. Carlo, M., Angus, D., McLaughlin, B., Snow, C., Dressler, C., Lippman, D., et al. (2004). Closing the gap: Addressing the vocabulary needs of English-language learners in bilingual and mainstream classrooms. Reading Research Quarterly, 39, 188- 215. Casalis, S., and M.-F. Louis-Alexandre. (2000). Morphological analysis, phonological analysis and learning to read French: A longitudinal study. Reading and Writing: An Interdisciplinary Journal, 12, 303–335. Cenoz, J. (2003). The effect of linguistic distance, L2 status and age on cross-linguistic influence in third language acquisition. In J. Cenoz, B. Hufeisen, & U. Jessner (Eds.), Cross-linguistic influence on third language acquisition: Psycholinguistic perspectives (pp. 8-20). Clevedon, UK: Multilingual Matters.     163 Cenoz, J., & Valencia, J. F. (1994). Additive trilingualism: Evidence from the Basque Country. Applied Psycholinguistics, 15, 195-207. Chen, Xi., Ramirez, G., Yang, L. C., Geva, E., & Ku, Y. – M. (2012). Comparing vocabulary development in Spanish- and Chinese-speaking ELLs: The effects of metalinguistic and sociocultural factors. Reading and Writing, 25(8), 1991-2020. Council of ministers of education, Canada (2009). Key factors to support literacy success in school-aged population: A literature review. http://www.cmec.ca/Publications/ Lists/Publications/Attachments/201/key- factors-literacy-school-aged.pdf Cummins, J. (1976). The influence of bilingualism on cognitive growth: A synthesis of research findings and explanatory hypotheses. Working Papers on Bilingualism, 9, 1-43. Cummins, J. (1979). Linguistic interdependence and the educational development of bilingual children. Review of Educational Research, 49, 222-251. Cummins, J. (2000). Language, Power, and Pedagogy. Toronto: Multilingualism Matters. D’Angiuilli, A., Siegel, L. S., & Maggi, S (2004). Literacy instruction, SES, and word- reading achievement in English-language learners and children with English as a first language: A longitudinal study. Learning Disabilities Research & Practice, 19(4), 202-213.     164  De Angelis, G., & Selinker, L. (2003). Interlanguage transfer and competing linguistic systems in the multilingual mind. In J. Cenoz, B. Hufeisen, & U. Jessner (Eds.), Cross-linguistic influence on third language acquisition: Psycholinguistic perspectives (pp. 42-58). Clevedon, UK: Multilingual Matters. Deacon, S. H., and J. Kirby. (2004). Morphological awareness: Just “more phonological”? The roles of morphological and phonological awareness in reading development. Applied Psycholinguistics 25: 223–238. Deacon, H., L. Wade-Woolley, and J. Kirby. (2007). Crossover: The role of morphological awareness in French immersion children’s reading. Developmental Psychology 43(3), 732-746. Dressler, C., & Kamil, M. (2006). First- and second-language literacy. In D. August & T. Shanahan (Eds.), Developing literacy in second-language learners: Report of the National panel of language-minority children and youth (pp. 197-246). Mahwah, NJ: Erlbaum. Droop, M., & Verhoeven, L. (2002). Language proficiency and reading ability in first- and second-language learners. Reading Research Quarterly, 38, 78-103. Dufva, M., & Voeten, M. J. M. (1999). Native language literacy and phonological memory as prerequisites for learning English as a foreign language. Applied Psycholinguistics, 20(3), 329-348. Dunn, L. M., & Dunn, M. (1997). Peabody Picture Vocabulary Test-Third Edition. Circle Pines, MN: American Guidance Service.  165  Dunn, L.M., Thériault-Whalen, C.M., & Dunn, L.M. (1993). Échelle de vocabulaire en image Peabody. Toronto, Ontario: Psycan. Durgunoglu, A. Y., Nagy, W. E., & Hacin-Bhatt, B. J. (1993). Cross-language transfer of phonological awareness. Journal of Educational Psychology, 85(3), 453-465. Duskova, L. (1984). Similarity – an aid or hindrance in foreign language learning? Folia Linguistica, 18, 103-115. Duursma, E. D., Romero-Contreras, S., Szuber, A., Proctor, P., Snow, C., August, D.,… Calderon, M. (2007). The role of home literacy and language environment on bilinguals’ English and Spanish vocabulary development. Applied Psycholinguistics, 28, 171-190. Errasti, M. P. S. (2003). Acquiring writing skills in a third language: The positive effects of bilingualism. International Journal of Bilingualism, 7(1), 27-42. European Commission. 