Transfer of Persuasive Writing Skills from English L2 to Arabic L1: A Reverse Perspective, Exercises of English Philology

Download Transfer of Persuasive Writing Skills from English L2 to Arabic L1: A Reverse Perspective and more Exercises English Philology in PDF only on Docsity! TRANSFER OF KEY LANGUAGE FEATURES FROM L2 ENGLISH ACADEMIC PERSUASIVE ESSAY TO L1 ARABIC ACADEMIC PERSUASIVE ESSAYS BY ARABIC EFL LEARNERS A Thesis Submitted by Abdalrhman Alsaghier, B.Ed., M Ed. For the award of Doctor of Philosophy 2020 II Abstract Transfer between languages is generally accepted as being predominantly a positive phenomenon. Less is known, however, of the effects of transfer on the quality of writing in English and in Arabic and the reverse transfer in writing from English to Arabic for bilingual higher education students. This study reports a contrastive analysis of Arabic scripts and English scripts on nine features of writing proficiency, of an unrehearsed persuasive essay writing task undertaken by 40 bilingual Arabic L1/English L2 fourth-year undergraduate students aged 20-24. This genre was selected because of its relevance to academic writing skills that the students were seeking to acquire. The marking criteria were adopted from the Australian National Assessment Program - Literacy and Numeracy (NAPLAN) test. The findings of a positive cognitive impact on students’ persuasive writing after receiving English-medium higher education in an Arabic university, were supported by reflective survey responses. A stimulated-recall methodology by interview prompted participants to deconstruct the purpose of specified features of writing. The results were revealed in a detailed analysis of the individual scripts, by additional comparison with monolingual Arabic writers. Positive transfer of key language features from English L2 to Arabic L1 writing was evident for audience, ideas, persuasion, devices, cohesion and paragraphing. V Table of Contents Abstract ...................................................................................................................... II Certification of Thesis ............................................................................................. III Acknowledgements ................................................................................................... IV Table of Contents ...................................................................................................... V List of Tables ............................................................................................................ IX List of Figures ........................................................................................................... XI List of Abbreviations ............................................................................................. XII Chapter 1: Introduction ............................................................................................ 1 An Overview ................................................................................................................ 1 1.1 Background ........................................................................................................ 1 1.2 Research problem ............................................................................................... 4 1.3 Research Questions ............................................................................................ 6 1.4 Purpose of the Study ........................................................................................... 6 1.5 Significance of the Study ................................................................................... 7 1.6 The Organization of the Thesis .......................................................................... 8 Chapter 2: Literature Review ................................................................................. 10 An Overview .............................................................................................................. 10 2.1 Concept of Traditional Contrastive Rhetoric ................................................... 10 2.1.1 Criticism of Traditional Contrastive Rhetoric ........................................... 12 2.1.2 Summary .................................................................................................... 14 2.2 Cognitive Theories ........................................................................................... 14 2.2.1 Theory of Bilingualism .............................................................................. 15 2.2.2 Theory of Multicompetence....................................................................... 17 2.2.3 The Interdependence Hypothesis ............................................................... 18 2.2.4 Threshold Hypothesis ................................................................................ 22 2.2.5 Summary .................................................................................................... 24 2.3 Differences between Writing in Arabic and English ........................................ 25 2.4 Arabic Persuasive Writing ................................................................................ 28 2.5 Research Studies on Transfer of Writing Skills from L2 to L1 ....................... 29 VI 2.6 Summary .......................................................................................................... 38 CHAPTER 3: Methodology .................................................................................... 39 Overview .................................................................................................................... 39 3.1 General Design of the Study ............................................................................ 39 3.2 Research Method Adopted for the Current Study ............................................ 40 3.2 Participants in the Study ................................................................................... 43 3.4 Methodological Triangulation .......................................................................... 44 3.5 Data Collection Instruments ............................................................................. 46 3.5.1 Phase 1: Writing Tasks .............................................................................. 46 3.5.2 Phase 2: Stimulated Recall Interviews....................................................... 50 3.5.3 Phase 3: Survey .......................................................................................... 52 3.6 Ethical Considerations ...................................................................................... 56 CHAPTER 4: Results of marking of English and Arabic written scripts .......... 57 Introduction ................................................................................................................ 57 4.1 Comparison of G1AdvEL2/AL1’ Performance versus G2IntermEL2/AL1’ Performance on Writing a Persuasive Genre in English ........................................ 58 4.1.1 The Use of Key Language Features in the English Essays – G1AdvEL2/AL1. ................................................................................................ 59 4.1.2 The Use of Key Language Features in the English Essays – G2IntermEL2/AL1.............................................................................................. 66 4.1.3 Overall Quality of the G1AdvEL2/AL1 and G2IntermEL2/AL1 English Essays.................................................................................................................. 72 4.1.4 Summary .................................................................................................... 73 4.2 Scores of Arabic Scripts ................................................................................... 74 4.2.1 The Use of Key Language Features in the Arabic Essays – the G1AdvEL2/AL1 Group. ..................................................................................... 74 4.2.2 The Use of Key Language Features in the Arabic Essays – G2IntermEL2/AL1.............................................................................................. 82 4.2.3 The Use of Key Language Features in the Arabic Essays – G3ArabicMonoL1 Students ................................................................................ 89 VII 4.2.4 A Comparison of the Use of Key Language Features in English and Arabic Essays by Students Majoring in English (G1AdvEL2/AL1 and G2IntermEL2/AL1) ............................................................................................ 95 4.2.5 Overall Quality of the L1 Arabic Essays Produced by Students Majoring in English (G1AdvEL2/AL1 and G2IntermEL2/AL1) and Students Majoring in Arabic (G3ArabicMonoL1) ................................................................................ 96 4.2.6 Summary .................................................................................................... 97 CHAPTER 5: Findings from the Stimulated Recall Interviews ......................... 99 Introduction ................................................................................................................ 99 5.1 Use of Key Language Features in L2 English and L1 Arabic Persuasive Essay Writing as Reported by the Interviewees ............................................................... 99 5.1.1 What Other Language Features did You Learn in L2 Essay Writing Lessons in Particular? ....................................................................................... 100 5.1.2 How do You Write to Communicate Your Opinion to the Reader? ........ 102 5.2 Bilingual Arabic L1 / English L2 Students’ Expectations in Persuasive Essay Writing .................................................................................................................. 102 5.3 The Influence of English Essay Writing Experience on Arabic Writing and Vice Versa ..................................................................................................................... 104 5.4 Summary ........................................................................................................ 106 CHAPTER 6: Findings from the Surveys ............................................................ 107 An Overview ............................................................................................................ 107 6.1 Arabic (L1) Writing Background ................................................................... 107 6.1.1 Summary .................................................................................................. 111 6.2 English (L2) Writing Background .................................................................. 112 6.2.1 Self-Evaluation of English Proficiency ................................................... 117 6.2.2 Summary .................................................................................................. 118 6.3 Personal Attitudes towards Writing Activity ................................................. 119 6.4 Writing Practices ............................................................................................ 120 6.5 Learners’ Perception of the Effect of English Writing Instruction ................ 120 6.6 Summary ........................................................................................................ 120 Chapter 7: Discussion, Conclusions, and Recommendations ............................ 122 Introduction .............................................................................................................. 122 X Table 6.4: Common topics found in Arabic writing instruction in secondary schools and university ........................................................................................................... 109 Table 6.5: Techniques for persuasive readers in essays reportedly covered during Arabic writing instruction received by participants ................................................. 110 Table 6.6: Organization of good persuasive essays studied by the students during their Arabic writing instruction ........................................................................................ 111 Table 6.7: Types of key language features reported to be mostly the focus of the students’ Arabic writing teachers ............................................................................. 111 Table 6.8: Weekly English writing instruction ........................................................ 113 Table 6.9: Types of writing studied by the students in English writing instruction 113 Table 6.10: English writing instruction topics ......................................................... 114 Table 6.11: Techniques for persuasive readers in essays reportedly covered during English writing instruction received by participants ................................................ 115 Table 6.12: Components of good essays reportedly studied by the participants during their English writing instruction............................................................................... 115 Table 6.13: Types of key language features reported to be mostly focused on by the students’ English writing teachers ........................................................................... 116 Table 6.14: Amount of English instruction that participants reported receiving throughout their education ....................................................................................... 116 Table 6.15: Number of hours of English instruction that participates took per week at high school and university ....................................................................................... 117 Table 6.16: Exposure to English in secondary and university studies ..................... 117 Table 6.17: Students’ self-evaluation of English proficiency (N= 209) .................. 118 Table 6.18: Writing qualities that the students believe they need to be a good writer .................................................................................................................................. 119 Table 6.19: The students’ reports of publishing apart from their writing in classrooms .................................................................................................................................. 120 Table 7.1: Group G1AdvEL2/AL1’s English scripts versus group G2IntermEL2/AL1’s English scripts ....................................................................... 129 Table 7.2: Group G1AdvEL2/AL1’s English scripts versus group G3ArabicMonoL1’s Arabic scripts ........................................................................................................... 133 XI List of Figures Figure 2.1 cultural thought patterns in intercultural education (Kaplan, 1966: p.21) 11 Figure 2.2 an overview of the research conceptual framework of the study ............ 