The role of mother tongue english language essay

INTRODUCTION

The role of mother tongue (L1) in second language learning and instruction has been reevaluated several times within the last few decades. L1 has been balancing on the language teaching scales from being viewed as a hindrance in the era of Behaviorism, to its insignificance as argued by Chomskyan linguists, then to it having a double positive or negative transfer effect as was believed by cognitivists, to the modern era where scholars interpret L1 as a part of repertoire of strategies of second language (L2) learners use in the course of L2 acquisition (Fries, 1945; Chomsky, 1965; Krashen, 1984; Selinker , 1983; Cook, 2001; Cohen and Brooks-Carson, 2001). Being a challenging endeavour, L2 writing confuses processes inherent in L1 writing (brainstorming ideas, generating content, evaluating appropriate vocabulary and grammar structures, revising and editing text) with L2 language processing issues. Given the complexity of writing and the role that L1 plays in modern language instruction, it is worthwhile to investigate how language learners use L1 from their repertoire of strategies to succeed in L2 writing. In Kazakhstan, English writing becomes extremely important because students have to prepare for the international high-stakes tests (i. e. IELTS) or university examinations in which essay writing plays an important role. Consistent with the high demand, Academic Writing courses have become established in the English language curriculum in the Universities, and even in the Departments where the main medium of instruction is not English, but L1. Currently, students are required to take from one to three years of Academic Writing instruction and offered elective standardized tests[1]preparation subjects which also focus on writing an academic essay (personal communication with the English Language Faculty, February, 2013). Sadly, Kazakh students perform unsatisfactorily on the writing exams scoring below the benchmark set by exemplary institutions.[2]Not surprisingly, English language teachers attribute this to the lack of exposure to authentic input, in other words to EFL (English as a foreign language) environment, claiming that target language only finds its use in the classroom, whereas outside learners never use it beyond a narrow range of topics, therefore students struggle to understand authentic speech and texts, let alone master quality writing. However, such underperformance in writing tasks could also be attributed to the present teaching methodology adopted at the Faculty. Present methodology does not allow for the use of L1, it is viewed a as a hindrance rather than valuable resource, so the students are explicitly discouraged from using L1 in a classroom (ibid, March, 2013). Such practice does not reflect what research has taught us about the major role of L1 in second language acquisition and, certainly, does not regard L1 as valuable resource in advancing L2 writing. In this context, it is reasonable to suggest that provided English teachers are better informed about the cutting edge research in second language acquisition and have greater understanding of the role of L1 in L2 writing, they will be able to better assist students on academic writing, and, hopefully, improve the standards. Thus, the focus of the proposed research study is not whether the mother tongue is involved in L2 writing production, since in the light of theories currently being debated by linguists, it is clear that leading research suggests that the process of foreign language (FL) acquisition is mediated by mother tongue (Cook, 2001; Meyer, 2008), rather it is reasons and conditions that predetermine the choice of L1 over L2 or vice versa in composing an academic writing piece. Certainly, FL learners have two languages (or more) at their disposal, which they constantly interchange to achieve their communicative goal both in speaking and writing. Drawing on this observation, the purpose of the proposed study is to discover when EFL learners rely on their L1 when composing in L2. More specifically, the study aims to explore at what cognitive stages (i. e. planning or revising) in their composing EFL learners use L1 to facilitate the writing task and how this strategic use of L1interacts with an argumentative writing task, as well as overall writing quality as assessed by holistic means, traditionally employed in English international tests (Smith, 1994). This essay is a critical analysis of the preliminary research design of the planned study and is organized as follows. I will begin with a review of past research on the nature of L2 writing emphasizing cognitive model of writing and the studies that addressed the role of L1 in L2 composing. Next, the research questions will be posed and clarified. Section 4 deals with the discussion of research paradigm and methodology adopted. Here, the design of the study will be presented in detail with regards to its methods and data collection procedures, population and the process of data analysis. Finally, the issues of validity and reliability of the data as well as ethical issues concerned with the proposed research project will be discussed.