2008. Multilingualism: an Asset for Europe and a Shared Commitment. Communication from the Commission to the Council, the European Parliament, the Economic and Social Committee and the Committee of the Regions. Brussels: Commission of the European Communities.  http://ec.europa.eu/education/languages/pdf/com/2008_0566_en.pdf Foote, R. (2007). Transfer in L3 acquisition: The role of typology. In Y.-K. I. Leung (Ed.), Third language acquisition and universal grammar (pp. 89-114). Britol, UK: Multilingual Matters. Frazier, P. A., Tix, A. P., & Barron, K.E. (2004). Testing moderator and mediator effects in counselling psychology research. Journal of counselling Psychology, 51(1), 115-134.  166   Garcia, G. E. (1991). Factors influencing the English test reading performance of Spanish-speaking Hispanic children. Reading Research Quarterly, 26(4), 371- 392. Gardner, R. C. (1985). Social psychology and second language learning: The role of attitudes and motivation. London, England: Edward Arnold. Gardner, R. C., and L. M. Lasynchuck. (1990). The role of aptitude, attitudes, motivation and language use on second-language acquisition and retention. Canadian Journal of Behavioural Science 22(3), 254-270. Genesee, F (1978). A longitudinal evaluation of an early immersion program. Canadian Journal of Education, 3(4), 31-50. Genesee, F. (1983). Bilingual education of majority language children: the immersion experiment in review. Applied Psycholinguistics, 4, 1-46. Genesee, F., Holobrow, N., Lambert, W.E., Cleghorn, A., & Walling, R. (1985). The linguistic and academic development of English-speaking children in French schools: Grade 4 outcomes. Canadian Modern Language Review, 41, 669-685. Genesee, F., & Jared, D. (2008). Literacy development in early French immersion programs. Canadian Psychology, 49(2), 140-147. Geva, E., & Clifton, S. (1994). The development of first and second language reading skills in early French immersion. Canadian Modern Languge Review, 50, 646- 667. Geva, E., & Siegel, L. S. (2000). Orthographic and cognitive factors in the concurrent development of basic reading skills in two languages. Reading and Writing: An Interdisciplinary Journal, 12, 1-30.  167  Geva, E., & Verhoeven, L. (2000). Introduction: The development of second language reading in primary grades – Research issues and trend. Scientific Studies of reading, 4(4), 261-266. Geva, E., & Zadeh, Z. Y. (2006). Reading efficiency in native English-speaking and English-as-a-second-language children: The role of oral language proficiency and underlying cognitive-linguistic processes. Scientific Studies of Reading, 10(1), 31-57. Germain, C., Lightbown, P., Netten, J., & Spada, N. (2004). Intensive French and intensive English: Similarities and differences. Canadian Modern Language Review, 60(3), 409-430. Gottardo, A., Yan, B., Siegel, L. S., & Wade-Woolley, L. (2001). Factors related to English reading performance in children with Chinese as a first language: More evidence of cross-language transfer of phonological processing. Journal of Educational Psychology, 93(3), 530-542. Griessler, M. (2001). The effect of third language learning on second language proficiency: An Austrian example. Journal of Bilingual Education and Bilingualism, 4(1), 50-60. Gunderson, L. (2007). English-only instruction and immigrant students in secondary school: A critical examination. Mahwah, NJ: Erlbaum. Gunderson, L., D’Silva, R., & Murphy Odo, D. (2012). Immigrant students navigating Canadian schools: A longitudinal view. TESL Canada Journal, 29(6), 142-156.  