24 Figure 2.3 the structure of argumentative texts. Adapted from Hatim (1991, p. 193) .................................................................................................................................... 27 Figure 3.1 examples of the implementation of methodological triangulation in research .................................................................................................................................... 45 Figure 3.2 data collection process for the present study ............................................ 46 Figure 3.3 NAPLAN persuasive writing scoring criteria and range of scores. Source: (NAPLAN, 2011). ....................................................................................................... 49 Figure 3.4 scale conversion of the NAPLAN scores ................................................. 49 Figure 3.5 research design to analysis of the respective key language features across the various student groups’ written persuasive texts ................................................. 50 Figure 7.1 the research design showing how the analysis of the respective key language features across the various student groups written persuasive texts revealed text differences ................................................................................................................ 125 Figure 7.2 Mind map of the research findings relate to research questions and new knowledge that is emerging from research finding .................................................. 127 XII List of Abbreviations L2 English as the second language of use L1 The social and academic Arabic language EFL English as a Foreign Language CR Contrastive Rhetoric IH Interdependence Hypothesis TH Threshold Hypothesis CUP Common Underlying Proficiency BICS Basic Interpersonal Communicative Skills CALP Cognitive Academic Language Proficiency PT1 participant number 1 PT2 participant number 2 PT3 participant number 3 PT4 participant number 4 G1AdvEL2/AL1 Advanced English as a second language/Arabic as first language G2IntermEL2/AL1 Intermediate English as a second language/Arabic as first language G3ArabicMonoL1 Arabic as first language monolingual 3 instruction was needed before any positive influences could be revealed on their learnings in Chinese. Kobayashi and Rinnert (2007) also supported Cummins’ interdependence hypothesis. They revealed evidence that the training and practice in writing experienced by EFL (English Foreign Language) Japanese students in overseas educational settings had effects on opinion writing in their native language (Japanese). They further disclosed that the transfer of English L2 to Japanese L1 did not occur automatically; rather, there were factors operating for the transfer to happen, namely language proficiency and disciplinary knowledge. Similarly, Agheshteh (2015) revealed the long-term influence of intensive study of an L2 on L1 composing skills with Iranian learners learning a foreign language for seven years. The Iranian learners studied English in the intermediate and secondary school and then at the university revealed the positive impacts learning English had on their L1 writing skill, were all the Iranian learners bilingual better performance than their monolingual counterparts in their L1 essay writing. Wang (2014) research added to this in emphasising that a certain proficiency level is needed to successfully transfer composing skills from L2 to L1. Unfortunately, only a few studies (Agheshteh, 2015; Akyel & Kamisli, 1996; Kecskes & Papp, 2000; Kobayashi & Rinnert, 2007; Y. Liu & Carney, 2012) have investigated the reverse transfer of writing skills from L2 to L1 among adult learners in settings where L2 is a foreign language. Furthermore, most of these studies were limited because they employed only one instrument for data collection and were less relevant to the present research into AFL L2 and Arabic L1, since they were conducted in European and American contexts. Thus, there is a problem about the cognateness of the languages being investigated. The study of transfer between two cognate languages is less likely to be able to reveal backward transfer since the languages generally have more matches than dissimilarities in terms of linguistic and rhetorical notions contrasted to transfer between non-cognate languages (Peukert, 2015). Thus, in the present research context of EFL and L2 Arabic, an empirical study with a comprehensive methodology is needed to explore reverse transfer between these two non-cognate languages. Palfreyman and Karaki (2017) stated that “English and Arabic are genetically unrelated and very different at various linguistic levels, including their 4 script”. Alshammari (2016) agreed with Palfreyman and Karaki, that the Arabic language has a very different rhetorical style from the English style, especially in persuasive writing. Moreover, Rass (2015) and Abbadi (2014) also claimed that English and Arabic are dissimilar not only in the textual structure of argumentation, but also in the linguistic strategies employed. Having reviewed available literature and research studies, it seems that few investigations have been carried out in an Arabic context, yet Arabic speaking countries choose to teach EFL and it is well established that students find learning EFL difficult both in schooling (Al-Mahrooqi & Denman, 2014; Al-Nasser, 2015) and in higher education (Alrashidi & Phan, 2015; Bailey & Damerow, 2014). Therefore, research is needed to investigate in depth how English (L2) affects Arabic (L1) (Aldosari & Alsultan, 2017; Aljohani, 2016; Hussien, 2014a, 2014b; Van De Wege, 2013). Taking this research gap into consideration, an empirical study is needed to contribute to the knowledge of Arabic speaking background EFL learners’ cross- language transfer of key language features from L2 to L1, particularly in the academic writing context. Ortega (2014) stated that further research is needed on bidirectional transfer in order to draw conclusions about cross-linguistic effects. In the present study, the term ‘transfer’ will be defined similarly to the way Kecskes and Papp (2003) conceptualized it, that is “any kind of movement and/or influence of concepts, knowledge, skills or linguistic elements (structures, forms), in either direction, between the L1 and the subsequent language(s)” (p. 251). In the context of this current investigation, the term ‘transfer’ particularly refers to the utilization in participants’ L1 of language features (e.g., as audience, text structure, ideas, persuasive devices) as a result of their exposure to English as the language of learning at university level. 1.2 Research problem There is a lack of empirical studies investigating how the key language features of writing cognitively transfer from second language to first language. Previous studies have to a large extent concentrated on the influences of first language writing skills on second language learning, ignoring the effects that second language could have on bilinguals’ native language. For L1 speakers of Arabic, it is often highly advantageous to acquire competence and proficiency in L2 English, particularly in writing skills. The 5 acquisition of such L2 English skills is not only advantageous for academic purposes, as the research in this thesis will show, but also meets needs concerning communication and reasoning that can be deployed in English as the lingua franca of business and organisations in a globalising world. English is the language of much information available online, and is also often used for intercultural communication in a variety of global contexts outside its applications in normatively L1 workplaces or domains. Therefore, there are very pragmatic reasons for research into this problem – not restricted to particular situations or domains. The concatenation of fields in which L2 English can be advantageous is cumulative – that is, while travel and the employment market may be motivations, the acquisition of L2 English as a global language or as a lingua franca opens up a multitude of horizons. Here the importance of writing and persuasive writing skills in particular, is key, and they are shaped by language transfer from L1 Arabic. It is in this space that this thesis frames its research problem. Investigators have long been interested in the ways languages can impact on each other within the bilingual’s mind. Generally, however, this interest has been unidirectionally concentrated in that the question has focused on the extent to which a learner’s mother tongue influences the second language; negative transfer is dominant in terms of lexical and/or grammatical mistakes (Al-Zoubi & Abu-Eid, 2014; Murad & Khalil, 2015). In exploratory studies by L2 English, French and Russian researchers into L1 Hungarian (Kecskes & Papp, 2000), L2 English into L1 French (Cook, 2003), L2 English into L1 Chinese (Y. Liu & Carney, 2012), and L2 English into L1 Iranian (Agheshteh, 2015) generally positive effects are expected. Gonca (2016) argued that an intensive L2 English teaching program allowed L2 students not only to progress their L2 knowledge, but also to enhance their L1 strategies. In addition Liu & Carney (2012) found that the intensive L2 English teaching program at university improved the Chinese college L2 users’ abilities in their L2 and L1 academic persuasive essays. The students employed a direct approach smoothly in both their L2 and L1 essays and seemed more flexible and willing to utilize transitions to indicate the movement of the ideas between the paragraphs in both their L2 and L1 academic persuasive essays. The combination of their L1 and L2 knowledge produced improvements in their skills in both languages (Enama, 2015). Improving students’ achievements in one language can bring about similar improvements in other languages (Abu-Rabia & Shakkour, 2014). 8 students employed direct approach smoothly in both their L2 and L1 essays and seemed more flexible and willing to utilize transitions to indicate the movement of the ideas between the paragraphs in both their L2 and L1 persuasive essays. Agheshteh (2015) revealed that Iranian bilinguals performed better than Iranian monolinguals on an L1 essay-writing test. This study is timely and beneficial for the following reasons. The results of this study might lend support to or criticism of the viability of the interdependence hypothesis offered by Cummins. The current investigation builds on a comprehensive methodology to test the transfer of key language features within adult EFL, in contrast or agreement with Cummins’ theory in relation to transfer of writing skills between children’s L1 and L2. The majority of children are not developmentally ready to understand the complexities of the effective presentation of an argument and tend to present only basic knowledge of argumentation, like stating their opinion and giving reasons in support of this opinion, in both oral and written work (Ferretti, MacArthur, & Dowdy, 2000). To the best of this researcher’s knowledge, there is no comprehensive study in the Arabic educational context that focuses exclusively on the transfer of key language features from L2 English to L1 Arabic. There is a dearth of study published about Arabic persuasion strategies (Suchan, 2014). The results of the present study will contribute to a better understanding of how adult bilingual Arabic L1/English L2 students transfer the key language features they applied in their English scripts to their subsequent persuasive writing in Arabic. Further, the findings will give valued information to Arabic speaking teachers and EFL teachers by showing how English composing instructions should be approached and assessed by L2 Arabic speakers. Additionally, the outcomes may give feedback to writing instructors in terms of presenting the strengths and problems for Arabic EFL students in their L1 and L2 writing. 1.6 The Organization of the Thesis This thesis has seven chapters. The first chapter, the present one, introduces the study, states key terms used in the investigation, the problem of the research, introduces the research questions, provides the aim of the study, presents its significance, and then offers a summary of the forthcoming chapters. The second chapter includes reviews of the theoretical concepts and relevant empirical studies associated with the issue of cross-linguistic influence in writing. The aim of the literature review is to show the 9 research gaps that indicate the significance and the purpose of this study. Chapter 3 provides detailed information on the research methodology. More specifically, in this section, the researcher provides information on the context of the study, the participants, data collection and processes, and the data analysis and justification of the research design. Chapter 4 provides the analysis and results of the participants’ Arabic and English essays. Chapter 5 offers analysis and results of the stimulated recall interviews conducted with the participants to explore their preferences for using a particular key language feature in their L1 and L2 essays. In addition, through the interviews, the researcher can extract information from the long or short-term memory of a particular writer, and ascertain why he or she structured the texts in certain ways. Chapter 6 presents the analysis and results of the survey to identify aspects of essay writing learning that the writers themselves had been exposed to in their teacher- training courses in both languages, to identify factors that might affect the reverse transfer of key language features from English L2 to Arabic L1. Finally, Chapter 7 provides the results of mind mapping, discussion of findings in relation to the research questions, the limitations of the study, the study’s conclusions, implications of the research and recommendations, and the study’s contribution to knowledge. 10 Chapter 2: Literature Review An Overview This chapter discusses the theory of traditional contrastive rhetoric and the close correlation between culture and L2 writing and the criticism against the theory of traditional contrastive rhetoric. The chapter assesses cross-linguistic transfer based on cognitive theories of bilingualism and multicompetence, and the interdependence hypothesis and threshold hypothesis. Although the cognitive theories guide the overall rationale of the present study, the research conceptual framework was built on Interdependence Hypothesis and Threshold Hypothesis as being especially relevant to the main purpose of this study in seeking evidence of bilingual Arabic L1/English L2 transfer of the key language features they applied in their English scripts to their subsequent persuasive writing in Arabic. This chapter also discusses the differences between writing in Arabic and writing in English, Arabic persuasive writing, and studies on the transfer of writing skills from L2 to L1. 