Ảnh hưởng của ngôn ngữ thứ nhất tiếng việt đối với việc viết tiếng anh một nghiên cứu về sinh viên chuyên ngành tiếng anh tại một trường đào tạo ngoại ngữ

The influence of mother tongue on second language learning has been a significant issue of research into second language acquisition, for mother tongue is an important part of learners’ linguistic background. The present research aims at (1) identifying the errors caused by L1 interference in written compositions, and (2) investigating learners’ perception of the reasons for such problems. The study employs error analysis to categorize errors in 106 written essays and organize interviews with students to explore their perceptions. The collected data from written essays and interviews are analyzed by quantitative and qualitative analysis. The results suggest that verb errors including verb voice and verb number are among the most prominent errors, while calque and collocation errors cause significant problems in lexis aspect. In terms of reasons behind L1 interference errors, two main causes are identified: the difference between Vietnamese and English, and learners’ L1 habit. The study indicates important pedagogical implications for English learning and teaching, which are expected to benefit stakeholders in ESL EFL contexts.

VIETNAM NATIONAL UNIVERSITY, HANOI

UNIVERSITY OF LANGUAGES AND INTERNATIONAL STUDIES

FACULTY OF ENGLISH LANGUAGE TEACHER EDUCATION

GRADUATION PAPER THE INTERFERENCE OF FIRST LANGUAGE

IN ENGLISH WRITING COMPOSITIONS: A STUDY OF ENGLISH MAJORS AT A

LANGUAGE UNIVERSITY

Supervisor: Trần Thị Thanh Phúc, PhD Student: Đinh Bảo Ngọc

Course: QH2014.F1.E1

HÀ NỘI – 2018

ĐẠI HỌC QUỐC GIA HÀ NỘI

TRƯỜNG ĐẠI HỌC NGOẠI NGỮ KHOA SƯ PHẠM TIẾNG ANH

KHÓA LUẬN TỐT NGHIỆP

ẢNH HƯỞNG CỦA NGÔN NGỮ THỨ NHẤT - TIẾNG VIỆT ĐỐI VỚI VIỆC VIẾT TIẾNG ANH: MỘT NGHIÊN CỨU VỀ SINH VIÊN CHUYÊN NGÀNH TIẾNG ANH TẠI MỘT TRƯỜNG

ĐÀO TẠO NGOẠI NGỮ

Giáo viên hướng dẫn:

Trần Thị Thanh Phúc, PhD

Sinh viên: Đinh Bảo Ngọc Khóa: QH2014.F1.E1

HÀ NỘI – 2018

ACCEPTANCE PAGE

I hereby state that I: Dinh Bao Ngoc, class QH2014.F1.E1, being a candidate for the degree of Bachelor of Arts (English Language Teacher Education) accept the requirements of the College relating to the retention and use of Bachelor‟s Graduation Paper deposited in the library

In terms of these conditions, I agree that the origin of my paper deposited in the library should be accessible for the purposes of study and research, in accordance with the normal conditions established by the librarian for the care, loan or reproduction of the paper

Signature

Date

ACKNOWLEDGEMENTS

The six-month process of conducting this graduation thesis has been such a valuable experience to me as a student researcher Without the support of my supervisor, my teachers and beloved people, the completion of this paper could not have been possible

Firstly, I would like to express my gratitude towards Dr Tran Thi Thanh Phuc, my supervisor, whose support and guidance has enriched my reservoir of knowledge, strengthened my research skills and encouraged me to finish this research

Secondly, I am thankful of all the participants of this study Their substantial assistance in the data collection process has facilitated the validity of the data and the study in general

Lastly, I owe great appreciation for my beloved family and friends for their constant support, especially my classmates at QH2013.E1 These people have provided me with important academic consultant, research resources and emotional encouragement, which accompany me till the end of this journey Thanks to their invaluable help, I have matured as a researcher and have had the determination to complete this research

1.4 Significance of the study 3 1.5 Organization of the study 4

CHAPTER 2: LITERATURE REVIEW 5

2.1 L1 transfer and L1 influence 5 2.2 Different approaches to second language acquisition 6 2.2.1 Contrastive Analysis 6 2.2.2 Error Analysis 8 2.2.3 Error analysis process 9

2.3 Overview of previous studies 16 2.3.1 Studies of L1 interference 16 2.3.2 Studies of Vietnamese influence on ESL learners 19

3.1 Research design 22

3.3 Data collection 23

3.4 Data analysis 25

CHAPTER 4: FINDINGS AND DISCUSSION 29

4.2 Frequencies of error types 32 4.2.1 Grammar errors 32

LIST OF ABBREVIATIONS

%: percent

CA: Contrastive Analysis

EA: Error Analysis

L1: First Language

L2: Second Language

EFL: English a Foreign language

ESL: English as a Second Language

TL: Target Language

LIST OF TABLES

Table 1 Grammatical errors (Dagneaux, et al., 1996, cited in Lopez, 2009) 11

Table 3 Types of lexical errors (James, 1998) 13

Table 7 Rank-order frequency of found grammar errors 31 Table 8 Rank-order frequency of found lexical errors 38