THEORETICAL BACKGROUND

Following the designs of L1 composition studies, researchers have investigated L2 writing processes and reported that the use of L1 is a ” common strategy” among second language learners (Krapels, 1990, p. 49). This research on L1 in L2 writing has addressed a number of topics: transfer of L1 strategies into L2 composition (Cumming et al., 1989; Jones and Tetroe, 1987); generating ideas, developing concepts, elaborating on content (Friedlander, 1990; Wang and Wen, 2002; Uzawa and Cumming, 1989); search and evaluation of vocabulary (Cumming, 1990); translating from L1 (Kobayshi and Rinnert, 1992; Uzawa, 1996); revising and editing (Lay, 1982; Qi, 1998; Wang and Wen, 2002). Undoubtedly, previous studies have significantly contributed to our understanding of L1 use in L2 composing processes; nevertheless they are limited in several ways. First, in current research the types of L2 writing tasks have been limited to ” knowledge telling” inherent in narration and description which simply involve retrieval of information from existing memory (Bereiter and Scardamalia, 1987, p. 5). Few studies have focused on argumentative tasks which have higher cognitive demands involving ” knowledge transformation” (ibid.), meaning abstract thinking, problem solving, etc. Since performing standardized English writing tasks is a challenge often faced by L2 writers, it is meaningful to investigate how students’ L1 use may be employed to succeed on argumentative essay writing[3]. Secondly, research regarding L1 use in L2 composing and writing is currently dominated by ESL (English as a Second Language) students while studies regarding EFL learners are still few. Leki (2000) claimed that due to influences of different environments of language learning, ESL and EFL students may display salient distinctions in many characteristics regarding English language acquisition. Finally, the context of Kazakhstan has never been explored and it is one of the modest contributions of this study to fill the gap in EFL research. With a few notable exceptions, the findings from recent studies tell us little about the role of L1 in composing activities involved in L2 writing. In their study of Chinese EFL writers, Wang and Wen (2002) argued that L1 is extensively used in generating and organizing ideas and L2 is most likely to be employed when undertaking task-generating and task-examining activities. Based on these findings, the researchers proposed an L2 composing process model which rests on the well-known Flower and Hayes (1981) model of L1 writing. In this more recent model, the task environment, the composing processor and the role of the writers’ long-term memory were borrowed from the original Hayes and Flower model but since the three operational processes, i. e. ‘ planning’, ‘ translating’ and ‘ reviewing’ have been heavily criticized for giving a false impression that writing is linear and sequential process with well-defined boundaries, Wang and Wen (2002) presented their own five categories of composing activities: task-examining, idea-generating, idea-organizing, text-generating and process controlling (p. 243). However, this model has not become popular or broadly adopted mostly because of the concerns over a small sample size of the study (16) and the reliability of using a think-aloud technique as a sole method for data collection. Unlike Wang and Wen’s study, which compared composing processes while performing narrative and argumentative tasks, the proposed study will focus on argumentation only and, due to its cognitively demanding nature, will rely on cognitive model of writing. Hence, Kellogg’s (1994) cognitive model of writing was selected to guide the discussion of the role of L1 in L2 composing, because this influential model recognizes the recursive rather than linear approach to writing and considers how cognitive factors influence composing processes. In this model writing consists of three important interactive and recursive processes: conceptualization, execution, and monitoring. Conceptualization involves planning the content of the writing and translating ideas into words. During planning, writers retrieve ideas from their long-term memory or from the cues provided by the task and organize them into a coherent order. In the translation of ideas into linguistic form three processes can be distinguished: retrieval of lexical items, syntactic encoding of clauses and sentences, and expressing cohesive relationships in the text. In the execution stage, writers use motor movements to create a handwritten or typed text. Finally, monitoring ensures that the created text adequately expresses the writer’s intention and, if mismatches are found, the text is revised (Fig. 1). Figure 1. Cognitive stages of writing. (Kellogg, 1994)The above outlines up-to-date research on L2 writing and the findings on L1 use as well as theoretical framework proposed for investigation provide necessary sources and background for the research questions and methodology of the present study.