2.1 Concept of Traditional Contrastive Rhetoric The theory of Contrastive Rhetoric originated in 1966 when Kaplan (1966) wrote the ground-breaking article, “Cultural thought patterns in intercultural education”. With a focus on the extra-sentential level, in this article, Kaplan designed a new theoretical instrument to study written texts for problem-solving in the area of second language writing. According to Kaplan, language and writing bind together culturally, making each language’s rhetorical nature peculiar to it. Moreover, writing in English as an L2 is subject to interference from the rhetorical and linguistic conventions of the learners’ L1 (Connor, 1996). Ferris, Hedgcock, and Hedgcock (2013) claimed that in the context of learning to write in an L1, cultural aspects lead people to have various expectations with regard to the organization of texts; consequently, L2 learners’ schemata vary from those of L1 learners, causing challenges in communication. For them, Contrastive Rhetoric aims at determining such anticipations and their effects on L2 literacy development that involve L2 writing skills. 13 processes for writing. Further, the act of generalizing the writing skills of learners based on their cultural identities was denounced. Even Zamel (1997) declared that teachers and researchers who see “L2 learners as bound by their cultures could be trapped by their own cultural tendency to reduce, categorize, and generalize” (p. 342). As a result, the L2 teachers became restricted as they were less able to understand their students as L2 writers (Zamel, 1997). To concentrate on the complexity, variability, and unpredictability of cultures became a major necessity for teachers and researchers. Around the same time, (Kubota, 1997) also presented a critical view of the cultural dichotomy between East and West by contributing empirical studies on English and Japanese writing. She showed that the Japanese writing organization (ki-shoo- tenketsu) was shunned in Japanese writing, because of linguistic and educational impacts from the West since the mid-nineteenth century. In another investigation, Kubota (1998) explored why Japanese university students chose to use the same rhetorical patterns in English and Japanese persuasive essays and how those patterns affected their writings. The outcomes of this exploration revealed that the writers were not implementing the same patterns in all their L1 and L2 writings. Furthermore, there was a tendency for L1-L2 transfer of writing strategies rather than negative transfer from L1specific rhetoric. For an explicit critique of Contrastive Rhetoric, one can turn to the work of Kubota and Lehner (2004). They observed critical Contrastive Rhetoric studies that “incorporate key concepts drawn from postmodernism, poststructuralism, postcolonialism, and critical pedagogy which are already being integrated in the larger field of applied linguistics"(p. 9) This view challenged the idea that English is the superior language and that writing stresses cultural thought rather than social concerns. As Kubota and Lehner (2004) alleged: “Contrastive Rhetoric has a ―reductionist, deterministic, prescriptive and essentialist orientation” (p. 10). They also suggested that researchers and teachers implement plurality, complexity, hybridity, and unpredictability in study and practice. Connor (2005), on the other hand, noted that Kubota and Lehner had underestimated Contrastive Rhetoric, while failing to see how much it has given L2 writers in terms of rhetorical approaches and writing style substitutions of L1 and culture. Kubota and Lehner (2005) noticed patterns of growing diversity in Traditional Contrastive Rhetoric as discussed by Connor. They sensed that it, nevertheless, depended on cultural dissimilarity, which negatively affects L2 14 writers. When considering discourses in both traditional and critical Contrastive Rhetoric, there is a much debate. Contrastive Rhetoric research puts focus on L1 to L2 transfer following cultural dissimilarities, but it does not cover from L2 to L1 transfer in the same way (Mohamed & Omer, 2000). Because traditional Contrastive Rhetoric research studies examine relations between the L1 and L2 of bilinguals in a fragmented way, they are not considered to be comprehensive, not facilitating the two languages’ interactions. Raimes (1991) also refused Kaplan's robust hypothesis that language controls thoughts and rhetoric by adopting a low version, declaring that cultural background influences cognitive processes which, in turn, control rhetorical preferences. Therefore, rhetorical preferences are shown by L2 users, but writing in L2 may not essentially be the consequence of direct transfer of rhetorical patterns of rhetorical patterns from the L1 “but can be due to other cultural dimensions such as L1 literacy practices, writing functions, writing conventions, the frequency and distribution of different writing genres” (Ismail, 2010). 2.1.2 Summary Some researchers stereotype L2 learners who exhibit cultural dissimilarities in their patterns of academic switching between their L1 and L2. These investigators find it palpable that L2 English writing students are incapable of critical thought. Hence, culture becomes equated to interference when it is an aid to L2 learners. Concerning perspectives on theory, critical and traditional Contrastive Rhetoric investigators have different views on cross-linguistic transfer of L1 and L2 writing features due to their belief in the inadequacy of cultural dissimilarities. In contrast, bilingualism studies concentrate on the interrelationship of L1 and L2 writing knowledge, and consider it as an interdependent system. This multicompetence theory declares that students’ linguistic knowledge commingles in an ordinary language system since it is developed through L1 or L2. There, linguistic knowledge transfer could be bidirectional, changing both ways from L1 to L2 or vice versa (Jarvis & Pavlenko, 2008; Y. Liu & Carney, 2012). 2.2 Cognitive Theories Theories of bilingualism, multicompetence, interdependence and threshold will be discussed in this section. 15 2.2.1 Theory of Bilingualism Grosjean (2008) declared that bilingualism is “the regular use of two or more languages (or dialects)” and bilinguals are “those people who use two or more languages (or dialects) in their everyday lives” (p. 10). Grosjean (2008) offered two different views of bilingualism—a monolingual fractional view and a bilingual holistic view. The function of the former is to classify and assess bilingual users’ abilities in conjunction with two or more language proficiencies. In this way, bilinguals have distinctive but balanced proficiencies in their languages. Thus, bilinguals show keen abilities in two or more languages even when they use their language skills for different purposes and contexts. Behaving as monolinguals, these bilinguals have the power to adjust the activation of a language system as they deem appropriate. On the other hand, the bilingual wholistic approach does not necessarily see language users as proficient in two or more languages. Hence, due to their interlocution, these bilinguals may vary their language choice based on the appropriate purpose and context. A holistic and dynamic relationship exists among bilinguals who vary their languages while engaging with other speakers, regardless of dissimilar contexts and purposes. The proficiency of bilinguals to speak two or more languages has a connection with the purpose of language use in different contexts. Grosjean (2008), therefore, contended that these bilinguals can exhibit imbalances in their language proficiencies when the context for and purpose of using the languages changes. In Malaysia where English is a second language native-like accent may not be the aim since people employ English to facilitate communication with those who are not mainly from native English speaking backgrounds. People who do not have native-like accent and pronunciation can be considered bilingual and use English well. Content and purpose make up the appropriateness of language usage. Since bilinguals are not a combination of two monolinguals, it is difficult to determine the communicative competence of bilinguals. Together with social factors, there is a language repertoire that is whole based with regard to language skills. Moreover, wholistic bilinguals often use cross-linguistic transfer in either forward or reverse style as they process languages. Bilinguals’ language proficiencies can differ greatly from monolinguals’ due to context and purpose. 18 The monolingual concept of bilingualism tends to characterize bilinguals as deficient monolinguals, while multicompetence gives bilinguals credit for mastery of two or more languages. Metalinguistic awareness entails one’s ability to practise two or more languages according to Bialystok (1991); it was defined by Richards and Schmidt (2013) as the act of reflection and analysis of two or more languages to acquire linguistic knowledge. In other words, bilinguals are not imitating native speakers; however, they use their languages through reflections and perspectives. 2.2.3 The Interdependence Hypothesis The Interdependence Hypothesis states that bilinguals develop Common Underlying Proficiency (CUP) instead of having separate proficiency systems for their two languages (Cummins, 1981, 1996, 2001, and 2000). This common underlying proficiency makes possible the transfer of cognitive/academic or literacy-related proficiency from one language to another (Cummins, 2005). Cummins stated that although there may be differences in terms of surface features such as phonological and orthographic aspects, there are deep conceptual features shared by the two languages. These features involve abstract reasoning, reading/writing strategies, and prior knowledge usage. A bilingual’s CUP helps to achieve literacy skills transfer of languages. J. Cummins (1980) originally identified two different areas of common underlying proficiency (CUP), which he called Interdependence Hypothesis (IH): Basic Interpersonal Communicative Skills (BICS) and Cognitive Academic Language Proficiency Skills (CALP). From Cummins’ perception, Basic Interpersonal Communicative Skills involve cognitively undemanding manifestations of language proficiency in interpersonal situations. In contrast, Cognitive Academic Language Proficiency refers to the dimension of language proficiency that is related to literacy skills. This type of communication is mainly related to literacy based activities such as reading and writing (Cummins, 2000). Since CALP deals with higher order thinking processes , it is considered part of CUP, and consequently as transferable between languages. Meanwhile, BICS refers to surface features of language skills such as vocabulary and orthography in either L1 or L2, which are language specific and should be learned in the particular language as they are not transferable across languages. However, the level of achievement in these surface linguistic features (BICS) may facilitate the transfer of CALP across languages. In Cummins (2005) the Basic Interpersonal Communicative Skills and Cognitive Academic Language Proficiency 19 Skills distinction was broken into five more specific types: transfer of conceptual elements, transfer of metacognitive and metalinguistic strategies, transfer of pragmatic aspects of language use, transfer of specific linguistic elements, and transfer of phonological awareness. Conceptual aspects of writing such as audience, ideas, persuasive devices, cohesion and paragraphing require further research in an Arabic- English context (Al-Shekaili, 2011). The interdependence hypothesis was formally expressed as follows by Cummins (1981): To the extent that instruction in Lx is effective in promoting proficiency in Lx, transfer of this proficiency to Ly will occur provided there is adequate exposure to Ly (either in school environment) and adequate motivation to learn Ly. (p.29). Cummins (1981, 1996, and 2000) revealed that transfer of writing strategies is unlikely to happen automatically. Three conditions must be fulfilled for transfer: (i) there must be effective language instruction to advance L1 and L2 conceptual knowledge; (ii) learners need adequate exposure to language in both society and the classroom; and (iii) there must be a high motivation for language learning. Accordingly, the Interdependence Hypothesis stipulates that language instruction shapes the cross-linguistic transfer of literacy skills since CALP can be developed as a result of this transfer (Cummins, 1981, 1996, 2000). Students who receive effective L1 literacy instruction often strive to acquire higher L2 literacy skills. Thus, extensive L2 instruction in the light of the Interdependence Hypothesis facilitates learners’ skills and knowledge in L1 and L2. This was reinforced by Ahmadi, Khoii, and Taghadosian (2015) investigation into the impact of teaching L1 and L2 to first graders in a bilingual classroom separately to gauge their L1 literacy learning. This study, which took place in Iran, intended to compare the fluency and accuracy of the language of first grade Persian L1 and English L2 in a bilingual school with the fluency and accuracy of the language of first grade Persian L1 in a monolingual school during a timeframe of 32 weeks. The results showed that there were statistically significant differences between the bilingual and monolingual groups. The students in the bilingual school outperformed the students in the monolingual school in reading fluency. However, studies on bilingual students in higher education may offer different results and perspectives from those studies conducted with first graders. Bilingual students in 20 higher education already have well-formed literacy in L1 and have included themselves in more dissimilar social environments than first graders have. Further research is needed on bilingual students to better understand the possible influence of language writing features among bilingual students. Of additional importance is Cummins (2000) concept of additive bilingualism. He argued that to be successful, the learner must have adequate exposure to two languages where literacy is promoted such that the learners’ literacy skills can be transferred across languages. Cummins (2000, p. 37) defined additive bilingualism as, “the form of bilingualism that results when students add a second language to their intellectual tool-kit while continuing to develop conceptually and academically in their first language.” Thus, this contributes a second condition. (Sierens & Van Avermaet, 2014) similarly stated that additive bilingualism is “the acquisition of a second language without losing the skills acquired in the first language, because society appreciates and acknowledges both languages as being equal. Bilingual learners are empowered to enrich L1 literacy skills while they