LIST OF FIGURES

CHAPTER 1: INTRODUCTION

This chapter presents the rationale of the study, states the research problems and specifies the scope and significance of the study

1.1 Background of the study

In the new era of globalization, the study of English as a second language has evidently claimed its international dominance For learners from different linguistic backgrounds, the process of second language acquisition encompasses the trace of their mother tongue In a broad sense, the mother tongue – the old knowledge – can influence the acquisition of second language – the new knowledge This phenomenon is defined as “L1 transfer”, which can exert both positive and negative effects on second language acquisition (Ellis, 1997, p.51) In other words, when a person learns a foreign language, existing knowledge of the mother tongue can create challenges for him or her in the learning process Such challenges are reflected in learners‟ errors, or interference errors, which are specified as those that “can be traced back in their mother tongue” (Lott, 1983, p 256) Since errors found in language learning can take roots from learners‟ mother tongue, the study of L1 transfer is essential in an effort to improve the quality of L2 acquisition (Ellis, 1997)

In the context of language learning in Vietnam, the challenges posed by L1 transfer can be considerable obstacles for Vietnamese ESL learners, considering the great disparity between Vietnamese and English Tran (2007) reports that English is an inflectional language with German-Indo European origin, while Vietnamese is a typical isolating language from the Austro-Asiatic language family The fact that Vietnamese is an isolating language, which means that words

do not change in form, certainly causes trouble to learners during their second language acquisition process (Mai, Vu & Hoang, 2003) Therefore, Vietnamese ESL learners consequently encounter numerous challenges: the habits of using L1

can affect their acquisition of L2 in different language components This reliance may result in problematic language utterances or expressions, reducing the quality

of the language products As errors in the language production (including spoken and written forms) are inevitable, language intelligibility can be hampered with such shortcomings (K Nguyen, 2015)

As a student majoring in English at one of the most prestigious Vietnamese universities, the researcher used to experience and have witnessed numerous cases

in which Vietnamese has a negative influence on students‟ English, particularly in written form From the researcher‟s experience, even English majored students find this influence quite challenging Therefore, the researcher has decided to conduct an investigation into this issue By examining sophomore students‟ writing pieces with proper error analysis process, meaningful implications could

be drawn to support the process of learning and teaching English

Another motive to conduct this study is to contribute to the validity of existing literature and enrich the scientific data in language transfer between English and Vietnamese Recently, various studies have been conducted on the influence of other first languages such as Chinese (Timina, 2013; Wang, 2015), Spanish (González, 2016) or Arabic (Hussein & Mohammad, 2011) However, since each language has its individual distinction, these empirical findings can barely provide generalizations about Vietnamese influence Despite a wide range

of studies into L1 transfer, the number of studies into Vietnamese‟s influence on ESL learners is quite limited Two worth mentioning studies are that of K Nguyen (2015) which examines Vietnamese‟s influence on English acquisition and use of people in Hanoi and that of Ha (2011) which studies L1 influenced written errors made by high school students Furthermore, undergraduate English majors are uncommonly chosen as subjects of existing studies, most of which focus on non-majored or high school students This research, which focuses on English majored

students‟ errors in English writing with relation to their L1 impact, hopes to bridge the gap in the existing literature of the field

1.2 Statement of research problem and questions

The study‟s primary purpose is to investigate the influence of mother tongue on written compositions by examining L1 interference errors found in the essays of English-majored students The study also attempts to explore the perception of the students on the reasons behind such errors In general, the study aims to answer the following questions:

i What are the L1 interference errors made by students in their English writing compositions?

ii What are possible reasons for L1 interference errors as perceived by students?

1.3 Scope of the research

The research is conducted on a small population of English majored students of one university in Hanoi It investigates the negative influence of Vietnamese as a mother tongue on English acquisition, specifically the errors in lexis and grammar made as a result of L1 influence Besides, within the scope of this BA thesis, the research focuses on studying the written products of the students and excludes other skill areas

1.4 Significance of the research

In achieving all the research goals, the researcher hopes to provide beneficial resources for ESL scholars, teachers and students, as well as those who share the interest in the topic Since the research reports the errors committed as a result of mother tongue influence, it would provide empirical evidence and updated data to resolve the inadequacy of study on L1 interference in written English It would hopefully bridge the gap in existing literature about Vietnamese,

in particular, as a hindrance on SLA Furthermore, the research would inform readers, possibly ESL teachers and students, of the reasons behind errors caused

by L1 negative influence on writing skill It would enable teachers and students to adjust teaching and learning strategies to avoid L1 negative influence on written compositions

1.5 Organization

The study report consists of 05 chapters as follows

- Chapter 1 – Introduction: The introduction presents an overview of the research including the background of the study, the research problem, its significance and scope

- Chapter 2 - Literature Review: The literature review includes a summary of major concepts and frameworks for this research, namely L1 transfer, error and error analysis, and a synthesis of related studies in L1 influence is also included

- Chapter 3 - Methodology: This section presents the research design of the study including the selection of participants, data collection, and data analysis procedure

- Chapter 4 - Findings and Discussion: This section reports the result and provides analysis and interpretation of the collected data