RESEARCH QUESTIONS

Taking into account the scarcity of research on writing in the EFL environment the following two research questions will be addressed in the proposed study: To what extent do intermediate Kazakh learners of English in an academic context rely on their L1 when composing a piece of argumentative writing? It is hypothesized that EFL learners rely extensively on their first language while composing an essay because they lack exposure to the target language and therefore are limited to some extent in the topics they can address in their writing. Also having been receiving formal writing instruction in English for a rather short period of time, the learners are likely to use their L1 to retrieve topic knowledge, as suggested by Friedlander (1990), who argued that positive transfer of L1-related content enhances the quality of L2 composition if EFL students are allowed to use language of knowledge acquisition while composing. In the case of Kazakh University students, it is reasonable to claim that students have acquired most of their general knowledge through native language, therefore they are likely to generate topic and elaborate on content with the help of L1. At what stages of composing a piece of argumentative writing do these learners most rely on their L1, and why? The Kellogg’s model of writing will enable us to determine the different stages of processing, and will therefore provide a framework for identifying the points at which learners employ L1 most. Then learners’ reasons and intentions of L1 use at these stages will be investigated and discussed along with the learners’ strategies of language switching across linguistic systems.

RESEARCH DESIGN

Research paradigm and methodology.

To answer the research questions, the present study will be informed by a Grounded Theory (GT) approach within the constructivist paradigm. In this paradigm, it is believed that knowledge is constructed through interaction ‘ between and among’ researcher and participants, it is emergent and grounded in actions and experience (Guba and Lincoln, 1994, p. 111). The distinctive feature of a constructivist researcher is that s/he expects the concepts, topics or categories of importance to emerge as they are constructed by the participants. This emergence is also a distinctive feature of the GT approach, which aims to build up a theory of a phenomenon (in the present case it is L1 use in L2 writing) while being ” grounded” in the collected empirical data and allowing data to generate ‘ insights, hypotheses, and generative questions’ to construct further knowledge about the studied phenomenon (Schwandt, 2001 as cited in Taber, 2009). GT is the only approach that ‘ does offer a system that is designed to develop a rigorous theory in contexts where little reliable theory exists’ (Taber, 2009, p. 216). For this reason, both constructivist paradigm and GT approach are perceived as the most suitable for the proposed study, as there is very little empirical research investigating the role of L1 in L2 discursive writing. Within constructivist paradigm, a case study is perceived as particularly fitting strategy of inquiry for rich in-depth data. According to Bassey (1999), a case study should recognize the ” complexity and ‘ embeddedness’ of social truth”. The proposed study of the nature of L2 writing implies a thorough investigation of the L1 role in a situated EFL environment; therefore it is embedded in local context, what makes it suitable for a case study inquiry. Also, given the scope of the MPhil study and the limited time available to the researcher, a small-scale case study is deemed as the most appropriate research strategy (Wilson and Stutchbury, 2009). Accordingly, the proposed study fits the case study scenario because it is a contextualized empirical enquiry of a single instance (that is L2 writing) within its real life context (i. e. classroom, examination) (Yin, 2003; Nunan, 2005). This specific educational and cultural context plays an important role in influencing learners’ writing practice and processing and researcher’s interpretations. Secondly, the study focuses on providing a comprehensive explanation of cognitive operations of L2 writers, which cannot be directly observed; therefore the researcher will employ highly qualitative techniques, such as verbal protocols and stimulated recall. Lastly, even though ” grounded” in data, researcher is not expected to let the data ” speak for itself”, but to interpret the data and construct and co-construct knowledge, thus subjectivity is an integral part of this research underpinned by constructivism. Both participants and the researcher are expected to be biased to some extent, which is affected by their own assumptions and interpretations.