168   Hammarberg, B. (2003). Roles of L1 and L2 in L3 production and acquisition. In J. Cenoz, B. Hufeisen, & U. Jessner (Eds.), Cross-linguistic influence on third language acquisition: Psycholinguistic perspectives (pp. 21-41). Clevedon, UK: Multilingual Matters. Haenni Hoti, A. U., Heinzmann, S., Muller, M., Oliveria, M., Wicki, W., & Werlen, E. (2011). Introducing a second foreign language in Swiss primary schools: The effect of L2 listening and reading skills on L3 acquisition. International Journal of Multilingualism, 8(2), 98-116. Harley, B., Allen, P., Cummins, J. and Swain, M. (1991). The development of second language proficiency. Cambridge, UK: Cambridge University Press. Holm, S. (1979). A simple sequentially rejective multiple test procedure. Scandinavian Journal of Statistics, 6, 65-70. Holmes, J., & Guerra Ramos, R. (1995). False friends and reckless guessers: Observing cognate recognition strategies. In T. Huskin, M. Haunes, & J. Coady (Eds.), Second language reading and vocabulary learning. Norwood, NJ: Ablex Publishing Corporation. Hoover, W. A., & Gough, P. B. (1990). The simple view of reading. Reading and Writing: An Interdisciplinary Journal, 2, 127-190.  169  Hufeisen, B., and N. Marx. (2007). How can DaFnE and EuroComGerm contribute to the concept of receptive multilingualim? Theoretical and practical considerations. In Receptive, multilingualism: Linguistic analyses, language policies and didactic concepts, ed. J. Thije and L. Zeevaert, 307-321. Amsterdam: John Benjamins. Jarvis, S., & Pavlenko, A. (2008). Crosslinguistic influence in language and cognition. New York: Routledge. Jessner, U. (2008). Teaching third language: Findings, trends and challenges. Language teaching, 41(1), 15-56. Jeynes, W. H. (2002). The challenge of controlling for SES in social sciences and education research. Educational Psychology Review, 14(2), 205-221. Jiménez, R. T., Garcia, G. E., & Pearson, D. P. (1996). The reading strategies of bilingual Latina / o students who are successful English readers: Opportunities and obstacles. Reading Research Quarterly, 31(1), 90-112. Katz, L., & Frost, R. (1992). The reading process is different for different orthographies: The orthographic depth hypothesis. In R. Frost & L. Katz (Eds.), Orthography, phonology, morphology, and meaning (pp. 67-84). Amsterdam: Elsevier. Kiefer, M. J., & Lesaux, N. K. (2007). The role of derivational morphology in the reading comprehension of Spanish-speaking English language learners. Reading and Writing, 21, 783-804.  170  Klieme, E. (2006). Zusammenfassung zentraler Ergebnisse der DESI-Studie [Summary of essential results of the DESI-study]. Frankfurt a. M., Deutsches Institut fur Internationale Padagogische Forschung. Retrieved from http://www.dipf./DESI_Ausgewaehlte_Ergebnisse.pdg Koda, K. (1989). The effects of transferred vocabulary knowledge on the development of L2 reading proficiency. Foreign Language Annals, 22(6), 529-540. Ku, Y., and R. C. Anderson. (2003). Development of morphological awareness in Chinese and English. Reading & Writing, 16, 399-422. Lambert, W. E., & Tucker, G. T. (1972). Bilingual education of children: The St. Lambert experiment. Rowley, MA: Newbury House. Leseaux, N. K., & Siegel, L. S. (2003). The development of reading in children who speak English as a second language. Developmental Psychology, 39(6), 1005- 1019. Leung, Y-k. I. (2005). L2 vs. L3 initial state : A comparative study of the acquisition of French CPs by Vietnamese monolinguals and Cantonese-English bilinguals. Bilingualism: Language and Cognition, 8(1), 39-61. Lu, L., & Liu, H. H. (1994). PPVT-Revised. Taipei, Taiwan: Psychological Press. Manis, F. R., Lindsey, K. A., & Bailey, C. E. (2004). Development of reading in Grades K–2 in Spanish-speaking English-language learners. Learning Disabilities Research & Practice, 19, 214–224. Meara, P. (1993). The bilingual lexicon and the teaching of vocabulary. In The bilingual lexicon, ed. R. Schreuder and B. Wetltens, 279-298. Philadelphia, PA: John Benjamins.  