- Chapter 5 - Conclusion: This section sums up the findings and presents pedagogical implications from the research, while stating the limitations of the research and suggesting improvements for future studies

CHAPTER 2: LITERATURE REVIEW

This chapter presents a review of the background literature of the study, including the conceptual framework and existing studies related to the research The section starts with an overview of the definition of language transfer and mother tongue influence, then moves on to review the error analysis approach in second language study The chapter also includes a review of related studies by international and Vietnamese scholars, in which the gap in existing literature is portrayed

2.1 Language transfer and L1 influence

First language, or mother tongue, referred to the language that humans acquired when they were at young age and that became a part of their development (Saville-Troike, 2006) Saville-Troike (2006) also defined second language (L2)

as the additional language that people aimed to learn later in life, which is also called target language (TL) Second language was of great importance in learners‟ education, employment and other fundamental fields

Since people acquired different languages throughout their lives, there were inevitable interactions among them One of these interactions was language transfer which was defined by Odlin (1989) to be the influence caused by similarities and differences between the target language and prior acquired knowledge (p 27) This prior knowledge was not necessarily native language knowledge but any other ones

According to Ellis (1994), there existed several perspectives on language transfer including behaviorist, mentalist and cognitive view (cited in Lu, 2010) A simpler model of language transfer classification was made in consideration of language transfer sources, results and levels by Huafei (2010, cited in Wang, 2015) Language transfer could be divided into interlingual and intralingual

transfer, positive, negative and neutral transfer, or into different linguistic levels including phonetic, lexical, syntactic, semantic, pragmatic transfer

In a multilingual context, language transfer could be observed in the relationship among several languages In second language acquisition area, Ellis (1997) specifically discussed the influence of L1 on L2 and called it “L1 transfer” which involved both positive and negative effects of L1 on L2 learning (p 51) He stated that L1 transfer could be one of the sources of errors in second language acquisition Saville-Troike (2006) explained this negative influence to be the situation in which L1 knowledge was inappropriately applied to produce L2 utterances and therefore caused what was considered “error” (p 19)

Transfer was also regarded as communication strategy or learning strategy Ellis (1997) explained that when learners with insufficient language knowledge encountered communication, they developed communication strategies For example, they might borrow a L1 word to supply for a TL word they did not know Mahmoud (2000) reported transfer as a communication strategy or learning strategy, by which hypotheses about the target language could be formed and tested James (1998, p 188) mentioned calque (literal translation from L1 into L2) and language switch (from L2 to L1) as two L1 based communication strategies which resulted in language errors

Meanwhile, Mitchell and Myles (2004) explained L1 negative transfer by discussing L1 as a set of habit They claim that when learners acquire L1, they developed a “well-established” set of habits in language (p 31) When learning L2, learners had to adopt a new set of habits, therefore, they were presented with either benefits or difficulties due to either the similarities or differences between the two languages From this belief, contrastive analysis approach was introduced

2.2 Different approaches to second language acquisition

2.2.1 Contrastive analysis

In the 1950s, Contrastive Analysis became a popular theoretical view on the issue Lado (1957), a pioneer researcher on contrastive analysis theory, stated that contrastive analysis attempted to anticipate and made meaning of L2 learners‟ difficulties based on the comparison between their L1 and L2 The approach was based on Structuralism theory: CA compared L1 and L2 at each linguistic level to point out the types of interference of L1 in L2 (Lado, 1957, cited in Saville-Troike, 2006) Besides, CA was created with an influence of Behaviourism in which language learning was equated with the establishment of habits In other words, the greater difference between L1 and L2 was, the more challenges L2 the learner might encounter (Saville-Troike, 2006; Mitchell & Myles, 2004)

Originating from distinctive language families, Vietnamese and English features differed substantially Tran (2007) presented a documentation analysis where he compared description of English grammatical features of major linguists like N Chomsky, M A K Halliday with that of Vietnamese linguists, namely Diep Quang Ban, Cao Xuan Hao, or Hoang Trong Phien and so forth He analyzed the major similarities and differences in verbs, nouns and sentences in syntactic, semantic, pragmatic aspects It appeared that the two languages differed in various aspects of verbs: from the structural aspect such as verb tense, finite verb form, the use of “to be”, to the pragmatic and semantic aspects For nouns, features such as countable and uncountable nouns, noun complements were put into consideration (Tran, 2007) The study provided a comprehensive report of the disparity between two languages based on their commonly documented language features Other cross-linguistic studies also indicated meaningful comparisons of Vietnamese and English features For instance, Tang (2007) published a research on cross-linguistic features of the two languages on phonology and lexical semantics Another research by T Nguyen (2015) presented the difference between spoken and written English in comparison with Vietnamese, offering some valuable insights into Vietnamese adverbials, adjectivals and subordination