acquire L2 literacy knowledge through instruction and vice versa (Cummins, 2000). The wealth of research over past decades has clearly shown “Positive links between additive bilingualism and students’ linguistic, cognitive, or academic growth” (Cummins, 2000, p. 37). Moreover, EFL students have been found both to succeed in learning English language skills, and be more likely to maintain their mother tongue through additive bilingualism. According to Kecskés and Papp (2000, 2003), metalinguistic awareness develops as bilingual learners mentally compare and analyse their two languages’ literacy skills and/or knowledge in their minds, thus developing their literacy in both languages. However, important for the present study and L2 learning in general, is the difficulty for language learners to gain sufficient exposure to L2 language use. This is certainly applicable to EFL students in Arabic speaking countries, thus it is likely to be problematic for them to receive enough English exposure. This aspect needs further investigation for the impact of L2 on L1 writing. Nevertheless, for students in an English major at a university learning environment, where learning English is high stakes, their exposure to formal, key English language features may be supported in the learning environment in a variety of ways. Cummins (2000) argued that in a context where there is enough promotion of literacy in the two languages either in the classroom or in another social environment, there is no reason for language skills not 23 Many research studies support the plausibility of Cummins’ Threshold Hypothesis (e.g., Chen et al., 2010; Kecskés & Papp, 2000 and Kobayashi & Rinnert, 2007a). Chen et al. (2010) investigated how L2 (English) instruction helped to develop the phonological awareness of L1 in Chinese students (Grades 1 and 3). They discovered that positive transfer during L2 instruction was not automatic. This positive transfer occurred after the students had been taught L2 for two years. This period was helpful for the Chinese students who required enough time to achieve the threshold and transfer L2 phonological awareness to their L1 capabilities. Also, supporting the Cummins’ Threshold Hypothesis, a comparative study by Kecskés and Papp (2000) placed Hungarian high school learners into three classes where they were involved in different approaches to learning a foreign language for two years. They organized an Immersion class in which students learned French as a foreign language, which was the language of instruction in several subjects. They also held an Intensive class, involving seven to eight classes a week in English (L2) and Russian (L2), but these languages were not used to instruct in other school subjects. The third class was included as a Control group where the students were taught a foreign language (French)) for only 2 or 3 hours per week. It was found that students in the Immersion and Intensive classes demonstrated stronger writing abilities in their L1 compared to the Control group. It was also concluded that rigorous FL/L2 instruction during a two- year period allowed learners to gain conceptual knowledge in the FL/L2 and L2 proficiency for cross-linguistic transfer to L1. Cummins’ Interdependence Hypothesis and Threshold hypothesis are particularly related to the main aim of the present study, which attempts to find if bilingual Arabic L1/ English L2 students transferred the key language features they applied in their English scripts to their subsequent persuasive writing in Arabic. The following diagram shows the conceptual framework for the research. The conceptual framework was built on the concepts of the Interdependent Hypothesis and Threshold Hypothesis, which are considered in detail above. 24 Figure 2.2 an overview of the research conceptual framework of the study 2.2.5 Summary The purpose of this section has been to discuss cross-linguistic transfer according to the theories of Grosjean (1996, 2008); Cook (1992, 1999); and Cummins’ (1996, 2000). Based on these three theories, it is pertinent to note that the communicative competence of bilingual learners must be evaluated by different standards than those used for monolinguals, who process only one language. The linguistic knowledge of bilinguals is generated by their conceptualization of their languages. Thus, bilinguals’ L1 and L2 knowledge has the potential to influence both languages. It is possible for bilingual learners to exhibit patterns of transfer (i.e., L1 to L2 or L2 to L1) due to language competence/knowledge. Since L1 and L2 linguistic competencies are Bilingua l writer (1) Interdependent Hypothesis Effective L1and L2 Instruction Exposure to the L2 L1 Shared language features of the L1 and L2 texts Motivation in learning L2 L2 L2 Proficiency Out put (2) Threshold Hypothesis 25 interdependent, the development of one language may indirectly or directly promote the other. The aforementioned theories guided the rationale of this study, particularly the Interdependent Hypothesis (IH) and Threshold Hypothesis (TH) presented by Cummins (1981, 1996, and 2000). Cummins’ (2001) research is especially relevant to this study’s main purpose, which seeks evidence of reverse transfer (i.e., L2 English to L1 Arabic) for key language features during prolonged L2 instruction. The current study is needed for a better understanding of how learners’ L2 language features, developed through their L2 writing instruction, affect and develop their L1 language features. It also needed to give support or feedback concerning the viability of Cummins’ (1981, 1996, 2000) interdependence hypothesis and threshold hypothesis. 2.3 Differences between Writing in Arabic and English When comparing written texts in different languages, it is important to take into account the different styles of writing. In the case of Arabic and English languages, they belong to two dissimilar language families. Arabic is a Semitic language, while English is Indo-European. The method of Arabic writing originated from the Holy Quran, (Besston, 1970; as cited in Alnofel, 2003). Also, the Qur'anic Arabic influences the written form of the Arabic language (Ostler, 1987). Koch (1983) admitted that Arabic written language represents argumentative writing. Arabic argumentation is dependent upon notions described in a few words; whereas in the western style, writing focuses on logical structures. To distinguish Arabic from English, Feghali (1997) classified four features of writing. First, Arabic writing contains much repetition (Drid, 2014). At different levels of language, such as phonological, morphological and lexical, syntactic, and semantic, repetition occurs (Koch, 1983). According to Johnstone (1991, p. 117), “An arguer presents truths by making them present in discourse: by repeating them, paraphrasing them, doubling them, and calling attention to them.” Instead of using counterarguments, which are found in English persuasive writing, Arab writers choose to use repetition in order to make an argument (Kamel, 2000). Similarly, in discourse, many Arabic writers have been found to hide their goals because they take an indirect approach, which contrasts directly with the contemporary western focus on explicitness in the genre-based approach (Nagao, 2018). In Arabic, to comprehend the message, a receiver relies on the physical context which is internalized during interactions (Hall, 1966). Also, typically, Arab speakers use more 28 Conversely, some researchers believe that L2 learners make errors because of their limited knowledge of the target language (Fakhri, 1994 as cited in Barry, 2014). Ismail (2010) sought an answer to the problem of negative L1 transfer among Arab learners who showed problems with ESL writing, in reference to Kaplan’s (1966) argument, stating L2 students used rhetorical strategies from L1 to L2. Examining the same persuasive task, Ismail (2010) studied thirty ESL and Arabic speakers’ writing samples and thirty samples of writing from native English speakers. Learners’ L1 and L2 samples did not exhibit any significant relationship in terms of errors. Also, there was no significant difference in the rhetorical performance of native Arabic and native English speakers. Hence, the study points to discrepancies in Kaplan’s (1966) L1 negative transfer concept. In summary, compared to English, the style of Arabic in terms of rhetoric is quite dissimilar, particularly for persuasive writing samples. The occurrence of negative transfer from L1 to L2 is, therefore, probable (Kaplan, 1966). Due to this difference in English and Arabic rhetoric writing, Arab students of English often have difficulty mastering the English rhetorical style. If instructors ignore the differences, there is a heightened chance that the Arab learners will show increased failure rates in their English study. 2.4 Arabic Persuasive Writing Persuasive writing is a skill that convinces the reader about any idea or opinion that the writer presents in writing. Many writers have the skill to influence readers by using different ways of persuasion. Persuasive writers implement different techniques and methods to support their claim and also improve their arguments. Normally in the advertising sector, such writing skills are utilized to gain the attention of customers. For Arabic persuasive writing, a different type of style can be used to present arguments and is normally used at the international level to handle governance issues. These styles include presentational and analogical styles. They also utilize the quasi logical style that makes the arguments stronger and more understandable for the listeners or readers. For Arabic as a foreign language, the importance of writing as a learning task is very complicated because it includes different skills and processes that explain how to develop a plan for writing, writing behaviour, and composition review. The writing process needs a high ability to recognize different stages and parts. Arabic 29 is the second language so the second language writer has to face different unique challenges in writing. There are also different factors related to writing complexity within the context of EFL (Bakry & Alsamadani, 2015). Many social and cognitive factors are also included in Arabic writing. In the experience of writing many different emotions and intellectual aspects are also part of the writing. In addressing AFL for writing purposes, the instructions are individualized to meet the requirements for learning the different aspects of the language. Arabic persuasive writing is also essential for all those who deal with the Arabs through their organizations and who use different communication formats to deliver their information. Normally all the information is delivered in the English language but Arabic is also utilized according to state requirements. Arabic persuasive writing skills are helpful for the students to develop a strong writing model while using a different model and also utilize different sorts of information according to requirements. Arabic persuasive writing provides complete proofs and processes that support the arguments with strong evidence and also helps to explain the information in a proper format (Ismail, 2010). The Arabic language is also utilized at different state-levels according to country requirements and in international dealing, the Arabic language has its importance. Therefore, it is very important to learn the writing techniques according to Arabic persuasive writing because such writing skills are essential at different levels and stages of life according to working requirements. At the professional level, such a writing format is also helpful to develop a strong way to communicate at the international level. Arabic persuasive writing is a very important part of international studies and with the English language, Arabic language understanding is utilized in different formats according to the different requirements of people (Khalil, 1989; panelAzizKhalil, 2015). 2.5 Research Studies on Transfer of Writing Skills from L2 to L1 For L2 acquisition to take place, language transfer must occur. In SLA research, teaching, and classroom areas, the importance of language transfer has often been neglected. Despite the history of the notion, the language transfer process has been re- evaluated many times in recent decades. During the 1940s and 1950s, language transfer was researched by those in the linguistics field who noted the influence of 30 behaviourism on the process of habit formation. Accordingly, L1 habits easily influence L2 learning as a result of native language transfer. One main problem is that L1 interference can hinder those learning a second language (Fries, 1945). Fries (1945), a renowned behaviourist, also posited that a student’s native language should be carefully compared to his or her target language in order to meet L2 theory and pedagogy standards. Yet, some feel that a learner’s native language can cause failure in the learning of L2 (Lado, 1957). Lado (1957) went on to propose the Contrastive Analysis Hypothesis (CAH), which offered an explanation for the role of L1 in L2 learning. CAH denotes that L2 learners’ L1 patterns promote the development of productive and receptive skills. Further, one can predict the ease of L2 learning by observing similarities and differences between L1 and L2. This kind of theory suggests that L1 creates problems for students of new languages, and it contends that transfer is a unidirectional (from L1 to L2) process, rather than a bidirectional (from L2 to L1) one (Gass, 2013). CAH has a great deal of influence on early research studies concerned about contrastive rhetoric. In Kaplan’s (1966) study, students found it cumbersome to write in a direct, rhetorical fashion, and this difficulty was linked to the students’ native languages as well as their academics. This shows that cross-linguistic transfer of writing skills must be unidirectional since one must focus on the students’ L2 written rhetorical structures and why they deviate from monolinguals’ writing norms in English. Moreover, this does not emphasize how the students’ knowledge of general writing skills (L1 and L2) must interact. Recently, several studies have not supported the notion of unidirectionality in language transfer; instead, they have concentrated on L2 to L1 from the perspective of bidirectional transfer of language skills (Ko, 2017; Schjerve & Vetter, 2012). For example, Pavlenko and Jarvis (2002) contended that L2 can affect L1 while L1 influences L2. After assessing the oral narratives of a group of 22 Russians who had lived in America for three to eight years, Pavlenko and Jarvis (2002) observed the emergence of a cross-linguistic effect from L1 and L2 and L2 and L1. Talebi (2013) investigated the reading of learners in three languages and found that developments in the processes of reading