Methods of research

The present investigation will use verbal reports (VR) as a core method for data collection. In essence, verbal reports are the comments that participants make while they perform a task or immediately after the task is completed. In their seminal work on protocol analysis Ericsson and Simon (1984) distinguished two types of verbal reports: concurrent and retrospective (p. 16). Concurrent reports (or think-alouds) are those performed during the task, while retrospective are the ones executed afterwards. Both have been extensively employed in SLA research to reveal and examine processes that are naturally hidden and not observable, such as learners’ cognitive processing and strategies use (Bowles, 2010, pp. 1-3). Concurrent verbal reports have been successfully employed in the following exemplary L2 writing studies: the role of translation from L1 and use of it as a facilitative strategy in L2 writing (Cohen and Brooks-Carson, 2001; Kobayshiand Rinnert, 1992; Jones and Tetroe, 1987), various L1 transfers (e. g. writing strategies, discourse patterns etc.) into L2 writing (Cumming, 1990; Uzawaand Cumming, 1989; Wangand Wen, 2001) and L1 as a source of topic knowledge (Friedlander, 1990; Lay, 1982). To ensure validity and richness of data, the present study will combine verbal reports with stimulated recall and semi-structured interviews. Each of these methods will now be critically analysed in terms of their appropriateness to the proposed study.

Concurrent or Think-aloud protocol

While being an invaluable tool in writing research, think-aloud method has been criticized for its ” artificiality, incompleteness… and the fact that it may even distort writer’s normal writing processes” (Hyland, 2002, p. 184). With regard to ‘ artificiality’, Ericsson and Simon (1984) have argued that thinking aloud activity is not an ‘ alien’ activity to people’s everyday life (p. 78); and in the present study the subjects are quite familiar with the activity in the context of their schooling. In Kazakh school practices, students are often instructed to think aloud and explain their decisions while solving mathematical equations on a whiteboard in front of the classmates and the teacher. Hence, the researcher regards thinking aloud as a common activity that informants have experience with. Next, the foreseen ‘ incompleteness’ of think-alouds will be tackled by a complimentary method of stimulated recall, which will be discussed in more detail in the next section. At last, the argument that verbalization is reactive, i. e. it alters the thought processes and therefore affects writing, has a long and controversial history. Cognitivist SLA scholars have long firmly believed that verbalization is a ‘ window into the mind of learners’, it precisely reflects the thought process without altering it (Bowles, 2010, p. 2), although many others raised concerns over validity of verbal reports and questioned accurate representation of thoughts. The design of the proposed study will separate planning and execution stages of writing performance, so the subjects will be engaged in two types of verbal reporting: (1) metacognitive VR during planning stage and (2) non-metacognitive VR during execution stage. While planning the informants will not be time-constrained and the writers are expected be very conscious of the decisions that they make and the reasons underlying such decisions, thus metacognitive processes will be dominant. Therefore, VR is perceived natural and not distorting normal thought processes. At the execution stage, on the contrary, metacognitive processes will be less accessible and conflicting with the task cognitive demands. For this reason, during execution stage, the participants will be asked to produce a non-metacognitive report, which is verbalizing thoughts per se, streaming them as they occur without attempting to explain or reason about them. Previous research suggests that concurrent reporting is reactive. However, as argued by Bowles (2010), non-metacognitive reports reflect the nature of cognitive processes fairly accurately, while slowing processing slightly. Thus, by asking participants to verbalize their thinking as it occurs, reactivity will be hopefully reduced. Although not flawless, concurrent report is perceived as the most appropriate for the planned study, since there is no other method, which can capture the reasoning of the subjects during writing activity and reveal cognitive and psycholinguistic processes underlying this activity. Having briefly presented the shortcomings of think-aloud method as an instrument of data collection, it is worthwhile highlighting its advantages. Gass and Mackey (2000) have argued that concurrent reports are able to reflect theoretical framework (p. 111). Also, this instrument allows going beyond the common practice of judging L2 learners competence on the basis of performance and product analyses revealing processes underlying the performance (Kormos, 1998).