However, there existed some problems with a contrastive analysis approach More recent studies indicated that contrastive analysis did not accurately predict learner‟s issues Odlin (1989) stated that the sole comparison of two languages could not anticipate certain errors made by students in L2 products To be specific,

he brought up the circumstance in which Spanish EFL learners produced the utterance “that very simple” without the verb “to be”, though Spanish and English shared similar grammatical structure (p 18) Another problem with contrastive analysis approach was the scale of the task of comparing and contrasting two languages Norrish (1983) claimed that conducting a contrastive analysis of two languages was not an easy task; it was challenging to devise a consistent framework for contrast: a mere analysis of one sentence in L1 and its equivalence

in L2 could result in diverse interpretations

2.2.2 Error analysis

In 1967, Corder (1967) introduced Error Analysis (EA), a theory that focused on examining the errors made by learners in their target language EA was based on the idea that learning a language was not the memorization of rules but the ability to learn and use language resource to create new utterances (Saville-

Troike, 2006) Grami (2012) stated that EA offered more explanations of learners‟

errors than CA did, for the latter only predicted the errors caused by mother tongue interference and excluded other factors While CA analyzed the language

to predict learner‟s problem, EA regarded the learner – or more specifically, the learner‟s errors – as the subject of analysis Error analysis, thus, provided more insights into other sources of learners‟ errors compared to CA

One important concept of error analysis was how an error was defined Ellis (1994) defined error as the language element that deviated from the L2 target language It is important though, to distinguish different terms relating to errors Norrish (1983) presented the difference among an error, a mistake and a lapse An error referred to something learners constantly did incorrectly because of lack in

knowledge A mistake was a result of inconsistent use of a language item; learners might use it correctly or incorrectly at times A lapse resulted from other factors unrelated to the learning process such as lack of concentration or fatigue Ellis (1997) also emphasized the need to clarify errors and mistakes, suggesting that the consistency in learners‟ production of wrong language might assist the identification of errors and mistakes

The identification of errors claims its role in the process of teaching and learning L2 Since errors are produced by learners in their learning process, they reveal learners‟ ability and their progress This point was supported by Gass and Selinker (2001), who stated that errors can act as “red flags” which could reflect learner‟s progress in their L2 learning (p 78) Furthermore, Saville-Troike (2006) regarded errors as “the window to language learner‟s mind” (p 39), emphasizing the important information that errors could reveal about language learners Zaphar (2016) stated that only by identifying errors could learners resolve errors, making

EA an essential task in L2 learning All in all, investigating errors was of great vitality in improving the quality of L2 acquisition To achieve the purpose of this study, it has been pinpointed that Error Analysis is an appropriate approach to the research methodology

2.2.3 Error analysis process

Ellis (1997) suggests that there are a number of steps to apply in the procedure of error analysis

- Collecting samples of learner language

- Identifying the errors

- Describing the errors

- Explaining the errors

- Evaluating/correcting the errors

C ollecting samples of learner language: The first step of the process is to select the sample for analysis The type of EA can be classified according to the size of the sample which can be massive, specific or incidental Massive sample refers to several language samples from various students, while specific sample is

a collection of one sample from a limited number of students Incidental sample consists of one language sample from one student Samples can be collected either

at one single time or over a period of time Since learners‟ errors can be affected

by various factors, it is essential to consider language factor (medium, genre, content) and learner factor (level, mother tongue, language learning experience) in the process of collecting samples

Identifying the errors: Next, errors must be identified With the collected samples, the initial task is to determine the language elements that deviate from the L2 or the target language

Describing the errors: At this step, the errors can be categorized by linguistic levels (morphology, syntax, and vocabulary), grammatical categories, or the way they deviate from the reconstructed utterances in the target language (Ellis, 1994; Ellis, 1997; Saville-Troike, 2006)

Explaining the errors: When the errors are categorized, the important task

is to explain their sources of errors This process gives important insights into the origin of errors and create meaningful pedagogical implications According to Gass and Selinker (2001), there are two types of errors in error analysis: interlingual and intralingual errors While the former refers to L1-related errors, the latter refers to those related to the target language To be specific, interlingual errors, or transfer errors are regarded as the errors caused by mother tongue interference For example, the students can write a sentence in English with a Vietnamese structure

Evaluating the errors: After finding the origins of errors, the last step is to

analyze the effects that errors exert on people addressed The process involves identifying who to be addressed, what errors to be evaluated and the people to

evaluate errors Within the scope of this study, the researcher focuses on identifying and describing lexis and grammar errors

2.2.4 Error classification

The errors can be categorized according to linguistic level (morphology, syntax, and vocabulary), grammatical categories or the ways they deviate from the reconstructed utterances in the target language (Ellis, 1997) In describing errors in

a learner corpus, the Center for English Corpus Linguistics, Université Catholique

de Louvain devised an error tagging manual to annotate such problems This manual was created based on the definition of seven error taxonomies including form, grammar, lexico-grammar, lexis, word redundant, word missing and word order, register and style (Dagneaux et al., 1996, as cited in López , 2009) Each mentioned linguistic area involved more specific categories For example, in grammar, the errors were described by 15 categories involving problematic verbs, nouns, adjectives, adverbs and so forth

Table 1 Grammatical errors (Dagneaux, et al., 1996, cited in López , 2009)

This checklist was a useful tool for describing errors in a large-scaled learner corpus However, the list of lexical error appeared to be lacking in certain typical L1 influenced errors which are central to this research

Errors can also be investigated based on different linguistic elements James (1998) stated that errors could be categorized based on three main levels, namely substance, text and discourse The summary of the error categorization is presented in the table below

Table 2 Levels of errors (James, 1998)

Levels

Substance errors Mechanical errors (punctuation, typographic errors, etc.)