in one language caused developments in another. In addition, Y. Liu and Carney (2012) examined persuasive essays written in both English and Chinese by high school and university students in China learning English. They 33 (1994) have shown that as student’s age, their higher level cognitive skills allows them to formulate stronger arguments. This represents the higher order thinking in terms of cognitive skills, metalinguistic awareness, and concept knowledge (Kecskés & Papp, 2000). When students receive effective L2 instruction, their L2 proficiency improvement indicates learning gains, which is factor one. This L2 proficiency mediates how well these learners transfer language skills from L1 to L2 (or L2 to L1). Research studies by Javadi-Safa et al. (2013), Xargia (2016) and Gonca (2016) corroborated the claim that L2 English writing success causes high abilities in students of L1 Persian, Greek, and Turkish writing. On the contrary, Wang (2014) and Cao (2016) found that Chinese EFL learners with advanced English proficiency are less likely to demonstrate reverse transfer of L2 English to L1 Chinese. Next, the foreign language environment represents factor two. Kecskés and Papp (2000) argued that when the instructional atmosphere is formal, then it is highly likely that foreign language transfer to students’ native language will occur. This is due to the focused attention given by the foreign language classroom to writing features, such as sentence structure and vocabulary. P. Liu and Ni (2016), Wang (2014), and Cao (2016) supported this idea and attempted to determine the transfer of writing skills from L2 English to L1 Chinese. In their studies, these researchers observed semantic, syntax, and discourse level in the L1 setting in conjunction with L2 users who formally studied English. The results revealed that exposure to an L2 cultural setting helps to influence the language system of L2 users, and an intensive L2 education in a dominant L2 setting can have an effect on L1. Wang (2014) and Cao (2016) offered the information that EFL students with intermediate English proficiency demonstrate reverse transfer (L2 English to L1 Chinese) in L1 settings, while L2 users are less exposed to L2 culture. By examining the writing of a Spanish L1 group, Maxwell-Reid (2010) studied argumentative Spanish texts composed by two groups of secondary students in an English Content and Language Integrated Learning program in an EFL environment. He then contrasted these texts with Spanish texts composed by learners of a Spanish curriculum. The groups had 24 students each. For data collection purposes, the participants’ written texts were used for methodological processing. As the Content 34 and Language Integrated Learning student texts had a great amount of English rhetorical aspects, many differences were found. These differences of text organization and clauses were predominant. Maxwell-Reid (2010) further asserted that L2 students’ writing skills, such as discourse and composition, develop in L2 and transfer to L1 writing effectively. Agheshteh (2015) noted the prominence of positive transfer of writing skills from L2 to L1. He investigated the effects of L2 English on Iranian bilinguals’ L1 writing skills by observing 61 participants—30 bilinguals and 31 monolinguals. For data collection purposes, he utilized an essay writing test taken in the participants’ L1. The findings of the test illuminated that Iranian bilinguals had stronger L1 essay writing abilities on the test than their counterparts, Iranian monolinguals. Agheshteh’s (2015) study is different from the present study in that it investigated the influences of L2 English on Iranian Bilinguals’ L1 writing ability, while the present study has examined the writing of Arab-speaking writers in an Arabic context. However, it is relevant to this study in that it framed this study mainly through the perspective of multi-competence and it also shares, in part, similar methodology with the present study in the use of text analysis. However, it differs in that the present study uses key language features such as audience, text structure, ideas, persuasive devices, vocabulary, cohesion, paragraphing, sentence structure, and punctuation, thus findings can be compared only to a limited extent. Akyel and Kamisli (1996) explored the Turkish and English writing skills of eight Turkish EFL students. They analysed the influences of EFL writing instruction for Turkish and English skill building and the learners’ attitudes about English and Turkish compositions. For data collection purposes, they utilized think-aloud protocols, student compositions, semi-structured interviews, and surveys. The study reflected that L1 and L2 writing had more similarities than differences. Moreover, L2 writing instruction had positive effects on the L1 and L2 strategies of these EFL students' Ll and L2 writing strategies and attitudes. In a similar study, KAYA (2013) looked closely at the transfer from L2 English to L1 Turkish. He explored the factors of writing rhetoric, introduction, body, conclusion and refutation, punctuation and coherence, unity scores of a pre-test in Turkish, an English test, a post-test in Turkish, and a delayed post-test in Turkish. All these factors were compared to find any significant differences between each pair’s scores. Accordingly, results revealed that 35 after intensive essay writing instructions in English, students’ writing improved in terms of coherence and unity as they wrote in Turkish in post and delayed post-tests. The participants began to use topic sentences and controlling ideas more effectively as they used connectives, which helped them go from one idea to the next. This raises a question of whether learners from different language families such as Arabic and English transfer writing strategies from L2 English to their L1 Arabic after they have practised intensive essay writing in L2 English. In preparation for the present study, it was difficult to find studies investigating the writing strategies used in English and Arabic persuasive essays written by tertiary Arabic learners of EFL. Sevgi (2016) wanted to study advanced-level language learners who employed similar cognitive strategies (Planning and Content Generation) for paragraph development in L2 (English) and L1 (Turkish). His study uncovered strong evidence of reverse transfer of writing skills. His study participants planned and wrote their paragraphs in L1 similar to the way they composed paragraphs in L2. Further, Uysal (2008) studied rhetorical patterns in Turkish and English essays, covering different subjects. The participants organized their Turkish essays in similar ways to how they organized their English essays. The main way they chose to organize these essays was to follow this pattern—thesis statement, explanation, and evidence for the argument. Focusing on Persian and English writing, Mazloomi (2011) examined how genre- awareness in the EFL essay writing classes influenced EFL learners’ L1 essay writing. Raising awareness of the structure of a five-paragraph English essay, 40 undergraduate junior students in two EFL classes went to eight sessions of treatment. The EFL learners’ essay writing improved greatly, with respect to the genre structure of five- paragraph essay compositions. Mazloomi (2011) concluded that genre-awareness of the English essay writing process significantly influenced the writing of EFL learners’ Persian essays. On the other hand, Mirzaee and Marzban (2016) did not observe any correlation between L1 (Persian) writing and L2 (English) writing. Similarly, Khodabandeh (2014), who observed Iranian students, focused on how bidirectional transfer takes place in English and Persian writings. He chose 104 sophomore students for participants. The participants were placed in three groups. First, some participants were randomly placed in an experimental group, receiving the 38 2.6 Summary There are controversial research findings in the studies reviewed above. L2 to L1 transfer of writing strategies has a positive, negative, natural or no positive and negative effect on L1 writing. However, there are mediating factors such as L2 proficiency, motivation, discipline in L1 and L2, age, foreign language environment, and effective L2 instruction affecting the transfer of L2 to L1 writing strategies. This finding also supports the main theoretical perspective of the current study that knowledge of writing is transferable between languages as long as efficient language instruction, adequate exposure to the L1 and L2, and an appropriate level of motivation to learn the language are achieved in the language learning context (Cummins 1996, 2000). 39 CHAPTER 3: Methodology Overview This chapter describes the research methodology of the study. It is divided into seven sections. The first section shows an overview of the chapter. The second describes the design of the study. Section three address the research method adopted for the current study. Section four gives details of the study’s participants. Section five provides an explanation of methodological triangulation. Section six describes data collection instruments. The last section discusses the ethical considerations. 3.1 General Design of the Study To conduct research effectively, one must properly develop a research design (Collis & Hussey, 2013), as the researcher needs to have a constant picture in mind with regard to the study; a research design must be visualized before it occurs. Thus, for this study, it was necessary to think about the purpose of the research, which was to examine the influence of L2 learning on L1. Specifically, this study’s aim, in terms of research design, was to discover if bilingual Arabic L1/ English L2 students transferred the key language features they applied in their English scripts to their subsequent persuasive writing in Arabic. By preparing a study, setting up the investigation, and accomplishing other tasks, research design is a crucial methodological process (Collis & Hussey, 2013). Collis and Hussey (2013) specified that effective research only takes place after a researcher has accurately chosen the research design. In the present study the researcher decided to combine both qualitative and quantitative methods to cross-validate the study findings. That is, the researcher analysed the results of each approach separately and then decided if the results from each method suggested the same conclusions. If they did, then the researcher’s confidence in the results and conclusion is strengthened. The researcher of the present study utilized an academic persuasive essay as a first method of data collection. The academic persuasive essay was used because it often involves organizing discourse by imposing a stepwise argumentation structure to a sequence of ideas, often through the use of persuasive devises (e.g. however, nevertheless, on the one hand,) (Uccelli, Dobbs, & Scott, 2013). Likewise, an academic persuasive essay goes beyond expressing emotions or reactions to events and requires that writers mark their stance towards specific ideas (Berman & Nir-Sagiv, 2004). Stimulated Recall 40 Interview was a second method of data collection. The Stimulated Recall Interview was used in order to examine students' cognitive abilities or to identify the writing processes and key language features they employed while writing the persuasive essays in both languages. A self-administered data collection method, where surveys were personally handed to the participants one by one, was used as the third method of data collection. The self-administered data collection method was selected for four reasons. In the first place, the collection of self-administered data can facilitate a high response rate and leverage quality (Uma Sekaran & Bougie, 2016). Secondly, it is a fast and versatile process (Grossnickle, 2001). Finally, it is highly confidential as participants do not need to reveal their identities (Burns, 2002). 3.2 Research Method Adopted for the Current Study To achieve the main aim of this study and answer the research questions, a mixed methods approach (qualitative and quantitative) study was performed. By adhering only to the qualitative method some important issues could possibly be overlooked. Thus, this study also employed the quantitative approach to strengthen the results. This was clearly advocated by Riazi & Candlin, 2014; Riazi, 2017. Although many studies exist on the topic of writing skills transfer from L1 to L2, this research study focuses on a new direction–the transfer of persuasive writing features from L2 English to L1 Arabic. To make the results more comprehensive, the researcher used qualitative and quantitative methodologies for study refinement. The process of mixing methods is known by several names, including blended research, triangulated studies, multi-method, and ethnographic residual analysis (Harrison & Reilly, 2011). Johnson, Onwuegbuzie, and Turner (2007) stated that the combination of quantitative and qualitative methods in one study has gained positive scholarly recognition. Creswell and Clark (2007) specify: Mixed methods research is a research design with philosophical assumptions as well as methods of inquiry. As a methodology, it involves philosophical assumptions that guide the direction of the collection and analysis of data and the mixture of qualitative and quantitative approaches in many stages in the research process. As a method, it focuses on collecting, analysing, and mixing both quantitative and qualitative data in a single study or series of studies. Its 43 3.2 Participants in the Study The research design required the researcher to select three groups of students from the higher education sector where undergraduate students’ first language was Arabic. The three groups of students involved were: Group1- G1AdvEL2/AL1 - Advanced English as a second language/Arabic as first language; Group2- G2IntermEL2/AL1 - Intermediate English as a second language/Arabic as first language; and Group3- G3ArabicMonoL1 - Arabic as first language monolingual. Participation in the study was on a voluntary basis. The first and second groups were fourth-year English majors and the third group was fourth-year Arabic language majors at an Egyptian university in Cairo. It was expected that transfer of the key language features would be most visible for the first and second groups because the students had already completed three years of English L2 writing instruction at university level. The fourth-year students were expected to provide more persuasive appeals in their essays, because they should have taken most of their university writing courses, including being taught the persuasive type of discourse (Cahyono, 2004). Y. Liu and Carney (2012) also stated that L2 speakers master writing persuasive strategies at more advanced developing stages because