171  Metsala, J. L. (1999). Young children’s phonological awareness and nonword repetition as a function of vocabulary development. Journal of Educational Psychology, 91(1), 3-19. McKeown, M., Beck, I., Onanson, R., & Perfetti, C. (1983). The effects of long-term vocabulary instruction on reading comprehension : A replication. Journal of Reading Behavior, 15, 3-18. Miller, J. F., Heilmann, J., & Nockerts, A. (2006). Oral language and reading in bilingual children. Learning Disabilities Research & Practice, 21(1), 30-43. Munoz, C. (2000). Bilingualism and trilingualism in school students in Catalonia. In J. Cenoz & U. Jessner (Eds.), English in Europe : The acquisition of a third language (pp. 157-178). Clevedon, UK : Multilingual Matters. Murphy Odo, D., D’Silva, R., & Gunderson, L. (2012). High school may not be enough: An investigation of Asian students’ eligibility for post-secondary education. Canadian Journal of Education, 35(2), 249-267. Nagy, W. E. (2007). Metalinguistic awareness and the vocabulary-comprehension connection. In R. W. Wagner, A. Muse., & K. R. Tannenbaum (Eds.), Vocabulary acquisition: implications for reading comprehension (pp. 52-99). New York, NY: Guilford Press. Nagy, W. E., & Anderson, R. C. (1984). How many words are there in printed school English? Reading Research Quaterly, 19, 233-253. Nagy, W., Berninger, V., & Abbott, R. (2006). Contribution of morphology beyond phonology to literacy outcomes of upper elementary and middle-school students. Journal of Educational Psychology, 98(1), 134-147.  172  National Reading Panel. (2000). Teaching children to read: An evidence-based assessment of the scientific research literature on reading and its implications for reading instruction. Washington, DC: National Institute for Child Health and Human Development. Odlin, T. (1989). Language transfer: Cross-linguistic influence in language learning. Cambridge, UK: Cambridge University Press. Ordonez, C. L., Carlo, M. S., Snow, C. E., & McLaughlin, B. (2002). Depth and breadth of vocabulary in two languages : Which vocabulary skills transfer? Journal of Educational Psychology, 94(4), 719-728. Paradis, J. (2011). Individual differences in child English second language acquisition: Comparing child-internal and child-external factors. Linguistic Approaches to Bilingualism, 1(3), 213–237. Pasquerella, A., Chen, X., Lam, K., Luo, Y. C., & Ramirez, G. (2011). Cross-language transfer of morphological awareness in Chinese-English bilinguals. Journal of Research in Reading, 34(1), 23-42. Proctor, C. P. August, D., Carlo, M. S., & Snow, C. E. (2006). The intiguing role of Spanish language vocabulary knowledge in predicting English reading comprehension. Journal of Educational Psychology, 98(1), 159-169. Proctor, C. P., August, D., Snow, C., & Barr, C. D. (2010). The interdependence continuum: A perspective on the nature of Spanish-English bilingual reading comprehension. Bilingual Research Journal, 33, 5-20.  173  R Development Core Team (2011). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org/. Ringbom, H. (2003). Lexical transfer in L3 production. In J. Cenoz, B. Hufeisen, & U. Jessner (Eds.), Cross-linguistic influence on third language acquisition: Psycholinguistic perspectives (pp. 59-68). Clevedon, UK: Multilingual Matters. Royer, J. M., & Carlo, M. S. (1991). Transfer of comprehension skills from native to second language. Journal of Reading, 34(6), 450-455. Sanders, M., & Meijers, G. (1995). English as L3 in the elementary school. ITL Review of Applied Linguistics, 107-108, 59-78. Sanz, C. (2000). Bilingual education enhances third language acquisition: Evidence from Catalonia. Applied Psycholinguistics, 21, 23-44. Sanz, C. (2008). Predicting enhanced L3 learning in bilingual contexts: The role of biliteracy. In C. Perez-Vidal, M. Juan-Garau, & A. Bel (Eds.), A portrait of the young in the new multilingual Spain (vol. 9, pp. 220-240). Clevedon: Multilingual Matters. Savage, R. (2001). The ‘simple view’ of reading: Some evidence and possible implications. Educational Psychology in Practice, 17(1), 17-33. Schoonen, R., van Gelderen, A., de Glopper, K., Hulstijn, J., Simis, A., Snellings, P.,…Stevenson, M. (2002). Linguistic knowledge, metacognitive knowledge and retrieval speed in L1, L2 and EFL writing. In S. Randsdell & M.-L. Barbier (Eds.), New directions for research in L2 writing (pp. 101-122). Dordrecht: Kluwer Academic.  