Text errors

Lexical errors Grammar errors Morphological errors

Syntax errors

Discourse errors

Coherence errors Pragmatic errors Receptive errors

Within the scope of this research, the research aims at investigating grammatical and lexical errors which both fall under the text error category

2.2.4.1 Lexis errors

Regarding error in lexis, James (1998) presented two subcategories which were formal errors and semantic errors Formal errors concerned problems with the forms, including formal misselection, misformation of words and distortion Semantic errors referred to meaning-related problems which comprised confusion of sense relations, collocation errors and stylistic errors The specific list is presented in the table below

Table 3 Types of lexical errors (James, 1998)

Category Types of errors Explanation

Formal

errors

Formal

misselection

Suffix type These errors refer to wrong

selection of words due to their similarities in sound or form These words may share the suffix, prefix, vowel, or consonant form (Laufer,

1992, cited in James, 1998) These errors include erroneous choice of word class

Prefix type Vowel based type

based type

Consonant-False friends

This refers to the instance when a formal resemblance of a L1 and L2 word leads to learner‟s misuse of it

Calque The learner use literally translates a

Misordering Learners put different parts into

Learners choose the wrong word in two co-hyponyms in a category

Using wrong synonyms

near-These errors refer to the wrong use of a close synonym to a target word, such as using “pharmacist” instead of “chemist” Collocation

errors

Semantic word selection

These are errors referring to the word combination that make

Statistically weighted preferences

Such errors concern problems in using words relating to statistical weight, like saying “big loss” is less appropriate than “heavy loss” Arbitrary

combinations

This refers to violations of fixed expression An example of error is

to say “make a try” instead of “make

an attempt”

Stylistic errors Verbosity This refers to wordiness and

unnecessarily lengthy expression Under-

Morphology errors referred to the failure to abide by the linguistic norm in

supplying any part of a word English had five lexical word types: noun, verb, adjective, adverb and preposition; however, preposition was regarded as having no morphology Therefore, instance of morphology errors could be observed in the rest of the word classes Some examples of such errors may be:

Noun: abolishment [correct: abolition]

Verb: breaked [correct: broke]

Adjective: beautifuller picture [more beautiful picture]

Adverb: see you soonly [correct: see you soon]

Syntax errors included the observed deviation from the linguistic norm in texts

“larger” than word structure such as phrase structure, clause, sentence and a group

of sentences

- Phrase structure errors: this type of error concerned problems in using phrases

In conducting error analysis, James (1998) suggested that the Qualifier model was employed to analyze this error type An example is given below

Modifier-Head-a cleverest boy in the class

Modifier 1 Modifier 2 Head Qualifier

We can see that the modifier 1 was problematic since the phrase was in superlative form In this case, the article at the position of modifier 1 was inappropriately selected

- Clause errors: these errors referred to the violation of norms in clause structures The errors could occur in the following conditions:

Superfluous phrase: He shaved himself the beard

Omitting: give to the dog [correct: give + noun phrase + to the dog]

Misordering: Watson sent to him the letter [correct: Watson sent the letter to him]

Misselecting: he seems crying [correct: to cry]

Blending: you would be likely to get [correct: You would most likely get]

or “along similar lines – according to similar law”

- Sentence structure errors: These errors occurred when clauses were selected and combined into larger units

- Inter-sentence structure errors: These errors involved cohesion problems which arose in the relationships among sentences These errors could be observed in five types of cohesive links identified by Halliday and Hasan (1976) including reference, substitution, ellipsis, conjunction and lexical cohesion (cited in James, 1998)

Among the listed error description framework, James (1998) provided a detailed specification of lexis errors by including errors from different aspects of language and ones resulted from interlingual and intralingual influence Nevertheless, his description of grammatical errors appears to be of a broader sense since the errors were classified under linguistic levels Adapting both the error tagging manual by Dagneaux (1996, cited in López, 2009) and James (1998)‟s frameworks to support the data analysis process seemed to be the optimal choice, for this research put emphasis on L1 interference This process was further elaborated in Chapter 3: Methodology

2.3 Overview of previous studies

2.3.1 Studies of L1 interference

L1 transfer could be observed in the production of English, namely learners‟ oral and written products Certain versions of language, which were intelligible in spoken language, may not be comprehended in written production Therefore, unlike spoken language, written discourse required a higher level of rigidity in grammar language use (T Nguyen, 2015) The errors found in written products, thus, were generally regarded as indicators of language learners‟ lack (Norrish, 1983) Various researchers have put effort in studying the influence of L1 on L2 writing Table 4 briefly summarized some recently conducted studies on the issue

Table 4 Overview of related studies

Study -

Author

First language

Research problems

Method Result / Conclusion

to three influencing factors: language proficiency, topic and writing modes

Writing compositions were collected

Instances of L1 impact were categorized and statistically analyzed according to three mentioned factors