of their higher L2 proficiency in writing. Overall, 290 students participated in the study. There were 250 bilingual Arabic L1/ English L2 fourth-year undergraduate students and 40 monolingual Arabic fourth-year undergraduate students. Of 290 students, 96 were the main participants in the first stage (English and Arabic writing tasks) of data collection. However, 16 students were excluded from the first stage of data collection since their essays had only about 5-6 sentences (35-45 words). These students’ exclusion was based on the fact that their writing was not suitable for study purposes. In other words, the written texts of these students did not exhibit the words of meaningful discourse. However, these 16 students completed the surveys at the third stage of data collection. Therefore, of the 80 students who participated in Stage one of the study, 40 were English majors and 40 students were Arabic language majors. Based on the students’ self-evaluation and their English language teachers’ report of their English language proficiency to discover whether English proficiency impact the participants’ essays overall quality., the 40 English students were divided into two proficiency groups to discover whether English proficiency affect the participants’ essays overall quality.. Twenty participants were classified as G1AdvEL2/AL1, and 20 as G2IntermEL2/AL1. The evaluation was 44 conducted in the first meeting after the students signed the participation agreement form. Three groups of students involved: G1AdvEL2/AL1, G2IntermEL2/AL1 and G3ArabicMonoL1. An overview of student samples contributed in the first stage of data collection are presented in table 3.2. Table 3.2: Overview of student samples, group nomenclature and distribution of written scripts in English and Arabic Student Groups Written text language English L2 Arabic L1 Group1- G1AdvEL2/AL1 x x Group2- G2IntermEL2/AL1 x x Group3- G3ArabicMonoL1 x In Stage 2, interviews were conducted with participants using saturation technique which opt for getting considerably diverse concepts from even a small sample (Guest et al., 2006, and Kvale, 1996). Following this technique, this study conducted interviews with four participants to get the perspective of the effect of the L2 on their L1, two from G1AdvEL2/AL1 and two from G2IntermEL2/AL1). In Stage 3, of the overall 290 students, 250 bilingual Arabic L1 / English L2 fourth-year undergraduate students, including the 40 English majors who participated in the first stage of the data collection, were the main participants. The researcher removed 41 surveys because of a great deal of missing data. Therefore, 209 valid surveys were left to be analysed. 3.4 Methodological Triangulation Neuman (2003) defined methodological triangulation as the combination of two or more research methods. The goal of methodological triangulation is to look closely at something from different angles (Neuman, 2003). For the production of rich results, the use of methodological triangulation makes research studies stronger (Greene & Caracelli, 1997). According to Spiggle (1994), four kinds of methodological triangulation exist. In the first model, the triangulation of measures utilizes more than one measure in the presence of the same phenomena. Next, in the second model, a triangulation of observation occurs as data is observed according to different fields. In the triangulation of theory, which is the third model, multiple theoretical perspectives take place. The fourth and final model is referred to as the triangulation of methods, 45 where the two methodologies are utilized equally; that is, each method works as part of a multimethod approach that tests a specific phenomenon at several levels. First Model. Qualitative research precedes quantitative measurement and the development of instruments Second Model. Quantitative methods are mostly used to complement qualitative ones. Third Model. Qualitative methods are used to explain quantitative research. Fourth Model. Qualitative and quantitative methods are used as equally valuable and simultaneously Figure 3.1 examples of the implementation of methodological triangulation in research (Stickler et al., 1992, p. 5) Tashakkori and Teddlie (2003) argued that the selection of triangulation models relies on the research method and research questions. As Figure 3.1 above indicates, data from a research study using both quantitative and qualitative methods uses mixed methods. The researcher of the present study decided to apply the fourth model of triangulation, when the two methodologies are used equally. The researcher has explicitly selected a multimethod approach, combining both qualitative and quantitative methods to involve as many aspects of the study purpose as possible. Qualitative Quantitative Qualitative Quantitative Results Results Quantitative Qualitative Results Qualitative Results Quantitative 48 Table 3.3: Proportion of script per category based on overall rating Overall category rating Number of scripts % Superior 7 17.5 Satisfactory 24 60.0 Developing 9 22.5 40 100 As shown in Figure 3.3, the scoring ranges for each of the nine key language features identified in the marking guide (NAPLAN, 2011) were used to score the students’ English and Arabic essays. For example, while the range of scores for ‘audience’ was between 0 and 6, that of ‘paragraphing’ where there is a lesser score, was between 0 and 3. According to Becker (2011) scoring rubrics “can be used to indicate how well a student has achieved mastery of aspects of L2 writing” (p. 114). However, for comparative purposes, each of the NAPLAN scores on the scale relevant to each category was converted to a scale of 1 to 10 (see Figure 3.4 for more details). The 40 English essays which were produced by bilingual Arabic L1/ and English L2 students (G1AdvEL2/AL1 and G2IntermEL2/AL1) were scored by the researcher, who applied the NAPLAN marking criteria based on the NAPLAN marking guide that provides detailed descriptions of the various levels of performance of each language feature with accompanying samples of students’ writing. The random sample of the English scripts were marked by the experts in the field. The 80 Arabic essays which were produced by the 40 bilingual Arabic L1/ English L2 students (G1AdvEL2/AL1 and G2IntermEL2/AL1) and 40 fourth-year monolingual Arabic majors (G3ArabicMonoL1) were scored by two Arabic speaking markers. The first was the researcher (M1) of this study, who is bilingual Arabic (L1) and English (L2) and the second, a monolingual Arabic speaking teacher (M2). This teacher had a postgraduate degree in Arabic education and literature from an Arabic university. In addition, he had more than five years of teaching experience. Training for him on scoring the Arabic essays was conducted four times, using a discussion format before the Group 3 participants’ essays were independently marked by M2. T-test analysis was conducted to compare mean differences of the key language feature scores of the English and Arabic essays produced by the two groups (G1AdvEL2/AL1 and G2IntermEL2/AL1). The purpose of the comparison was to discover whether students who included the key language features (audience, text structure, ideas, persuasive devices, vocabulary, cohesion, paragraphing, sentence structure, and 49 punctuation) in the English texts were also those who employed these features in the Arabic essays. The t-test was also used to test whether the mean and standard deviation differences of the L1 Arabic essay produced by G1AdvEL2/AL1 and G2IntermEL2/AL1 versus G3ArabicMonoL1 in the nine key language features were statistically significant. The purpose was to find whether the L1 Arabic essays written by English majors (G1AdvEL2/AL1 and G2IntermEL2/AL1) were higher or lower than those produced by monolingual Arabic majors (G3ArabicMonoL1) in the nine key language features. Figure 3.3 NAPLAN persuasive writing scoring criteria and range of scores. Source: (NAPLAN, 2011). 0 1 2 3 4 5 6 0.0 1.7 3.3 5.0 6.7 8.3 10 Developing Developing Developing Satisfactory Superior Superior Superior 0 1 2 3 4 5 0.0 2.0 4.0 6.0 8.0 10.0 Developing Developing Developing Satisfactory Superior Superior 0 1 2 3 4 0 2.5 5 7.5 10 Developing Developing Satisfactory Superior Superior 0 1 2 3 0.0 3.3 6.7 10.0 Developing Satisfactory Superior Superior Figure 3.4 scale conversion of the NAPLAN scores 50 Figure 3.5: research design to analysis of the respective key language features across the various student groups’ written persuasive texts 3.5.2 Phase 2: Stimulated Recall Interviews Easterby-Smith, Thorpe, and Jackson (2012) claimed that personal interviews must “probe deeply, to uncover new clues, open up new dimensions of a problem and to secure vivid, accurate inclusive accounts that are based on personal experience” (p. 131). Interviews provide researchers with the advantage of being able to ask for details and to develop follow-up questions. To put focus on L2 to L1 key language features transfer, the researcher of the present study chose stimulated recall interviews as a data collection technique. To better understand writers’ preferences for using a particular features in their L1 and L2 essays, a stimulated recall interview should be conducted as soon as possible after the task is finished (Fox-Turnbull, 2009). It is well-known Group 1 – G1 AL1/L2English- Advanced Level E English L2 Persuasive text Arabic L1 Persuasive text EnglishL2/ ArabicL1 bilingual persuasive writing Group 2 – G2 AL1/L2English- Intermediate Level Arabic L1 Persuasive text English L2 Persuasiv e text Baseline of Arabic L1 monolingual persuasive writing Group 3 – G3MonoL1 Arabic Arabic students Persuasive texts 53 study, EFL university students (more than 50 participants) responded to the survey. The researcher made no significant changes to the survey, with the exception of revising word choice for clarity’s sake. The researcher of this study utilized a self- administered survey and collected the data from the English Education Department. The next sections detail the survey of the G2IntermEL2/AL1 and G1AdvEL2/AL1 groups. 3.5.3.1 The Students’ Survey Structure A modified version of a survey developed by both Uysal (2008) and Rinnert, Kobayashi, and Katayama (2015) to elicit information about the participants' L1 and L2 writing backgrounds was adopted for the study. The survey given to students (presented in Appendix D) consisted of seven parts. Part One collected demographic information. Parts Two and Three asked participants about writing instruction in Arabic and English. Part Four sought to identify students’ personal attitudes toward writing; Part Five asked students about exposure to writing, whether in the classroom or in society. Part Six was designed to allow the participants to self-report about their English proficiency. Part Seven was designed to find out if the participants believed English writing instruction had an effect on their Arabic writing. 3.5.3.2 Response Rate A total of 250 surveys were distributed to higher education male and female students who were majoring in English and had been invited to complete the survey over the seven week period between November 16, 2016 and January 8, 2017. The researcher utilized a self-administered method, which entailed distributing the surveys to the respondents and watching them as they completed the surveys. Because of problems such as inaccuracy and incompleteness, the researcher chose to eliminate 41, leaving 209 returns, which was an 83% response rate for the study, as shown in Table 3.4. Table 3.4 Response rate in this study Total surveys hand out 250 100% Total respondents in this research 209 83.6% Invalid surveys 41 16.4% Valid surveys in this research 209 83.6% 54 3.5.3.3 Missing Data Missing data often occurs in research studies and can demonstrate statistical analysis errors (Duffy, 2006; Nguyen & Tsoy, 2017; Oketch, 2017). Having missing data is highly problematic even for small samples. Some examples of the repercussions of missing data include loss of sample representativeness, unbiased estimation, and exaggeration in variance and error in finding estimates of true values. When data is missing, the ability of the researcher can be reduced to the point where he or she cannot make accurate choices concerning the subject matter (Dong & Peng, 2013; Duffy, 2006; Oketch, 2017). Missing data typically occurs because of partially answered or completely unanswered questions. When surveys are missing, data is more likely to occur. For this reason, researchers in social science research have two alternatives: they can either omit data from their analyses, or use an estimation method to track missing data. Stevens (2012) advised the replacement of missing data by the use of mean scores on the variance. In addition, Norušis (2006) proposed that the problem of missing data can be solved when one removes a sample(s). Moreover, Hair, Black, Babin, Anderson, and Tatham (2006) contended that missing data can be classified in terms of ignorable data. As a research design method, ignorable data is characterized by data that is unanswered. For instance, in question number one, participants had to skip the next question if the answer was ‘no’. for that reason, the skipped question (number two) must be ignorable missing data. Furthermore, if the missing data score is less than 5 percent, then missing data is not a substantial problem (Tabachnick and Fidell (2007). Taking all of these points into consideration, the researcher decided to omit surveys that had missing data. 3.5.3.4 Sample Size and Data Collection Process Phase 3: Survey Concerning the size of the research sample, details are shown below. Bryman and Bell (2003) asserted that the two main aspects to be considered in the selection of samples are time and cost. Thus, one must carefully adopt a timeframe that is applicable to the given methodology if quality research is to be produced. With a self-administered data collection method, the research can be argued to produce quality research and sufficient data. Further, 250 respondents, the target sample size, were allowed to ask questions. According to the literature, self-administered surveys achieve nearly 100% response rate (Uma. Sekaran, 2000). Making an estimate, the researcher found that a 55 response rate of 100% would result in a sample of n = 250. The 250 were bilingual Arabic L1/English L2 fourth-year undergraduate students aged 20-24. They had learned English for more than three years at university, and their English proficiency was at intermediate and advanced levels. Thirty-six of them completed the first and third stages (writing takes and survey) of data collection and four of them completed all stages (writing takes, interviews and survey) in the study. The rest (210 participants) were joining only the third stage (survey) during the data collection process. The