174  Singer, J. D., & Willett, J. B. (2003). Applied longitudinal data analysis: Modeling change and event occurrence. Oxford, UK: Oxford University Press. Snow, C. E., Burns, M.S., & Griffin, P. (1998). Preventing reading difficulties in young children. Washington, DC: National Academy Press. Stanovich, K. E. (1986). Matthew effects in reading: Some consequences of individual difference in the acquisition of literacy. Reading Research Quaterly, 21(4), 360- 407. Storch, S. A., & Whitehurst, G. J. (2002). Oral language and code-related precursors to reading: Evidence from a longitudinal structural modal. Developmental Psychology, 38, 934-947. Swain, M. (1981). Bilingual education for majority and minority language children. Studia Linguistica, 15, 15-31. Swain, M., and S. Lapkin. (1982). Evaluating bilingual education: A Canadian case study. Clevedon, England: Multilingual Matters. Swain, M., & Lapkin, S. (2005). The evolving socio-political context of immersion education in Canada: some implications for program development. International Journal of Applied Linguistics, 15(2), 169-186. Swain, M., Lapkin, S., Rowen, N., & Hart, D. (1990). The role of native language literacy in third language learning. In S. P. Norris & L. M. Phillips (Eds.), Foundations of literacy policy in Canada (pp. 185 – 206). Calgary, Alberta: Detselig Enterprises Ltd.  175  Taylor, S.K. (2006). ESL in bilingual education with a twist: Immigrant children in Canadian French Immersion. Teachers of English to Speakers of Other Languages, 8(1). Turnbull, M.K., Lapkin, S., & Hart, D. (2003). Grade 6 French immersion students’ performances on large-scale reading, writing, and mathematics tests: building explanations. Alberta Journal of Education, 49(1), 6-23. Vancouver School Board. (2011). Early French immersion: K to 7. Retrieved from http://www.vsb.bc.ca/programs/early-french-immersion. Van Gelderen, A., Schoonen, R., Gloper, K., Hulstijn, J., Snellings, P., Simis, A …Stevenson, M. (2003). Roles of linguistic knowledge, metacognitive knowledge and processing speed in L3, L2, and L1 reading comprehension: A structural modeling approach. International Journal of Bilingualism, 7(1), 7-25. Verhoeven, L., & Leeuwen, J. (2012). The simple view of second language reading throughout the primary grades. Reading & Writing, 25, 1805-1818. Wang, M., C. Cheng, and S. W. Chen. (2006). Contributions of morphological awareness to Chinese-English biliteracy acquisition. Journal of Educational Psychology 98(3), 542-553. Wang, M., Ko, Y., and J. Choi. (2009). The importance of morphological awareness in Korean-English biliteracy acquisition. Contemporary Educational Psychology 34 (2), 132-142.  176  Wechsler, D. (2007). Wechsler individual achievement test – (2nd ed.). San Antonio, TX: Psychological Corp. Woodcock, R. W. (1991). Woodcock Language Proficiency Battery-Revised. Itasca, IL: Riverside Publishing.          177  Appendices Appendix A: Parent questionnaire  We would like to collect information about your family and about your child who is participating in this project. If possible, the PRIMARY CAREGIVER, or the person most knowledgeable about the child, should fill this out. We are interested in general information regarding your child and would also like to establish the home language(s) the child has been exposed to, this includes languages other than English that are spoken primary at home, but also with other caregivers or friends. Please answer only the questions that are relevant to you and your child. For example, if your home language is English, then ignore questions that refer to the home language.  All your answers are confidential and will only be used for the purposes of this project. Please return this to your teacher within a week. If you have questions or need help filling out this questionnaire, please contact your child’s teacher or the Principal Researcher at UBC, Stefka H. Marinova-Todd at (604) 822-0276. We will be glad to assist you.  Please write, check or circle the appropriate response to the following questions.  Your name: ______________________________________________  Phone Number: ___________________________________________   PLEASE ANSWER THE FOLLOWING QUESTIONS ABOUT YOUR GRADE 6 CHILD.  