4. 2. 2. Stimulated recall.

The VR data will inform what the students were thinking; to answer why they were thinking in a specific way and specific language stimulated recall interview will follow. Stimulated recall (SR) is introspective method which invites subjects to recall thoughts they had while performing the task by enhancing their memory through aids and ‘ tangible’ prompts (Gass and Mackey, 2000, p. 12). As a methodological tool SR has a number of definite advantages, all those relevant to this study include: has a considerable potential when studying cognitive strategies (Gass and Mackey, 2000; Lyle, 2002); allows participants to explain their decision-making (Gass and Mackey, 2000); requires minimum training into research goals (Lyle, 2002)In the present project this research method is fit for purpose as it will allow gaining more insights into writers’ cognitive processes through inviting participants to recall certain moments of their thinking activity which occurred while simultaneously performing the task and the VR. A short audio-recorded SR session will be conducted with each participant of the study immediately after completion of the essay. As the researcher will be present during the planning and the execution of the task, she will take field notes, noting instances of L1 use, as well as long pauses in think-alouds. These field notes and the participants’ written product (essay, planning notes, any corrections) will serve as stimuli in answering interview questions which will be designed to reflect specifically the focus of the study – the use of L1. SR has been appealing to researchers interested in uncovering cognitive processes of learners. In the writing research, Bosher (1998) used video SR as alternative to VR to examine L2 writing processes of ESL students, more recently Sasaki (2000) combined the VR and SR to investigate Japanese EFL students’ writing processes. Similarly, in the proposed study SR will be employed as an auxiliary to VR protocols method of data inquiry in order to overcome some shortcomings of concurrent thinking and to gain learner’s perspectives and interpretations of their thinking so as to answer research question two: when do students rely on their L1 most in L2 writing and why that happens. Nevertheless, advantageous as it is, SR, like any other research tool, has validity and reliability concerns. First, it must be carried out as soon as possible to avoid memory loss, so it is responsibility of the researcher to minimize the gap between the event and recall. Thus, in present study the recall would be immediate. Next, participants may sensor and distort their thoughts so as to present themselves more favorably or may ‘ create’ explanations instead of recalling. A final procedural concern is the language of the recall sessions. The language will remain English, because by the time of the interview the subjects will have been communicating in English for about one hour and it should not be a problem for them to continue. However, if the student is tired and prefers to speak his/her L1, the researcher will allow such option.

Semi-structured interview.

The interview will follow the SR session simply to establish validity of think-aloud protocols and students’ general experience of L1 use in writing. Appendix 2 contains a set of interview questions.

Participants

As far as population sampling is concerned, it could be characterized as ‘ purposive sampling’ (Descombe, 2007, p. 17). The researcher will conduct fieldwork in a public University, at the Faculty of English Language Teaching. Having personal contacts, the researcher has obtained an official permit to get access to the prospective participants over the 3-week period in April, 2013. The sample is also defined as purposive, because the researcher has been made aware of students’ backgrounds, level of language, their general workload and availability. Given the constraints of time both for data collection from each participant and transcription of verbal reports and further analysis, the maximum practical number of participants is restricted to 12. The prospective subjects are first year undergraduate students enrolled in TEFL program (Teaching English as a foreign language). They have relatively homogeneous educational background, sharing the following characteristics: taught by non-native English speakers in an EFL environment, have little or no exposure to English outside the classroom, have been taking an Academic Writing course aimed at IELTS bond 6. 0, share approximately the same level of English comparable to B2 on CEFR scale (Common European Framework of Reference for Languages). Also, these students are expected to take one of the English International tests (usually IELTS) in due course where scoring in discursive essay is essential.