Language proficiency, topic and modes are three factors affecting students‟ use of L1

in L2 writing; language problems result from difference in L1 (Arabic) and L2 (English); students use L1 in L2 irrespective of their L2 proficiency González,

2016

Spanish The study aimed to

identify most common L1 interference errors, examine error theories, identify error patterns for pedagogical solutions

A small corpus

of English learners at a Spanish high school was collected Error analysis method was employed during which

L1 interference and interlanguage were causes of errors Existing errors provided important pedagogical

benefits

errors were identified and calculated

Contrastive analysis and error analysis:

error were identified and classified

Statistical analysis were employed

Difference between the structures of L1 and L2 influenced the learners‟ L2 writing

Wang, 2015 Chinese The study aimed to

calculate the most frequent attributive clause errors caused

by negative transfer and how L1 interference

influenced Chinese students‟ writing

Error analysis:

collected writing pieces were examined for

identification of wrong

attributive clauses

The study found 5 error types caused

by negative L1 transfer L1 consciously and unconsciously influenced students‟ writing

Zaphar,

2016

Urdu The research aimed

to explore if remedial lessons based on error analysis could making English learning more effective

This study followed

experimental research design

The process included pre-test, remedial lessons, post-test on a group

of Pakistani students, then the test results were compared

Remedial lessons (including frequent practice of grammar rules) were effective

in reducing writing errors Error analysis was useful in pinpointing the problematic

language areas

Hussein and Mohammad (2011) conducted a study to investigate L1 Impact

on L2 Writing The study involves the participation of sixteen university students whose L1 was Arabic The subjects wrote 250-300 word essays responding to about 30 topics under three writing modes: process, comparison and contrast, and opinion The results of the study included the instances of L1 influence on L2 writing in different topics and modes of writing Through the calculations of errors,

the researcher concluded that writing mode, topic and student level of language proficiency were three factors affecting the negative influence of L1 on L2 More familiar topics and modes to students seemed to witness more L1 influenced traces

A more recent research is González‟s error analysis in Spanish EFL acquisition (González, 2016) The research collected 79 written compositions, each of which was composed by one participant including secondary students, high school students and adults learning EFL This study‟s result was a list of 20 common grammatical errors made in the writing, from which the researcher offered pedagogical suggestions The most common errors, namely misuse of pronouns, lack of auxiliary verbs, subject omission and so on were categorized into the grammatical errors Timina (2013) conducted a close investigation into the errors made in English compositions of Taiwanese EFL majored students (Chinese speakers) To identify the most common errors caused by the rhetorical difference between L1 Chinese and L2 English and the errors in grammar and lexis, the study examined errors in more aspects than the previously discussed one, including rhetorical, lexical and grammatical aspects The study adopted error analysis process with the support of contrastive analysis in comparing characteristics of L1 and L2 to find rhetorical errors It identified errors in essay organization structure, content and argumentation resulting from the difference English and Chinese‟s rhetorical characteristics Besides, there were also instances of grammatical and lexical errors such as verb tense, wrong word, incomplete structures or sentences This revealed that L1 and L2 difference significantly interferes with students‟ writing

Another study by Wang (2015) investigated the most common attributive clause errors caused by negative transfer and how L1 interference influenced Chinese students‟ writing 60 writing pieces were collected from sophomore English majors in a Chinese university for examination and identification of errors Since the researcher focused on clause errors, the error analysis process was

presented in depth The author first calculated students‟ use of complex sentences, classified the clause types and finally calculated the error frequencies The result showed that more than 80% of the errors from five error categories stem from mother tongue influence The limitation of this study was that only one writing piece was collected from each student, which might not represent their true competence

Regarding benefits of error analysis, Zaphar (2016) presented a further developed research on the effect of remedial teaching strategies based on error analysis The study adopted experimental research design in which written essays

of 37 Pakistani ESL students were examined to identify the common errors The students subsequently underwent a pre-test, 8-week training and post-test The findings indicated that students‟ most common errors were found in present and past simple tense and were classified into three categories: Overextension of analogy, Interlingual (L1) and intralingual (L2) errors 154 L1 interference errors were found in the pre-test, ranking second among the error categories The result

of the post-test show a decrease to 94 writing errors, revealing that error analysis was a useful process in developing remedial teaching strategies

2.3.2 Studies of Vietnamese influence on ESL learners

In Vietnamese context, there exist several related studies on the L1 influence on the acquisition of English as a second language K Nguyen (2015) presented phonological and grammatical differences between English and Vietnamese, and discussed language transfer as the factors influencing English acquisition He presented the influence of Vietnamese on English acquisitions of people in Hanoi in two areas of language: pronunciation and grammar In terms of grammar, the linguistic difference in noun phrase structures and verb tenses were discussed as influential factors to English acquisition It was concluded that Vietnamese usually made errors in using English noun phrases since the structure

in Vietnamese was somehow simpler An example would be word order:

Vietnamese were familiar with using adjectives after nouns, while English nouns needed to be complemented by prior adjectives An instance is presented below