1. What is the full name of your child who is currently in Grade 6? ________________________________________________________ 2. What is your child’s birthdate?  _____/______/_______     day/month/year  3. In what country was your child born?  Canada    Other (where? ____________________)  4. How many years has your child lived in Canada or another English-speaking country?   __________ years.  5. Since your child has lived in Canada, about how much time per year does he/she spend in a country other than Canada?  178   none 1 to 7 days   2 to 3 weeks  1 month  more than 1 month  6. At what age did your child start receiving consistent exposure to the home language (other than English and French)? ______________.  7. At what age did your child start receiving consistent exposure to English? ____________.  8. What language(s) would you like your child to speak when he/she graduates from high school?  English  French  Home language      Other   9. What language(s) would you like your child to read and write when he/she graduates from high school?  English  French  Home language     Other  10. Who speaks the home language (other than English or French) to your child? (Please mark all that apply)  mother    father     mother’s side grandmother   mother’s side grandfather  father’s side grandmother   father’s side grandfather  other relatives   brothers/sisters   his/her friends  neighbors      shopkeepers    other  11. Who speaks English to your child?  (Please mark all that apply)  mother    father     mother’s side grandmother   mother’s side grandfather  father’s side grandmother   father’s side grandfather  other relatives   brothers/sisters   his/her friends  neighbors      shopkeepers    other  12. Who speaks French to your child?  (Please mark all that apply)  mother    father     mother’s side grandmother   mother’s side grandfather  father’s side grandmother   father’s side grandfather  other relatives   brothers/sisters   his/her friends  neighbors      shopkeepers    other  13. Estimate the amount of time your child speaks in the home language (other than English and French) on a weekly basis  179  0%  0-10% 10-20% 20-30% 30-40% 40-50% 50-60% 60-70% 70-80% 80-90% 90-100%  14. Estimate the amount of time your child speaks in English on a weekly basis 0%  0-10% 10-20% 20-30% 30-40% 40-50% 50-60% 60-70% 70-80% 80-90% 90-100%  15. Estimate the amount of time your child speaks in French on a weekly basis 0%  0-10% 10-20% 20-30% 30-40% 40-50% 50-60% 60-70% 70-80% 80-90% 90-100%  16. List all the  languages that your child can speak (including home language, English French, other languages)  1)__________________2)_________________3)_________________4)_______  17. List all the  languages that your child can has been able to read in the past or can currently read (including home language, English, French)  1)__________________2)_________________3)_________________4)_______  18. How old was your child when s/he first learned to read in a language other than English or French? _____________  19. How old was your child when s/he first learned to read in English? ________  20. How old was your child when s/he first learned to read in French? _________  21. Does your child:  watch TV   in what language:  ____________      listen to the radio  ____________      play computer games  ____________  22. Do you have books in the home language (other than English and French) for your child in your home?  yes        no  If yes, how many:  NONE      BETWEEN 1 AND 10      BETWEEN 10 AND 25      MORE THAN 25  23. Do you have English books for your children in your home?  180   yes        no  If yes, how many:  NONE      BETWEEN 1 AND 10      BETWEEN 10 AND 25      MORE THAN 25  24. Do you have French books for your children in your home?  yes        no  If yes, how many:  NONE      BETWEEN 1 AND 10      BETWEEN 10 AND 25      MORE THAN 25  25. How often does your child read books in the home language (other than English and French)?   ALMOST EVERY DAY   3 TO 5 DAYS A WEEK   1 TO 2 DAYS A WEEK   1 TO 2 TIMES A MONTH   LESS THAN 1 TO 2 TIMES A MONTH   NEVER  26. How often does your child read English books?   