surveys were given at the end of the Arabic session to be completed at home and brought back in the next day, as time did not permit for completing the writing essay and survey in a one session. After collecting 250 surveys from 51 male and 158 female students, the researcher chose to remove 41 returns because of a great deal of missing data; therefore, 209 valid surveys were left to be analysed. The focus of the data for the main survey depended on Arabic and English writing instruction, personal attitudes about writing, general writing exposure, English language level, and learners’ perception about how L2 English writing instruction affected L1. Each survey had twelve pages and a cover page. Consequently, the researcher selected 209 completed surveys from the total 250 distributed surveys. Table 3.5: The profiles of the participants (N= 209) Frequency Percent Country: Egypt Other Country 209 0 100.0 0.0 Gender: Female Male 158 51 75.6 24.4 Current Education: Bachelor (B.A) Master (M.A) Doctor (Ph.D.) 209 0 0 100.0 0.0 0.0 Literary Strand in Secondary School: Male Female Science Strand in Secondary School: Male Female 34 127 17 31 16.3 60.8 8.1 14.8 Major at university: 58 language features of Audience, Text structure, Ideas, Persuasive devices, Vocabulary, Cohesion, Paragraphing, Sentence structure and Punctuation. The first section of this chapter presents an overview of the scores and percentage of use of key language features in the English scripts, comparing the results for G1AdvEL2/AL1 and G2IntermEL2/AL1. The second section reports the scores and percentages of use of the nine key language features in the Arabic scripts written by these two groups of students to discover to what extent those who had high scores in the application of key language features in their English essays were the same as those who received high scores in their Arabic essays, and those who scored low in their English essays were also those who scored low in their Arabic essays. Section three reports the research results in relation to the main goal of the thesis, that is, the investigation as to whether there is evidence for reverse transfer of key language features from written English to Arabic scripts for G1AdvEL2/AL1 and G2IntermEL2/AL1. The analysis includes scores for the use of key language features in Arabic scripts written by G1AdvEL2/AL1, G2IntermEL2/AL1 and G3ArabicMonoL1. 4.1 Comparison of G1AdvEL2/AL1’ Performance versus G2IntermEL2/AL1’ Performance on Writing a Persuasive Genre in English This section used the same method as O’Neill’s (2011) study of persuasive writing assessment. Based on overall impression, the forty English scripts were divided into three categories after reading and marking. Seven of these were Superior (17.5%), twenty-four Satisfactory (60%) and nine, developing (22.5%) (Table: 4.1). None of the scripts fully deviated from the main topic or genre. 59 Table 4.1: Proportion of script per category based on overall rating Overall category rating Number of scripts % Superior 7 17.5 Satisfactory 24 60.0 Developing 9 22.5 40 100 Scoring for the nine categories adopted from NAPLAN (2011) were used to measure the overall quality scores of the students’ English and Arabic essays. Since the marking of each key language feature according to the NAPLAN guide allocates the performance to different levels across the set of nine categories, for comparative purposes, ratings were converted to a scale of 1 to 10. 4.1.1 The Use of Key Language Features in the English Essays – G1AdvEL2/AL1. Table 4.2: percentages and scores for key language features which belong to G1AdvEL2/AL1, as identified in their English essays Number of scripts Original Score New Score Overall rating Percentage 0 - 6 0 - 10 Audience 7 3 5.0 Satisfactory 35 9 4 6.7 Superior 45 4 5 8.3 Superior 20 20 100 0 - 4 0 - 10 Text structure 16 3 7.5 Superior 80 4 4 10.0 Superior 20 20 100 0 - 5 0 - 10 Ideas 4 3 6.0 Satisfactory 20 16 4 8.0 Superior 80 20 100 0 - 4 0 - 10 Persuasive devices 13 3 7.5 Superior 65 7 4 10.0 Superior 35 20 100 60 0 - 4 0 - 10 Cohesion 15 3 7.5 Superior 75 5 4 10.0 Superior 25 20 100 0 - 3 0 - 10 Paragraphing 7 2 6.7 Superior 35 13 3 10.0 Superior 65 20 100 0 - 5 0 - 10 Vocabulary 8 3 6.0 Satisfactory 40 8 4 8.0 Superior 40 4 5 10.0 Superior 20 20 100 Sentence structure 0 - 6 0 - 10 7 4 6.7 Superior 35 9 5 8.3 Superior 45 4 6 10.0 Superior 20 20 100 Punctuation 0 - 5 0 - 10 6 3 6.0 Satisfactory 30 9 4 8.0 Superior 45 5 5 10.0 Superior 25 20 100 4.1.1.1 Audience According to NAPLAN (2011), the skill focus of ‘audience’ means to orient, engage, and persuade the reader. Writers with a score range of 6.7, 8.9 and 10 (Superior) may orient, engage and persuade readers successfully, whereas those scoring 3.3 (developing) may only orient and support the reader in understanding the script. Writers who score 0 and 1.7 (also developing) may only respond to the audience’s needs by conveying the message either in the form of simple text or symbols. The score of 5 was allocated to writers with a Satisfactory script. It was found that the majority (65%) of G1AdvEL2/AL1 were able to engage the reader by providing detailed information on a range of situations in which the benefits of learning a language outweigh playing sport, particularly by acknowledging a wider audience; and they were labelled Superior. In contrast, 35% of the participants could only orient the 63 for Superior. The Developing stage of a writer’s script ranges from no evidence, to one or two instances of persuasive devices that may also be of the same type. The opinion is either confused or satisfactory and persuasive devices used are very simple such as ‘I think’, ‘I reckon’, ‘because’ etc. The Satisfactory stage uses three or more instances of persuasive devices to support the writer’s position; however, the types of devices used are generally ineffective. The use of persuasive devices at the Superior stage range from effective but not sustained, to effective and sustained, which can genuinely persuade the reader (NAPLAN, 2011). From a total of 20 scripts, 13 (or 65%) were rated 7.5, and 7 (35%) received the highest rating of 10. Therefore, overall, all the scripts were rated Superior. Extract numbers 1, 3, 6, 7, 17 and 19 below, show the more sustained effective use of persuasive devices. Personal opinion (I share the view), conditional mood, (if you know the language of the intended country during touring, you will be able to overcome some communicative difficulties . . .), emphatic statements, authoritative statements, repetition (distant civilisations/different nationalities, learning language/bilingualism) and modality (you will be able to overcome). (Bilingual_Adv_P1) Figurative language (“People are similar to engines and motors”), direct address of the reader (constant use of “we” and “you”), emphatic statements, authoritative statements, repetition (foreign language/different nationalities) and modality (“If we don’t do sports we will be forced to rust”). (Bilingual_Adv_P3). Statements of varying intensity are juxtaposed effectively to support position (“However its benefits, sports need much effort, need a lot of money . . . Not all people can do that”, “We should learn new languages and seek knowledge if we want to be respected by others.”. (Bilingual_Adv_P6). Reference statements (“medical studies have shown”), authoritative statements (“companies who are looking to expand into overseas markets are constantly looking for . . .”), and repetition (sport/discipline/competition). (Bilingual_Adv_P7). Conditional mood, authoritative statements (“Even those that say they don’t care…will have noticed these problems”), emphatic statements (“Therefore it helps with career enhancement”) and modality (“. . . but this is simply not the case anymore . . .). (Bilingual_Adv_P17). 64 Personal opinion (“In my opinion”), emphatic statements (“Thus the benefits of sport could be replaced by other practices”), authoritative statements (“This can be achieved through challenging the brain’s power of thinking”), and reference statements (“Dr Ludwig Wittgenstein . . .”). (Bilingual_Adv_P19). 4.1.1.5 Cohesion The skill focus of cohesion concerns the control of multiple threads and relationships across the text, achieved through the use of referring words, ellipsis, text connectives, substitutions and word associations (NAPLAN, 2011). On the scale of 0 to 10, the scores of 0 and 2.5 were allocated to Developing, 5 to Satisfactory, and 7.5 and 10 to Superior scripts. At the developing stage, writers may only use symbols or at most short script that is likely to confuse readers. However, satisfactory scripts may contain some correct links between sentences with accurate usage of referring words or longer sentences with partially controlled cohesion. On the other hand, Superior scripts use well controlled or a range of cohesive devices with clear meanings and continuity of ideas (NAPLAN, 2011). It was found that 75% scored 7.5 and 25% 10, making an overall rating of Superior for the 20 selected essay scripts. See extracts numbers 1, 4, 7 and 19. Connectives (Moreover, however, as, secondly, which in return), substitution (it makes him acquire) and word associations (language/bilingualism, cultures/customs and traditions, foreign machines/communication technologies, travelling abroad/touring). (Bilingual_Adv_P1) Connectives (“Furthermore”, “Therefore”, “Although”. and word associations immigration/integrate”. (Bilingual_Adv_P4). Connectives (Furthermore, thirdly), substitution and word associations (travel/translation, culture/perspective, fate/competition/physical. (Bilingual_Adv_P7) Connectives, “Moreover”, “However”, “Thus”. Word associations include “expert/philosopher”. (Bilingual_Adv_P19). 65 4.1.1.6 Paragraphing The skill focus of paragraphing is about the segmenting of texts into paragraphs that assist the reader in following the line of argument. On the 0 to 10 scale, the score of 0 was labelled Developing, 3.3 as Satisfactory, and 6.7 and 10 as Superior. The three Developing essay scripts may have no correct use of paragraphing. At the Satisfactory level, the writing is organized into at least one correct paragraph with focus on one, or a set of ideas that assist the reader in understanding the text. At the Superior stage, paragraphs are logically constructed with topic sentences and details supporting the argument (NAPLAN, 2011). Based on such criteria, 35% scored 6.7, while 65% scored 10, thus achieving an overall rating of Superior. 4.1.1.7 Vocabulary The skill focus of vocabulary is about the range and precision of contextually appropriate language choices. On the scale of 0 to 10, the scores of 0, 2, and 4 were assigned to Developing, 6 to Satisfactory, and 8 and 10 for Superior. The Developing essay script may use symbols or drawings, a very short script, and mostly simple words, including two or three precise words or word groups. Satisfactory scripts must have at least four precise words or word groups such as ‘citizen’ or ‘duty of care’ etc. Superior scripts demonstrate sustained and consistent use of precise words and word groups. Superior scripts which scored 10 did not contain any inappropriate or inaccurate word choices and the language style was well-matched to the style of argument (NAPLAN, 2011). Out of a total of 20 scripts, 40% (or 8 scripts) were rated Satisfactory, 20% scored 10 and 40% scored 8. Therefore, 60% scripts achieved an overall rating of Superior for G1AdvEL2/AL1. See extract numbers 7, 11, 17 and 19 below. (“cauldron of rich . . .”, “brain-related disease . . . Alzheimers and dementia”). (Bilingual_Adv_P7). (“Prosperity”, “refute”, “neglect”). (Bilingual_Adv_P11). (“Specifying, refute, immigration, integration”). (Bilingual_Adv_P17). (“multilingual/monolingual/polyglot”, “crucial”, “furtherance”). (Bilingual_Adv_P19). 68 Punctuation 0 - 5 0 - 10 2 2 4.0 Developing 10 16 3 6.0 Satisfactory 80 2 4 8.0 Superior 10 20 100 4.1.2.1 Audience The G2IntermEL2/AL1 scored 5.0 and 6.7 for Audience on a 0 to 10 scale and were given an overall rating of Satisfactory and Superior respectively. Out of a total of 20 participants, 75% achieved Satisfactory and 25% Superior script writing skills. Therefore, the majority of the G2IntermEL2/AL1 were shown to have Satisfactory skills, meaning that these writers could orient the reader with evidence of persuasive text. The readers were able to follow the text fairly easily. In comparison, Superior writers with Superior skills could support reader understanding and begin to engage and persuade the reader through language choices (NAPLAN, 2011). The extracts below are representative of how the majority of participants (75% satisfactory scripts) introduced their writing. They could orient the reader; however gaps existed in the information due to poor sentence structure. There is an attempt to support the reader, however not enough consistent elaboration. Better examples include: “But as language learners, there are some important things we can learn from top athletes”; “a foreign language is more than just a boost to your CV”. (Bilingual_IntM_P21) “The number of people that visit France annually . . .”, “It is also important for making real connections” (Bilingual_IntM_P22) “We can recognise and be aware of other cultures . . .” “. . . thinking that people who learn a language imprison themselves in their room.” (Bilingual_IntM_P26) 4.1.2.2 Text structure On the scale of 0 to 10, 40% of G2IntermEL2/AL1 scored 5.0 and the remaining 60% scored 7.5. Therefore, the overall rating of Superior was given to the majority and Satisfactory to the remaining writers, meaning that the majority of writers were able to produce introduction, body, and conclusion in the text structure. They were also 69 able to detail longer texts that may have contained one weaker component. The writers were able to develop the body with reasons and supporting evidence (NAPLAN, 2011). In contrast, satisfactory writers either missed one of the components of introduction, body and conclusion or all were presented in a weak manner and needed further development (NAPLAN, 2011). 4.1.2.3 Ideas On the scale of 0 to 10, 70% of participants cored six and were identified as Satisfactory; only 30% scored 8.0, obtaining an overall rating of Superior. Therefore, mostly scripts written by the G2IntermEL2/AL1 were Satisfactory. These writers supported their ideas, showing at least one idea with more developed elaboration. On the other hand, the Superior scripts not only elaborated on ideas but also stated issues that were for and against the stated position (NAPLAN, 2011). The following extracts are representative of the way G2IntermEL2/AL1 students introduced their ideas. Ideas are sometimes elaborated, mostly contributing effectively to the writer’s position and occasionally reflecting on wider world issues, such as: “Some people have more aptitude for learning languages including children . . .”, “Learning a language helps in raising a good citizen” or “If you have to be in the class, you may as well learn the language”, “If you understand the language and culture then . . .”, “. . . We’ll be aware of other countries’ culture and tradition” or “They think that practising a sport may enable the person to be a champion or a famous man”. These ideas are not strong enough for a score of 8 (or Superior level) due to some ineffective/unelaborated/confusing arguments. Indeed, Bilingual_IntM_P21 argued that “Require high fitness to continue to exercise as well as respect for learning language”, “as is the language of science and technology and scientific research”. 