Materials

As the study is intended to simulate writing situations that students encounter in typical classroom and examination assessments, the essay topic has been selected from the Faculty’s pool of topics for Academic Writing (Intermediate B1-B2) course assessment. Friedlander (1990) argued that topics that may provoke thoughts in L1 should be avoided if we are to ban L1 from classroom instruction completely. Thus, in the present research the topic should not require knowledge primarily acquired through L1 (i. e. childhood), but at the same time should not contain notions and topics that target language emphasizes (i. e. learning a foreign language). The research aims to simulate natural flow of thoughts, not jeopardizing the source of primary knowledge. For this reason the topic selected reflects the broad topic of Education, which the participants have addressed multiple times throughout their schooling in L1 environment and have been focusing throughout this academic year in EFL environment. Argumentative writing is this research is broadly defined as convincing the reader of your point of view. The subjects have been trained that argumentation requires an awareness of formal register, an ability to manipulate abstract concepts and the rhetorical conventions (personal communication with the Faculty, March, 2013). Guided by this, students will be asked to present a written argument by agreeing or disagreeing with a given statement and supporting their answers with relevant examples and reasons (see Appendix 1).

Procedure

The data collection sessions will include warm up, planning stage, execution stage, stimulated recall and semi-structured interview. Warm up (max 15 minutes)The warm up stage is perceived essential by scholars (Ericsson and Simon, 1984; Bowles, 2010) because it helps to familiarize participants with the equipment and ensure that all informants understand the procedure. In this study a warm up stage will include oral instructions for verbal reporting and a practice of thinking aloud. Ericsson and Simon (1984) suggested that most difficulties with think-alouds have to do with requirements imposed on participants during the report. For this reason the researcher will follow the designed protocol suggested by Bowles (2010): 1. Reiterate the reason the participants are asked to think-aloud; 2. Provide instructions about how they should think aloud; 3. Include a warm up task during which participants practice thinking aloud and have time to ask the researchers any questions about the process before beginning the operational study (p. 114-115). In accordance with this protocol, the oral instructions will be worded in the following way: In this study we are interested what you think and how you make decisions when writing an academic essay (1). In order to do this I am going to ask you to THINK, REASON OUT LOUD as you write an essay. What I mean by THINK ALOUD is that I want you to tell me everything you are thinking from the time you first see the topic of an essay until you finish writing you final sentence. Do not plan what you say or explain what you are saying. Just act if you were alone in the room speaking to yourself. Use the language your thinking occurs in. (2). Try to think aloud throughout your performance. If you are silent for a long period of time I will ask you to keep talking. As a third warm-up component, students will be given an opportunity to practice the think-aloud technique with a mock composition (” Advantages and disadvantages of…”). No modeling of the think-aloud will be provided in order to avoid potential danger of the informants restricting their thoughts to the model (Smith, 1994). Moreover, it is essential for the present research to discover whether students think in their L1 during writing processes, and if so how much and when they are most likely to use it. Hence, modeling of concurrent protocol might influence the choice of language. So once the students are familiarized with the technique, they will be given the assignment, which will first ask them to plan their answer and then proceed with the writing. Planning stage (max 15 minutes)The reason for introducing this planning stage is that participants will find it much easier to think aloud. It might or might not assist them to perform concurrent VR when they continue on to execution, but it would prepare them so they will have had some experience of thinking aloud on the essay topic in easier, more natural circumstances then while they are actually writing. In her turn, the researcher will be making notes of L1 use and hesitations. Participants will also be encouraged to take notes at this stage. In addition to oral instruction, written explanation of the think-aloud procedure will be presented along with the task on an answer sheet (see Appendix 1). As instructed by Bowles (2010), they will be written in ‘ plain language’ so as to ensure complete comprehension (p. 113). Naturally, I will pilot-test the verbalization instructions (along with other materials) on a small number of participants whose data will not be included in the final sample. Pilot-testing will help to ensure that verbalization instructions are written clearly, in a way that participant understands and can follow. Any ambiguities that are found during pilot-testing will be corrected prior to the operational study. Execution Stage (40 minutes)Once instructions are presented and the student has taken time to plan, s/he will be given 40 minutes to write an essay. Such timing is common for both writing examinations at the Faculty and during IELTS or entering examinations to educational institutions. Although Ericsson and Simon (1987) warn that subjects who are verbalizing can be expected to take more time for the task (p. 78), it is believed that having allowed students to plan ahead and make necessary notes, students will manage to complete the essay in a given time. The researcher will be present and observing writer’s performance, taking notes of long pauses or any peculiarities and general writing/speaking behaviors of the subjects. While student is composing aloud, no interruption will be made, except for an occasional ” keep talking” reminder if a student pauses for longer than a 2-3 minutes, such interruption should cause very little, if any disruption (Ericsson and Simon, 1987, p. 83). The whole process will be audio-recorded. Stimulated recall (5-10 minutes)Gass and Mackey (2000) suggested that participants are minimally trained for SR and it is important to design procedures that create the strongest links possible between the focus of the study and the procedures for creating the recall. So the SR will be conducted with no delay and will focus on L1 use. Informed by field notes, subjects will be referred to specific passages in their essay and will be asked provide reasons for thinking in L1 or L2 in those passages. The recently produced text and the researcher’s field notes will be the stimuli that would hopefully trigger recall of the processes that were involved in the text production. These stimuli are believed to enable subjects ‘ to relive an original situation with great vividness and accuracy’ (Bloom, 1954 as cited in Gass and Mackey, 2000)Semi-structured interview (10 minutes)Finally, the subjects will be asked to comment primarily on the extent, to which they use their L1 in L2 writing, and on any difficulties they may experience with thinking-aloud. This semi-structured interview will be short and specific, and will also be audio-recorded (see Appendix 2 for a list of interview questions). Accordingly, each session with informants will take approximately 1, 5 hours, during which the subjects will meet the researcher individually in a quiet office.