English: picturesque ancient pentagonal crumbling church towers

Main noun: towers

Vietnamese: tháp nhà thờ cổ kính ngoạn mục hình ngũ giác dáng xiêu vẹo

“When I was small, I often swim in the river alone” („swam‟ should be the correct

verb) Besides, lexical errors were also identified but not specifically categorized

Pham (2016) presented a research on the interference of Vietnamese on ESL learner at university level He gathered a total of 120 letters, narrative and descriptive writings from 40 university students to analyze the errors caused by negative transfer The study reported that verb tense errors were dominant in narrative writing, while word choice and subject-verb agreement were the most problematic errors in descriptive and letter writing (20.35% and 19.52%) Notably, omission of “to be” appeared to be a serious problem with more than 100 instances,

which was attributed to the absence of this verb‟s equivalent in Vietnamese Regarding lexis, wrong word selection prevailed such as the use “big rain” for

“heavy rain” or “wide shoulder” for “broad shoulders” Redundant word expressions resulted from L1 were also documented For example, “prohibit not to smoke” was used instead of “prohibit smoking” Vu (2017) also investigated negative transfer on English writing of 15 Vietnamese tertiary EFL students at intermediate English level, in which interlingual errors were identified in the use

of word forms (57.6%), articles (26.1%), prepositions (11.0%), pronouns (4.3%) and possessive case (1%) The study pointed out that negative influence L1 could result in underproduction, overproduction and production errors

The participants of these studies were respectively high-school students, non-majored and intermediate college students It was astounding that English majored students were not common subjects of such studies, while these students had intensive exposure to academic written English With a demanding English study program, these students possibly encountered various challenges owing to their mother tongue, and thus can be a resourceful group of participants for research Besides, these studies solely focused on error analysis without probing into learners‟ perceptions on the issue Hence, this research, which targets English majors as the subject, hopes to bridge the gap in the existing literature

CHAPTER 3: METHODOLOGY

The following chapter presents the methodology of this research paper It includes the description of the subject, the justification for the data collection instruments and depiction of data collection and data analysis process

3.1 Research design

This research employed a mixed-method approach to research design, incorporating both quantitative and qualitative research In this study, quantitative data was first obtained through error analysis process, and the subsequent interview offered reasons perceived by participants for the identified errors This combination of two approaches could limit each one‟s drawbacks and provided a more comprehensive understanding of the research problem (Creswell, 2014) Mixing two methods also allowed triangulation which augmented the validity and richness of the research result (Zohrabi, 2013) Besides, mixed method approach was claimed to provide insights into identified trends and offer answer to the

“what” and “why” questions (Ivankova & Creswell, 2009, p 145) Since this study included identification of L1 errors (the “what”) and their underlying reasons (the

“why”), mix-method approach appeared to be the optimal choice of research design

3.2 Participants

Teachers: Prior to the data collection stage, the researcher contacted 08

teachers of the sophomore students to ask if their students made a lot of errors related to L1 influence Based on the responding teachers‟ opinions, the researcher decided to conduct the study on three classes of students that were reported to commit many errors influenced by mother tongue Two teachers of these participants were invited to categorize the error types (later specified in the data collection process) These two female teachers were direct supervisors of students‟ writing lessons during the studied period Before categorization, the two teachers

attended meetings with the researcher to thoroughly understand the error categorization chart The researcher was in charge of giving explanations for the each error types to ensure consistent understanding of both teachers

Students: The subjects of this study were 53 second-year students in a

language education university in Hanoi According to their study program description, all the participants were English majored students with an expected proficiency being B2 level of the CEFR framework after semester 3 of the program These students aimed at achieving C1 level at the end of the studied semester (semester 4) Since the time of data collection was the earlier half of semester 4, the students‟ competence was estimated to be at B2+ level

3.3 Data collection

3.3.1 Data collection instrument

Written essays: The collected documents were argumentative essays

written in class by sophomore students These essays were written as part of a test preparation course, which means that the writing condition resembled that of a real standardized test Each essay was written in class under a time limit of 45 minutes

A total of 106 compositions were collected from the participants in one month

Interview: The research invited 10 students with most L1 influenced errors

to participate the interview The researcher conducted a semi-structure interview with each of these students to investigate their perception on errors influenced by their mother tongue According to Mackey and Gass (2005), interview was a useful instrument to explore unobservable phenomena The instrument also provided confirmation and valid interpretation (Johnson & Turner, 2003) In this case, an interview with the students was useful in exploring their thought on error causes and L1 influence, which were not transparent via the written essays The interactive form of the interview allowed the researcher to ask for explanation of

vague or over-general answer (Mackey & Gass, 2005)

3.3.2 Data collection procedure

The data collection process will be divided in three phases:

Phase 1: The researcher asked for teachers‟ opinions and chose the participants The researcher asked for students‟ consent and collected their essays

Phase 2: The pilot study was conducted on 20 chosen essays The data was coded two times based on two framework by Dagneaux et al (1996, cited in López , 2009) and James (1998) The result showed a necessity to combine these two frameworks so that the study could present adequate error types identified in the written compositions