ALMOST EVERY DAY   3 TO 5 DAYS A WEEK   1 TO 2 DAYS A WEEK   1 TO 2 TIMES A MONTH   LESS THAN 1 TO 2 TIMES A MONTH   NEVER  27. How often does your child read French books?   ALMOST EVERY DAY   3 TO 5 DAYS A WEEK   1 TO 2 DAYS A WEEK   1 TO 2 TIMES A MONTH   LESS THAN 1 TO 2 TIMES A MONTH   NEVER  28. Does your child go to the library?  181      yes        no If yes, how often does your child go to the library? _________________________________  29. Do you or any other caregiver in the home, or outside the school, speak French?  Yes   No If yes, who speaks French and how much time does your child spend speaking French?   30. Does your child receive reading instruction before or after school?  Yes   No  If yes, what languages do they take place in? Home language English ngFrench  reOther: _____________  PLEASE ANSWER THE FOLLOWING QUESTIONS ABOUT YOURSELF  The following questions relate to the mother (or the female head of household). If this does not apply to you, then note here and continue with question 36 .  31. What country/province was the mother/guardian born in? ______________________ How long has she lived in Canada? ____________.  32. What is the mother’s native language?  ________________  33. What language(s) does the mother speak with her child? (Select one). English/French never English/French seldom English/French 50%       Home language always          Home language usually          Home language 50%  English/French usually English/French almost always       Home language seldom      Home language almost never  34. On a scale of 1 to 10, how well can the mother speak (please circle): In home language 1  2  3  4  5  6  7  8  9  10    Not all   Very well  In English   1  2  3  4  5  6  7  8  9  10       Not all   Very well  In French  1  2  3  4  5  6  7  8  9  10       Not all   Very well 35. On a scale of 1 to 10, how well can the mother read (please circle):  182  In home language 1  2  3  4  5  6  7  8  9  10     Not all   Very well  In English   1  2  3  4  5  6  7  8  9  10       Not all   Very well  In French  1  2  3  4  5  6  7  8  9  10       Not all   Very well  36. On a scale of 1 to 10, how well can the mother write (please circle): In home language 1  2  3  4  5  6  7  8  9  10     Not all   Very well  In English   1  2  3  4  5  6  7  8  9  10       Not all   Very well  In French  1  2  3  4  5  6  7  8  9  10       Not all   Very well  37. What is the mother’s highest level of education completed at this time?       none       some primary education       completed primary education       some high school       graduated from high school       some college or trade school       received associate’s degree or trade certification       received bachelor’s degree   (Major: __________________________)       some graduate study       received graduate degree         other  The following questions relate to the father (or the male head of the household): If this does not apply to you then note here and this is the end .  38. What country/province was your child’s father/guardian born in? __________________ How long has he lived in Canada? ____________.  39. What is the father’s native language? _______________________  40. What language(s) does the father speak with her child? (Select one).  183  English/French never English/French seldom English/French 50%       Home language always           Home language usually          Home language 50%  English/French usually English/French almost always       Home language seldom      Home language almost never   41. On a scale of 1 to 10, how well can the mother speak (please circle): In home language 1  2  3  4  5  6  7  8  9  10    Not all   Very well  In English   1  2  3  4  5  6  7  8  9  10       Not all   Very well  In French  1  2  3  4  5  6  7  8  9  10       Not all   Very well   42. On a scale of 1 to 10, how well can the father read (please circle): In home language 1  2  3  4  5  6  7  8  9  10     Not all   Very well  In English   1  2  3  4  5  6  7  8  9  10       Not all   Very well  In French  1  2  3  4  5  6  7  8  9  10       Not all   Very well  43. On a scale of 1 to 10, how well can the father write (please circle): In home language 1  2  3  4  5  6  7  8  9  10     Not all   Very well  In English   1  2  3  4  5  6  7  8  9  10       Not all   Very well  In French  1  2  3  4  5  6  7  8  9  10       Not all   Very well  44. What is the FATHER/GUARDIAN’S highest level of education completed at this time?  184        none       some primary education       completed primary education       some high school       graduated from high school       some college or trade school       received associate’s degree or trade certification       received bachelor’s degree   (Major: __________________________)       some graduate study       received graduate degree       other   Thank you very much for your cooperation!  Adapted from: Paradis, J. (2011). Individual differences in child English second language acquisition. Linguistic Approaches to Bilingualism, 1(3), 213-237.  185  Appendix B: Motivation to learn French questionnaire Please check the box that applies to your answer       Totally disagree Disagree Neutral Agree Totally agree 1) I enjoy listening to music and watching television in French   2) All students should learn French in school   3) I find that it is a waste of time to study French   4) In enjoy speaking to my friends in French   5) I find French difficult to learn   6) I will probably continue to use French once I am finished school   7) I work on my French homework very regularly   8) I never quite feel sure of myself when I am speaking French   9) I really work hard to learn French   10) I feel uncomfortable speaking French   11) I enjoy learning French    186  Appendix C: Derivational morphology test Production Task Instructions and Items. “I want you to help me with a word game. I’m going to say a word and I want you to change it to fit the sentence.  Let’s try one.” 1. Help.  My sister is always _________How could you change help to fit that sentence? [helpful] If they get that, say, ‘Right. My sister is always helpful’ If they miss the first one, explain that they need to change it to fit the sentence (help  has to become helpful) and repeat the complete sentence with the correct answer. Give next examples in the same way. 2. Farm. My uncle is a ____.  How could you change farm to fit that sentence? [farmer] 3. Improve. His work shows great ___________[improvement] “Ok. Do you have any questions? Great. Remember you just have to change the word to fit the sentence.”  Not to the experimenter. You can repeat the item for a child. Please write down the child’s response (it can later be coded as correct or incorrect).   Word Sentence Correct response Child’s response Correct?( ) 1 music.  That lady with the piano is a _____ Musician 2 popular The singer enjoys his _____  Popularity 3 Growth She wanted her plant to _____ Grow 4 slow I was glad that I wasn’t the _____ Slowest 5 science I want to grow up to be a _____ Scientist 6 four The horse came in _____ Fourth 7 local The birds migrated to a new _____ Location 8 produce The play was a grand _____  Production 9 vary The time of his arrival is _____  Varied 10  wide We measured the river’s _____ Width 11 assist The teacher will give _____ Assistance 12 calm The teacher asked us to walk _____ Calmly 13 magic He was a very good _____ Magician 14 strong He wanted to show off his _____ Strength     187  Continued Appendix C: Derivational morphology test  15 discuss Mum and Dad had a long boring _____ Discussion 16 appear He cared about his _____ Appearance 17 remark The speed of the car was _____ Remarkable 18 major He won the vote by a _____ Majority 29 density The smoke in the room was very _____  Dense 20 mystery The dark glasses made the man look ___  Mysterious 21 permit Father refused to give _____   Permission 22 Jacket Millie has three _____ Jackets 23 human The kind man was known for his _____ Humanity 24 skip Yesterday at recess, the girls _____ Skipped 25 description The picture is hard to _____ Describe 26 runner How fast can she _____ Run 27 cloud I really hope that it’s not _____ Cloudy TOTAL    

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0073901/manifest

Comment

Related Items