Validity and Reliability

Undoubtedly, the principles of validity and reliability are fundamental cornerstones that are at heart of any good quality research. Validity, in the broadest sense, refers to the extent to which the study ‘ actually investigates what it purports to investigate’ (Nunan, 2005). In other words, validity refers to credibility of conclusions reached through interpretation of data (internal validity) and generalizability of such conclusions (external validity) (Evans, 2009). Since the proposed study is a case study adopting a qualitative methodology, generalization of findings is not a major concern, as argued by a number of theorists (Evans, 2009; Yin, 2003). However, as suggested by Lincoln and Guba (1985) ‘ thick description’ will be provided, that is the researcher will provide rich contextual information ‘ to enable someone interested in making a transfer’ (as cited in Evans, 2009, p. 118). The concurrent verbal report, as has been described is a core method of data collection and arguably threats validity of this research the most. A number of challenges that VR poses are believed to be addressed in the current design of the study. Namely, sensitivity to instructions is dealt with by reference to relative literature (Ericsson and Simon, 1984; Bowles, 2010) that enabled developing clear instructions that would be presented in plain and terms-free language to the participants. Without a doubt, the biggest challenge is that thinking aloud might influence regular thought processes and interfere with writing activity. The researcher acknowledges the threat and, although it is impossible to avoid reactivity and increased cognitive load, a number of steps have been proposed to facilitate this activity: the warm up mock essay exercise and the planning stage, where VR is gradually introduced and practiced before being executed. Additionally, the other two methods will complement the validity of the VR – the SR will ensure completeness of the report, by inviting the students to comment on certain parts of the produced text, and interview questions will establish how familiar the informants are with the technique, what their general impressions and difficulties of this method. Finally, the SR and interview sessions will be kept rather short and highly focused on the issue of L1 use; no delay in SR and strong stimuli are expected to enhance the internal validity of data. Hereby, I will focus on ensuring internal validity via: Compensation for potential pitfalls of data collection methods. Methodological triangulation – combination of VR, stimulated recall and interview as methods of data inquiry, which will be used hand-in-hand complementing their respective limitations and exploiting benefits. Reliability in this study, which is ‘ the degree to which there is consistency in the results’ (Gass and Mackey, 2000, p. 364) will be achieved through the following steps: Piloting the data collection and procedure with 2 students from the same background to ensure suitability of instruments, timings, instructions etc. Creating identical environment for all the participants and making sure that methods of data collection, materials and procedure remain consistent throughout the projectAsking the same set of question in the semi-structured interviews (see Appendix 2)Providing detailed description of data collection, procedure and data analysis so as to facilitate replication of the study. Minimizing researcher’s bias. Having defined reliability as a measure of consistency over time and over similar samples and taken the above measures as a warrant, it is expected that the obtained data will be representative of the population and if collected on a different occasion with the same instruments will remain consistent.

Ethical considerations

A number of safeguards will be taken to ensure that the proposed empirical study is ethical. First, before fieldwork gets underway, being external researcher, I will obtain an informed consent from the gatekeepers: the Dean of the Faculty and teachers of the prospective informants (Taber, 2007, p. 139 italics in the original). Next, a voluntary informed written consent will be solicited from the participants (see Appendix3). For this to be an informed consent the students will be briefed about the research goals and procedures that they are to be engaged in, how the data will be used and reported (BERA, 2011). However, the researcher will present the research goal in a ‘ deliberately vague’ manner, by not disclosing that the topic of L1 use is under investigation, as this may affect the quality of verbal reports and compromise the findings (ibid. p. 135). Finally, anonymity and confidentiality of research participants is of paramount importance in the conduct of the study. Thus, anonymity will be ensured by giving the informants alphabetical codes. The researcher will recognize informants as ” original ‘ owners’ of …beliefs and utterances” whose data is a ” gift” and treat the data with care and respect (ibid. p. 135), storing in securely on an external hard drive and ensuring its safety and confidentiality. Additionally, during the research, the researcher will try to minimize the impact on the normal workload of subjects. Writing sessions and interviews will be carried out during self- study sessions. Also, the researcher will do her best to put the participants at ease, by providing favorable conditions and short breaks during sessions to preclude distress or discomfort (BERA, 2011). According to Wellington (2000), not only should ethical considerations be at the heart of any research, but also ” should continue through to the write-up and dissemination stages” (p. 3). For this reason, upon the completion of data analysis, the participants will be informed of the findings of the empirical inquiry and offered a set of instructions that may help them to improve writing.

DATA ANALYSIS

Once the data is collected, full transcription of data will follow. To facilitate the process of transcription, the researcher will use Express Scribe software, which is specifically designed to assist transcriptions of audio-recordings. Subsequently, each VR will be split into sections to reflect the Kellogg’s cognitive stages of writing and ‘ thematic analysis’ will be performed, which is commonly defined as the process of analysing data according to commonalities, differences and relationships across a data set (Gibsonand Brown, 2009). Having identified the themes, coding scheme will be devised. As discussed earlier, no analytical framework exists to reflect the instances of L1 use, so a priori coding of data is impossible, therefore the researcher will reserve to empirically generated codes. Such codes are generated through iterative examination of data sets and derive from researcher’s general interest, i. e research questions (ibid.). To ensure reliability of coding, interrater reliability check will be carried out. A random set of data will be pulled out and two raters will be asked to code the data using the developed coding scheme. Further analysis will seek to explain when L2 learners rely on their L1 most in writing, what strategies they employ and what causes them to switch across linguistic systems.

CONCLUSION

The main purpose of the present essay is to justify the research design of the proposed MPhil study in terms of its methodology and procedures. The choice of research methods, their critique and suitability for research goals, sampling and procedures have been fully addressed and critically evaluated. The proposed study intends to shed light on the use of L1 by Kazakh writers in academic writing and put forward some recommendations for introducing L1 in L2 writing instruction with the aim of improving writing standards in the target context. Given the growing need to sit international high stakes tests and study abroad, the study appears to be significant both socially and pedagogically.