4.1.2.4 Persuasive devices Most scripts of G2IntermEL2/AL1 were scored 5 for their use of persuasive devices. On the scale of 0 to 10, 85% scored 5.0 and were rated Satisfactory; 15% scored 7.5 and were rated Superior. The Satisfactory scripts had three or more instances of persuasive devices that supported the writer’s position; additional devices were found ineffective. In contrast, Superior writers used devices to persuade but did not sustain the effectiveness as they were not able to address the reader’s reason, values or emotions (NAPLAN, 2011).The following extracts are representative of how the 70 majority of G2IntermEL2/AL1 participants (those rated Satisfactory) introduced the persuasive devices in their English essays. Bilingual_IntM_P21 scored 5. He used various persuasive devices but these devices were not effective for a score 7.5. He used personal opinion (“we”), modality “shouldn’t let it discourage us from continuing to improve” and a value statement “Acquiring a second language enables us to develop various mental abilities at all ages”. Bilingual_IntM_P21 scored 5. She presented ideas in her writing but with not enough elaboration consistently. e.g. “France has been the number one tourist destination . . .”. The reader then has to make a supposition as to why learning a language is important. She also used personal opinion “you”, modality: “other languages will help when. . .”, and a value statement: “If you understand the language and culture then you will be prepared . . .”. Bilingual_IntM_P26 scored five. She used personal opinion such as “we”, modality: “If you are a student, it’ll improve your mental abilities . . .”, and a value statement, “. . . if you walk thirty minutes a day, that is enough”. 4.1.2.5 Cohesion On the scale of 0 to 10, 60% of G2IntermEL2/AL1 scored 5 and were rated Satisfactory, whereas 40% scored 7.5 and were rated Superior. Therefore, the majority of G2IntermEL2/AL1 were rated Satisfactory. These scripts had some correct links between sentences, with mostly accurate wording. However, such scripts contained simple connectives and needed correction after re-reading. The Superior scripts showed controlled use of cohesive devices and sustained evidence of text flow within a piece of writing. Connectives such as ‘however’, ‘even though’, ‘finally’ and others were also used appropriately. The following extracts are representative of how the majority of participants (those rated Satisfactory) introduced their writing: The participants showed sound control of the use of simple connectives as shown by the following examples: “if”, “there are” (Bilingual_IntM_P22) “If” “and”. ” (Bilingual_IntM_P24) “If”, “It also” (Bilingual_IntM_P25) 73 to measure the mean difference of the nine quality features of essays between the two groups. The P-value test of the nine elements needed to be less than .05 to be statistically different. Table 4.4: The mean comparison of quality scores for G1AdvEL2/AL1 and G2IntermEL2/AL1 English essays Category G1AdvEL2/AL1 (N=20) G2IntermEL2/AL1 (N=20) P-value (2 tailed) M SD SE M SD SE Paragraphing 8.8450 1.61489 .36110 6.3600 1.04650 .23400 .000 Persuasive devices 8.3750 1.22340 .27356 6.8750 1.11065 .24835 .000 Cohesion 8.1250 1.11065 .24835 6.0000 1.25656 .28098 .000 Sentence structure 8.0800 1.22500 .27392 4.8300 .76026 .17000 .000 Text structure 8.0000 1.02598 22942 6.5000 1.25656 .28098 .000 Punctuation 7.9000 1.51831 .33950 6.0000 .91766 .20520 .000 Ideas 7.6600 .69767 .15600 6.6000 .94032 .21026 .000 Vocabulary 7.6000 1.53554 .34336 5.6000 1.23117 .27530 .000 Audience 6.4250 1.23454 .27605 5.4250 .75524 .16888 .004 As shown in Table 4.4, all the mean differences were statistically significant, with the G1AdvEL2/AL1 students outperforming the G2IntermEL2/AL1 students. The P- value produced low and relatively high statistical significance. For the Audience feature, low statistical significance was found as shown by its P value (P = 004 for the Audience). There were relatively high statistical significance for text structure, ideas, persuasive devices, cohesion, vocabulary, paragraphing, sentence structure, and punctuation. In summary, the G1AdvEL2/AL1 students scored higher in all key language features. 4.1.4 Summary The English proficiency levels were related to the application of key language features in the English scripts. G1AdvEL2/AL1 was rated as Superior in applying the key language features (Audience, Text structure, Ideas, Persuasive devices, Vocabulary, Cohesion, Paragraphing, Sentence structure and Punctuation) in their English scripts and G2IntermEL2/AL1 received low essay scores in using the same key language 74 features in their scripts. In the next section, the main goal of the thesis is to ascertain whether there is evidence for reverse transfer in the Arabic essays produced by the G1AdvEL2/AL1 and G2IntermEL2/AL1 students. 4.2 Scores of Arabic Scripts In the previous section, it was shown that English proficiency affected the use of key language features in the English persuasive scripts. This section examines the Arabic scripts written by Arabic and English majors for evidence of reverse transfer. This section is based on five tables of survey data. The data from Tables 4.5, 4.6 and 4.7 show the percentage and scores of Arabic scripts in terms of the nine key language features adapted from (NAPLAN, 2011). In the three tables, each of the NAPLAN scores relevant to each category is converted to a scale of 1 to 10. Table 4.8 shows a comparison of English and Arabic scripts by the G1AdvEL2/AL1 and G2IntermEL2/AL1 students on key language features. Table 4.9 shows the mean comparison of quality score of Arabic scripts written by English and Arabic majors. 4.2.1 The Use of Key Language Features in the Arabic Essays – the G1AdvEL2/AL1 Group. Table 4.5: G1AdvEL2/AL1 Key language feature percentages and scores for the Arabic essays Number of scripts Original Score New Score Overall rating Percentage 0 - 6 0 - 10 Audience 7 3 5.0 Satisfactory 35 13 4 6.7 Superior 65 20 100 0 - 4 0 - 10 Text structure 15 3 7.5 Superior 75 5 4 10.0 Superior 25 20 100 0 - 5 0 - 10 Ideas 6 3 6.0 Satisfactory 30 14 4 8.0 Superior 70 20 100 0 - 4 0 - 10 Persuasive devices 15 3 7.5 Superior 75 75 5 4 10.0 Superior 25 20 100 0 - 4 0 - 10 Cohesion 4 2 5 Satisfactory 30 16 3 7.5 Superior 70 20 100 0 - 3 0 - 10 Paragraphing 8 3.3 8.0 Superior 40 12 3 10.0 Superior 60 20 100 0 - 5 0 - 10 13 3 6.0 satisfactory 65 Vocabulary 5 4 8.0 Superior 25 2 5 10.0 Superior 10 20 100 0 - 6 0 - 10 Sentence structure 14 9 8.3 Superior 70 6 6 10.0 Superior 30 20 100 0 - 5 0 - 10 Vocabulary 13 4 8.0 Superior 65 7 5 10.0 Superior 35 20 100 0 - 6 0 - 10 Sentence structure 14 9 8.3 Superior 70 6 6 10.0 Superior 30 20 100 0 - 5 0 - 10 Punctuation 8 3 6.0 Satisfactory 40 12 4 8.0 Superior 60 20 100 4.2.1.1 Audience Overall, the ability of seven of the G1AdvEL2/AL1 participants to orient, engage and persuade the reader through their written scripts was rated as Satisfactory, with a 5/10 rating. The remaining thirteen participants’ essays were rated as Superior at 6.7/10. 78 “They get to enjoy social and economic benefits, as well as the mental benefits of learning a new language”; “It sharpens skills on reading, negotiating and problem solving”). (Bilingual_Adv_P6) (“It makes us keeping up the latest news and it shows us how to deal with the modern technology”, “Learning a new language helps us to learn about new cultures, customs and traditions”). (Bilingual_Adv_P15) 4.2.1.4 Persuasive devices All twenty participants’ scripts were rated as Superior under the category of using persuasive devices that are used to enhance the writer’s position and also persuade the reader (NAPLAN, 2011). However, in terms of allocating specific ratings there was a difference as only five participants were rated 10/10, while the others were rated at 7.5/10. The main reason for this specific distinction was based on the effectiveness and relevance of the persuasive devices that they included in their scripts. The following extracts are representative of how the majority of G1AdvEL2/AL1 students (Superior) introduced their writing in Arabic: Personal opinion (“I would prefer”), Direct address of the reader (constant use of (“نحن” “we”, "ك" “your”), Personal stories 79 “For instance, here in Egypt, we have special games that other countries do not know them as “Alstgmaih” but people who learn Arabic and read about Egyptian’s culture and habits would know it.” Repetition / (Knowledgeable with other worlds/gate to other worlds). Modality “They believe that they could get money in an easy way”. (Bilingual_Adv_P1). Several instances of direct address of reader (“نحن” “we”, "ك" “your”), Reference statements “ “University of Chicago”, Emphatic statements “It boosts the brain power.” “It sharpens skills on reading…” Value statements “It can help us to make choices that could profit us further down the road”. (Bilingual_Adv_P6) 80 Personal opinion, “I like learning new languages for some reasons and aims” Conditional mood “If a worldwide university offers a scholarship….” Authoritative statements “It is exactly what many want and desire”, Value statements, “Some people are not interested in sport…” Modality (“It is more exciting….”). (Bilingual_Adv_P3) 4.2.1.5 Cohesion Through the findings, it was identified that cohesion is a major factor that needs more emphasis when writing persuasive essays. Even though all participants’ scripts were ranked Superior in this context, sixteen of them were able to get a 7.5, mainly due to the fact that these sixteen used other connectives such as however, although, therefore and additionally within their essay (NAPLAN, 2011) making the content of these essays tightly linked and enhancing readability. Unlike the remaining four, they were able to control multiple threads and relationships across the text. The following extracts are representative of how the majority of G1AdvEL2/AL1 students introduced cohesive devices in Arabic: .”Thus, In addition, also, in conclusion“ (وهكذا ، باإلضافة إلى ذلك ، أيضا ، في الختام) (Bilingual_Adv_P4). "الغزو / مستعمر" Moreover”, “Furthermore”. Word associations“ "عالوة على ذلك" “invasion/colonist”. (Bilingual_Adv_P7). 83 0 - 4 0 - 10 Cohesion 17 2 5.0 Satisfactory 85 3 3 10 Superior 15 20 100 0 - 3 0 - 10 Paragraphing 3 1 3.3 Developing 15 17 2 6.7 Superior 85 20 100 0 - 5 0 - 10 Vocabulary 15 3 6.0 Satisfactory 75 5 4 8.0 Superior 25 20 100 0 - 6 0 - 10 Sentence structure 7 2 3.3 Developing 35 10 3 5.0 Satisfactory 50 3 4 6.7 Superior 15 20 100 0 - 5 0 - 10 Punctuation 4 2 4.0 Developing 20 9 3 6.0 Satisfactory 45 7 4 8.0 Superior 35 20 100 4.2.2.1 Audience Among all twenty, six of the G2IntermEL2/AL1 scripts were scored ‘Satisfactory’ and allocated 5/10 for their capacity to orient, engage and persuade the reader. These participants were able to orient the reader through their essays; however, their ability to engage and persuade the reader needed improvement through effective use of language and persuasive techniques. The fourteen remaining participants of this cohort were ranked 6.7/10 in terms of their ability in orientation, and engaging and persuading the reader through language choices. However, none of these G2IntermEL2/AL1 students was able to score 8.3 or 10/10. The main reason for this was their inability to establish a strong credible voice within the chosen context and stance. Their scripts also demonstrated a lack of sustainability in language choices and in the use of persuasive techniques (NAPLAN, 2011). Better examples include: 84 “Everyone wants to put his family in [a] high position”, “Man can’t depend on sport to build his future or to improve his position in the society to be a decent citizen”. (Bilingual_IntM_P21). “Learning sport is a complement to learning language”, “Essential for life continuous”. (Bilingual_IntM_P22). “Countries in the world have become more and more interdependent and new technologies have erased many existing borders”. (Bilingual_IntM_P26). 4.2.2.2 Text Structure As the skill focus of this section denotes, the participants were meant to show explicitly an introduction, body and conclusion in their Arabic essays. This text structure was clearly visible in fifteen scripts and they were allocated 10/10 and categorised as ‘Superior’. The remaining five scripts were categorised as ‘Satisfactory’ and scored 5/10. They were tested not only for having the separate sections, that is, introduction, body and conclusion merely for the sake of having these sections, but they also needed to include specific content in a coherent, controlled manner. 85 4.2.2.3 Ideas Five of the twenty participants, were rated 8/10 for the selection, relevance and elaboration of ideas that are essential for making a persuasive argument. However, the main reason that stopped them from achieving the highest rating in this category was due to their inability to include a more balanced perspective on the given topic by including a refutation of other positions or opinions. Twelve participants’ scripts were categorised as Satisfactory in this aspect because they had all included many ideas that were not adequately elaborated. Three of the participants (or 30%) scored 2/10 and were rated Developing. At this stage, the ideas may have no or insufficient evidence, or may range from one to few ideas that are either unrelated or not elaborated, thus preventing the reader from making a reasoned connection between ideas and argument. The following extracts are representative of how the majority of G2IntermEL2/AL1 introduced their ideas in Arabic: Ideas are sometimes elaborated, mostly they contribute effectively to the writer’s position, and occasionally reflect on wider world issues. “Some people have more aptitude for learning languages including children…”, “Learning a language helps us to defend values and ideas related to religion and faith” Not strong enough for a score 8, due to some ineffective/unelaborated/confusing arguments. “Require high fitness to continue to exercise as well as respect for learning language”, “As is the language of science and technology and scientific research” (Bilingual_IntM_P21).