Phase 3: Error analysis process was employed during the analysis of writing pieces For objective and valid results, two teachers of the participants were invited to examine all the essays against the given categorization table of errors They were responsible for highlighting the L1 interference errors in the essays and categorizing them into certain types All student names were kept anonymous from the graders during the process to avoid possible bias If any disagreement arose between two raters on one error, they discussed to finalize one final decision

Phase 4: The semi-structured interview was conducted with each of the 10 students with the biggest number of errors The researcher was responsible for clarifying any possible questions and confusion in the interview The interview focused on exploring students‟ awareness of all the identified L1 interference errors, the possible reasons for the error All of the interviews were conducted in Vietnamese to enable students to easily elaborate their thought and avoid misunderstanding During the interview, the researcher showed the students their own writing pieces with marked errors and inquired what they were thinking when writing their essays Sometimes, on-spot clarifications were requested from the

researcher to confirm the interview data The interview question sets could be found in Appendix 2

3.4 Data analysis

The research employed a mixed-methods approach to data analysis procedure Firstly, quantitative analysis was conducted to find out the frequencies

of the found errors and their statistical weight against the total number of errors

In terms of the interview results, thematic analysis was employed to classify the data into different themes Thematic analysis was described as a flexible analysis method in investigating perceptions and pinpointing similar and contrasting features (Nowell, Norris, White & Moules, 2017) Appropriate interpretations were made using the analyzed data from the interview

3.4.1 Error analysis

106 writing compositions of around 250-300 words were collected for analysis The teachers identified the grammatical and lexical errors in each piece and tagged them with specific labels referring to their error categories

The error classification list was crafted based on the different text errors described in James‟ (1998) framework It is important to note that only grammatical and lexical error types resulting from L1 were included in the final categorization table Therefore, lexis distortion errors described by James (1998), which solely referred to the errors influenced by the target language, were excluded from the checklist However, as previously mentioned, his description of error levels seemed to be overly general to categorize specific errors in students‟ writing This raised the need to develop a checklist with more specific types of grammatical errors Meanwhile, the error tagging manual by Dagneaux et al (1996) originally concerned different taxonomy of language including form, grammar, lexico-grammar, word order/redundant/missing, lexis, register and style

(cited in López , 2009) However, as stated in chapter 2, this lexical error list did not seem to cover all L1-related lexical errors

Thus, a pilot study was conducted on 20 compositions with a view to

devising a more detailed list of errors The pilot study showed that students made

L1-related grammar errors which were described in the error tagging manual by Dagneaux et al (1996, cited in López , 2009) Based on this result, the researcher

decided to utilize the list of grammar errors from this framework Observably,

certain overlapping error descriptions could be detected in the two checklists, so the researcher did make some alterations to devise the final one In other words, similar errors listed in two frameworks were described as one Besides, applying the suggestion of Hemchua and Schmitt (2006), the researcher added preposition partner error under the collocation error category Lastly, “omission of „to be” was added to the list since a high frequency of such errors were found in the pilot study This error type was also identified in other studies such as Ha (2011), Pham (2016) Besides, Tran (2007) claimed that omitting “to be” before an adjective is a unique characteristic of Vietnamese It is important to note that this type only refers to omission of “to be” as the main verb The final error categorization chart

is presented in Table 5 below

Table 5 Error categorization chart

Using general term for specific term

28

Semantic error - Collocation errors

Semantic word selection

After being categorized, the total number of L1 errors of each type were calculated through mechanical counting The quantity of each error was then ranked in terms of number and percentage (against the total number of errors), and was presented in a set of charts and graphs The specific examples of each error were also presented accordingly to supply the demonstration of each error

3.4.2 Interview data

The coded interview responses from students were used to report the reasons for different types of students‟ errors Students‟ phrases expressing the error reasons were extracted from the interview transcripts; among which the related answers were grouped into themes In particular, answers such as

“translation”, “word-by-word transfer” were put under “literal translation” theme After that, trivial answers that fited in no themes were also recorded These responses were also compared with error analysis result to confirm if L1 were

exactly the sources of the founded errors This task of triangulating data aimed to facilitate the validity of the research result The researcher came to conclude about the most common interference errors, synthesized students‟ viewpoints on the reasons for their errors and drew out pedagogical implications to improve the situation

CHAPTER 4 FINDINGS AND DISCUSSION

This chapter presents the collected data via error analysis and interview and discussion of research results in light of existing literature In answering research question 1, quantitative data are reported to specify the errors made as a result of mother tongue interference The interview provides response to research question

2 about the reasons for commitment of L1 interference errors Since research question 2 helps to clarify question 1, the data from error analysis and student interviews are combined for better demonstration of the findings The report is followed by a discussion of the result where comparisons and contrast are made in relation to previous studies

4.1 Overview of found errors

After the investigation into the essays, the two raters found 494 L1 influenced errors, with an average of 4.66 errors in each of the 106 examined essays During the error analysis, the raters only took notes of the errors caused by L1 influence and excluded errors resulted from other sources Table 6 below presented the detailed number and rate of each error against the total To be specific, the number of errors were found throughout mechanical counting, when the error rate was calculated by the following formula: