Raising the Concern with Written Error Gravity in the Foreign Language Classroom. Reflecting on The Tunisian Context

Textbook , 2026 , 97 Pages

Didactics for the subject English - Pedagogy, Literature Studies

The objective of this study is to examine how written errors are evaluated in English as a Foreign Language (EFL) classrooms, with particular emphasis on the role of teaching experience and assessment criteria in shaping judgments of error gravity. The research is situated in the Tunisian higher education context, where systematic guidelines for prioritizing written errors remain limited.

Grounded in applied linguistics and pedagogical research, the study explores how teachers perceive, categorize, and respond to learners’ written errors. It adopts a mixed-methods approach combining quantitative and qualitative data. Fifty essays written by third-year management students were analyzed following an established error analysis model. In addition, twenty Tunisian EFL teachers participated in a scale-based questionnaire, while a subset of teachers contributed to semi-structured interviews, allowing for deeper insights into their evaluative reasoning.

Table of contents

Dedication

Abstract

List of tables

List of abbreviations

CHAPTER 1 : Introduction
1.0. Introduction
1.1. Background of the study
1.2. Statement of the problem
1.1. 3. Research objectives
1.4. Research questions
1.5. Research hypotheses
1.6. Structure

CHAPTER 2 : Literature review
2.0. Introduction
2.1. Defining errors
2.1.1. Error Analysis:
2.1.2. Methodology of error analysis
2.1.2.1. Identification of errors
2.2.2. Description of errors
2.4.2.4. Evaluation of error seriousness
2.1.3. Benefits of error analysis
2.1.4. Criticism of Error Analysis
2.2. Error Treatment in writing
2.2.1. Definition of Error Treatment
2.2.2. Overview of Feedback in Learning Theory and Instructional Design
2.2.3. Feedback in the Teaching of ESL/EFL Writing
2.2.4. Issues of Error Treatment in SLA
2.2.4.1. , Negative Perspectives of Error Treatment
2.2.4.2. Positive Perspectives of Error Treatment
2.3. Error Gravity Research
2.3.1. . Which error to prioritize?
2.3.1.1. , Errors that are relevant to the pedagogical focus
2.3.1.2. Errors that occur frequently
2.3.1.3. Errors that hinder intelligibility
2.3.1.4. . Errors that affect teachers’ acceptability
2.3.1.5. Errors that cause irritation
2.3.1.6. . Errors that affect comprehensibility
2.3.2. How to correct errors?
2.3.2.1. Direct VS. Indirect Corrective Feedback
2.3.2.2. Focused Vs. Unfocused Corrective Feedback
2.3.2.3. Reformulation
2.3.3. Who should correct learners’ errors?
2.3.3.1._______ Native versus non ... native speaker
2.3.3.2. Experienced versus novice teachers
2.3.4. . Criticism of error gravity research

CHAPTER 3 : Methodology
3.1. Introduction
3.2. Corpus of the study
3.3. Participants
3.3.1. Profile of the participants
3.3.2. Population selection
3.4. Methodology
3.4.1. Type of the study
3.4.2. Procedures
3.4.3. . Instruments of data collection
3.4.3.1 Teachers’ questionnaire
3.4.3.1.1 Rationale and Objectives
3.4.3.1.2 Questionnaire design considerations
3.4.3.1.3 Description of the questionnaire
3.4.3.2 . Teachers’ interview
3.3.3.2 . 1Rationale and objectives
3.3.3.3 .2 Description of the interview
3.5. Summary

CHAPTER 4 : Findings and discussion
4.1. Introduction
4.2. Findings
4.2.1. The impact of teachers’ experience on error gravity judgment variation of the five error categories
4.2.1.1. Analysis of STETs and LTETs judgment of error gravity using Acceptability criterion
4.2.1.2. Analysis of STETs and LTETs judgment of error gravity using Irritation criterion
4.2.2. The impact of criteria selection on teachers’ error gravity judgment variation of the five error categories
4.2.2.1. . Analysis of STET’s judgment of error gravity using acceptability and irritation criteria
4.2.2.2. Analysis of LTETs judgment of error gravity using acceptability and irritation criteria
4.2.3. The impact of teaching experience and criteria selection in shaping severity hierarchies of the five error categories
4.2.3.1. Analysis of the hierarchical order of errors according to acceptability criterion for the two groups
4.2.3.2. Analysis of the hierarchical order of errors as judged by two groups of teachers according to irritation criterion
4.2.3.3. Analysis of the hierarchical order of errors according to acceptability and irritation criteria of STETs
4.2.3.4. Analysis of the hierarchical order of errors according to acceptability and irritation criteria of LTETs
4.2.4. Analysis of the data provided by teachers’ semi- structured interview
4.3. Discussion
4.3.1. Teachers’ questionnaire
4.3.1.1. The effect of teachers' experience on their error gravity judgment of the five error categories
4.3.1.2. The effect of criteria selection on teachers’ error gravity judgment of the five error categories
4.3.1.3. The impact of teaching experience and criteria selection in shaping severity hierarchies of the five error categories
4.3.2. Teachers’ interview

CHAPTER 5 : Conclusion
5.1. The main findings
5.2. Research limitations
5.3. Implications for teaching
5.4. Conclusion

Appendices

Dedication

I dedicate this work to those who hold palpitating hearts full of linguistics...

Abstract

The English as a Foreign Language (EFL) composition teacher’s problems center on identification, correction, and evaluation. “The first two tasks are relatively easy to handle, but the third one is not and has only recently begun to be addressed in studies of error gravity. The difficulty in judging student errors is that the teacher needs guidelines for determining their seriousness, but these do not presently exist. » (Sheorey, 1986: 306). In order to judge the gravity of written errors, teachers generally resort to different sets of criteria. In addition, the majority of error gravity research studies focus on the judgment of native versus non native speakers. However, little concern has been paid to the role of experience in shaping gravity judgment in the Tunisian context. Therefore, this research attempted to investigate the impact of experience and criteria selection in shaping teachers’ judgment of the gravity of written errors. Following Keshavarz’s model (2012) of error analysis, 50 essays of third year management students were analysed. Twenty Tunisian teachers participated in the evaluation process. Teachers were classified according to their experience to long teaching experience (LTETs) and short teaching experience teachers (STETs). In order to gauge their judgments, they were to react on a scale item questionnaire using two different criteria: irritation and acceptability. Moreover six teachers with different experience participated in a semi-structured interview. The ultimate objective was to raise teachers’ awareness of prioritizing errors while testing and marking through majoring the impact of two main variables: teaching experience and criteria selection on the prioritization process. In fact, this research relied on both qualitative and quantitative research methods. The quantitative method used the scale item questionnaire, which was distributed to two different groups of teachers: 10 (STETs) and 10 (LTETs), to major their reactions to the seriousness of the existing errors. However, the qualitative method is based on a semi­structured interview, which is addressed to four teachers: 2 (STETs) and 2 (LTETs). The findings showed that teachers tended to behave differently in reacting to the students’ written errors. Therefore, they differed in terms of leniency/severity dichotomy in assessing the gravity of the errors. Generally speaking, LTETs seemed to be more lenient in evaluating students’ errors than STETs in accounting for two criteria: acceptability and irritation. Finally, the study ends up by six different error hierarchies ranging from the most serious to the least serious error types and categories.

List of abbreviations

ESL: English as a Second Language

SLA: Second Language Acquisition

EFL: English as a Foreign Language

L2: Second Language

EA: Error Analysis

IL: Interlanguage

CF: Corrective Feedback

EG: Error Gravity

NSTs: Native Speaker Teachers

NNSTs: Non- Native Speaker Teachers

STETs: Short Teaching Experience Teachers

LTETs: Long Teaching Experience Teachers

List of tables

Table 2.1: Celce-Murcia’s (1985) table of More Effective and Less Effective Teacher Correction Strategies

Table 2.2: Ferris and Robert’s (2001) Description of Error Categories

Table 2.3: Error Chart

Table 3.1: Profile of long- teaching- experience teachers

Table 3.2: Profile of short- teaching-experience teachers

Table 3.3: Error Analysis research procedures

Table 4.1: xamples of morphological errors

Table 4.2: Examples of Lexical errors

Table 4.3: Examples of syntactic errors

Table 4.4: Examples of Mechanical errors

Table 4.5: Examples of Miscellaneous errors

Table 4.6: Descriptive Statistics for Errors Severity scoring of error categories in Acceptability criterion

Table 4.7: Descriptive Statistics for scoring severity of error categories in Irritation criterion

Table 4.8: Descriptive Statistics for STETs judgment of Acceptability and irritation criteria

Table 4.9: Descriptive Statistics for LTETs judgment of Acceptability and irritation criteria

Table 4.10: Error gravity hierarchy of STETs group in Acceptability criterion

Table 4.11: Error gravity hierarchy of LTETs group in Acceptability criterion

Table 4.12: Error gravity hierarchy of LTETs group in Irritation criterion

Table 4.13: Comparative error gravity Hierarchies of LTETs group in both criteria

Table 4.14: Box’s M test Results

Table 4.15: Discriminant Analyses: Classification Results

CHAPTER 1

INTRODUCTION

1.0. Introduction

The first chapter introduces the whole book. First, this introductory part gives an overview of the background of the study, the statement of the problem, the research objectives, questions, and hypotheses, and finally, the structure. As for the Second chapter, it provides the relevant literature concerning error treatment in writing and the basic concepts related to the research topic. Concerning the third part, it will explain the methodological tools used to collect data about teachers' variability in treating students' written errors. The fourth chapter has to do with data analysis and the discussion of the findings. The last chapter will summarize the whole book, provide some pedagogical implications, and it will present the study limitations for future research recommendations.

1.1. Background of the study

The present study investigates variability among teachers in assessing the gravity of language errors found in the written exams of third-year management students. The interest in undertaking this error analysis grew out of the different remarks made by teachers concerning students' writing performance. EFL Teachers in general and EFL teachers in the English department in particular often complain that at an advanced academic level students produce different error types in their written performance. Therefore, in writing "the back-and­forth negotiation is not possible; there is only ‘one pass A'. Because there is no possibility of negotiating the meaning of written documents, the inevitable problems of misunderstandings are exacerbated "(Abisamra 2003: 3-4)

This work is an exercise in applied linguistics. It can be positioned within the cognitive approach to investigate foreign language learning. It is also anchored in the pedagogical studies concerned with teaching and improving classroom practices. It adopts the active research where the teacher positions her-/ himself as an insider observer of the learners’ behavior. The main focus is on the analysis of the learners’ written output with a particular attention to the erroneous productions. The ultimate objective would be to better understand the learning process and facilitate the teaching operation.

1.2. Statement of the problem

To facilitate the learners’ writing task easier and more effective, teachers should have some priorities in dealing with written errors. Therefore, treating errors according to hierarchical order enables the teacher to prioritize his teaching objectives. Hence, this research will be an attempt to investigate teachers’ differences in reacting to the study written errors in order to see if there is a correlation between teachers’ experience and their tolerance of certain errors in accounting for two different evaluation criteria.

1.3. Research objectives

1. To investigate teachers’ conception of learners’ errors.

2. To understand how teachers react to learners’ errors when detecting them in their writings.

3. To gauge teachers’ tolerance of learners’ errors.

4. To determine teachers’ categorization of learners’ errors.

5. To assess whether such categorization of learners’ errors has an implication on teaching and syllabus design. Therefore, prioritizing errors lets the teacher know which error to tackle first. This informs teaching because it affects syllabus and curriculum design, teaching and evaluating methods (feed-forward instead of feedback) and decision making. Accordingly, this research seeks to shed light on the behavior of a specific community of teachers in relation to learners’ errors in writing. It is expected that new perspectives in syllabus design and pedagogical approach be recommended.

1.4. Research questions

With the purpose to reach the objectives stated above, this study tries to answer the following questions:

1. How do STETs and LTETs react to the written errors according to their types in the light of the teaching experience?

2. How will STETs and LTETs react to the written errors according to their types in the light of criteria selection?

3. How will written errors be ranked by STETs and LTETs in light of teaching experience and criteria selection?

4. What are the teachers’ attitudes towards error prioritization as measured by teachers’ interviews?

1.5. Research hypotheses

The main hypothesis of this study is that teachers’ judgmental differences in evaluating learners’ written output are due not only to teaching experience, but also to the angle that we choose to perceive error from.

1- Positive directional hypothesis:

There are a lot of differences among teachers in responding to the gravity of written errors.

2- Null hypothesis:

There are no obvious differences among teachers in treating the seriousness of the written errors.

1.6. Structure

The present study consists of five chapters. The first chapter will be reserved for the introduction, which includes the background to the study, the statement of the problem, the research objectives, questions and the hypotheses. The last point in this chapter is about the organization of the book. The second chapter will be dedicated to undertaking a review of the relevant literature dealing with the writing skill, and the major works about teachers’ differences in reacting to written errors. As for the third chapter, it will shed light on methodology. It will present the reasons behind selecting the subjects, their profile and the different methods of data collection and the procedures used in the present study. The forth chapter will be devoted to the results and the discussion of the findings. Finally, the concluding part of the study will sum up the main findings and provide some implications for teaching. The chapter ends with the limitations of the study without losing sight of certain recommendations for future research.

CHAPTER 2

LITERATURE REVIEW

2.0. Introduction

Reviewing the literature about error treatment and weighing the gravity of the written error is the concern of the following section.

2.1. Defining errors

Terms like ‘mistake’ and ‘error’ have sometimes been used interchangeably. These terms are associated with Corder (1967) pioneered in establishing a distinction between errors on the one hand and mistakes or lapses on the other hand. Thus, errors are more deeply anchored in the brain. Therefore, they are related to competence, while mistakes happen because of lack of attention or distraction due to tiredness, excitement, sadness or fear. In the same vein, Richards (1984:95) states that a mistake is made by a learner when writing or speaking which is caused by the lack of aspects of performance like fatigue, whereas an error is caused by the lack of competence.

Norrish (1983) has another distinction between errors and mistakes. For him, errors are "a systematic deviation when a learner has not learned something and consistently gets it wrong." Norrish (1987:8) defined a mistake as an inconsistent deviation, which means that the learner fluctuates between the right and the wrong." Richards and Schmidt (2002:184) approach error from the vantage point of the native speaker, i.e., the way s/he regards incomplete learning. An error refers to a systematic error of competence, both covert and overt, that deviates from the norms of the target language (Eun-pyo, 2002:1). Ellis (1996:710) distinguishes between covert and overt errors on the basis of Corder’s (1971) model. They define covert errors to be grammatically correct but not interpretable within the context of communication, whereas overt errors refer to the obviously ungrammatical utterances. In addition, Cunningsworth (1987:87) adds that errors are systematic deviations from the norms of the language being learned.

According to Ellis (2003: 17), “errors reflect gaps in a learner’s knowledge; they occur because the learner does not know what is correct”. In other words, they occur systematically and they reflect the learner’s incomplete knowledge. Errors are related to competence, whereas mistakes are related to performance. Ellis (2003: 17) claims that mistakes “reflect occasional lapses in performance”.

These lapses could be slips of the tongue, slips of the pen, or slips of the ear. The latter "may provide some clues to how the brain tries to make sense of the auditory signal it receives" (Yule, 2010:161). As an example, "great ape may be heard as grey tape".

Thus, drawing clear cut boundaries between errors and mistakes has always been problematic for (EFL) teachers. Moreover, deciding upon the criterion to adopt in making this distinction is more problematic. Therefore, James (1998) has opted for three criteria. The first was the criterion of self-correctability. So, learners are able to correct their mistakes, but they are unable to correct their errors to the exception of advanced learners. The second criterion was consistency, i.e., the learner commits an error if s/he consistently uses a deviant form for a correct target language (TL) but if the deviant forms are random they are mistakes. The third criterion is the frequency of occurrence. That is, errors with low frequency are considered to be mistakes and those with high frequency are errors (Sari& Mulyani, 2020).

2.1.1. Error Analysis:

Since the late 60s, there has been a raising concern about learners’ errors. Therefore, the contrastive analysis with its two versions took place to explain why some features of a target language were more difficult to acquire than others. However, it does not provide an in depth analysis and explanation of students’ errors (Esser, 1980). Therefore, Error analysis comes as a more focused method in approaching learners’ errors. Pit Corder is the ‘Father’ of Error Analysis (EA). It is in his article entitled “The significance of Learner Errors” (1967) that Error Analysis took a new turn. Errors used to be ‘flaws’ that needed to be eradicated. Therefore, errors used to be seen negatively in the area of language learning and teaching. Analyzing learners’ errors systematically paths the way to determine areas that need reinforcement in teaching (Coder , 1967).Richards and Schmidt (2002:184) define it as “The study and analysis of the errors made by second language learners.” By the same token, Crystal (2003:165) claims that error analysis is “a technique for identifying, classifying and systematically interpreting the unacceptable forms produced by someone learning a foreign language”.

Error Analysis is the examination of those errors produced by students in both the spoken and written medium. Corder, (1967) writes:

The study of error is part of the investigation of the process of language learning. In this respect, it resembles methodologically the study of the acquisition of the mother tongue. It provides us with a picture of the linguistic development of a learner and may give us indications as to the learning process. (Corder, 1967: 125)

This quote highlights the importance of studying learners’ errors because the latter reflect the process of learning.

In line with corder’s (1967) thoughts, Ellis (2003) states that:

There are good reasons for focusing on errors. First, they are a conspicuous feature of learner language, raising the important question of ‘Why do learners make errors?’ Second, it is useful for teachers to know what errors learners make. Third, paradoxically, it is possible that making errors may actually help learners to learn when they self-correct the errors they make. (Ellis, 2003: 15)

This quote explains the theoretical and the practical importance of studying learners’ errors. By the same token, Corder, (1974: 122-154) argues that error analysis has two main objectives: one theoretical and another applied. The theoretical object serves to ‘elucidate what and how a learner learns when s/he studies a second language’ (Corder, 1974:123). And the applied object serves to enable the learner ‘to learn more efficiently by exploiting our knowledge of his dialect for pedagogical purposes’ (Corder, 1974:123).

Corder (1974:125) notes that Error Analysis (EA) is useful in second language learning because it draws the attention of teachers, syllabus designers, and textbook writers to errors which ".. can tell the teacher how far towards the goal the learner has progressed and consequently, what remains for him or her to learn". So, students’" errors are valuable feedbacks” (Xie & Jiang, 2007:13), because they are stone miles or “red flags” that provide evidence of the learner’s knowledge of the second language(Selinker,1972).Therefore, one of the main aims of error analysis is to help teachers to be more focused on the assessment of their learners' productions. As a matter of fact, remedial work would be vital to help students avoid the most common errors.

Richards et al.(1996:127)focused on the identification of learning strategies, to look for the causes of learner’s errors, which help in determining universal hardships in language learning. Similarly, Michaelides, (1990:30) asserts that the systematic analyzing students’ errors systematically holds a great importance for students, researchers, and teachers. The latter can have a more reliable picture of his students’ beliefs and practice of the target language system. Willcott (1972:73) conducted an error analysis to diagnose Arabic native speakers’ problems with the syntax of written English.

This study is going to focus on the study of errors rather than mistakes because the former reflect Students’ competence. In fact, studying errors related to competence enables the teacher and the researcher to understand the cognitive mechanisms and processes that cause or prevent errors.

2.1.2. Methodology of error analysis

Keshavarz’s (2012) provided a four step to evaluate error seriousness. Hence, the following four-step approach was designed:

• Identification of errors
• Description of errors
• Explanation of errors
• Evaluation of error seriousness

For practical reasons, this study duplicated this approach to investigate EA in the Tunisian context.

Keshavarz (2012:168) asserted that EA is “a procedure used by both researchers and teachers which involve collecting samples of learner language, identifying errors, classifying them according to their nature and causes and evaluating their seriousness”. The following sections will shed more light on these four steps.

2.1.2.1 . Identification of errors

Ellis (2008: 47) argued that the concept of “error” is still vague and constitutes a subject of debate. As was shown in the previous section, researchers have not yet decided whether pragmatic failure can be considered an error. They have not reached an agreement about the distinction between errors and mistakes. The problem of overt and covert errors remains unresolved. But above all, researchers have not yet reached a consensus on considering “the infectious” use of a foreign language as erroneous. This last point is of particular importance to this study. Indeed, this idea may be behind calls for tolerance of errors for non- native speakers. This will be discussed in more depth in the results and discussion section. For a better understanding of the nature of errors, the next section will present a detailed description of errors through James (1998) and Ellis (2008) characterization.

2.1.2.2 Description of errors

Ellis (2008, p. 52) states that two descriptive taxonomies can be used in describing errors: surface strategy taxonomy and linguistic taxonomy. According to James (1998: 104-114), the type of errors has been classified into two types: linguistic category classification and the surface structure taxonomy.

• The Surface Structure Taxonomy

This is the second type of descriptive taxonomy proposed by Dulay, Burt and Krashen (1982:150). This taxonomy is based on “the way surface structures are altered”. James (1988) claims that there are four ways in which writers can modify target forms. They are respectively: Omission, Addition, Misformation and Misordering

• Linguistic Category Classification

This taxonomy classifies errors in the light of linguistic categories. First, it indicates the level at which the error is located: in phonology, grammar, lexis, text or discourse. Second, it is about the class of grammatical errors, i.e. noun, verb, adjective, adverb, preposition, or conjunction.

James (1998:129-160) classified errors into three levels: substance errors, text errors, and discourse errors. This study focuses on text errors only because some feature of errors are similar with the common ESL writing errors of Ferris (2005:53) model. In James’ (1998) classification, text errors are comprised of morphology errors, syntax errors and lexical errors, which are described as follows.

1. Morphology Errors

Morphology Errors are discussed in terms of grammar. (James: 1988). In English, morphology errors include noun, verb, adjective, adverb and preposition.

2. Lexical Errors

Lexical Errors are considered to be the most frequent category for some learners. Also, Native speakers consider the lexical errors in learners’ inter language to be more disruptive and irritating than other types (James, 1998: 143-144). Lexical errors can be divided into two main types: formal errors of lexis and semantic errors in lexis (James: 1998). Formal errors of lexis include misselection, misformation, and distortions, whereas, Semantic Errors in Lexis include Confusion of sense relations and Collocation errors.

3. Syntax Errors

Syntax Errors are related to phrases, clauses, sentences and paragraphs. They include Phrase Structure Errors, clause errors and Sentence Errors.

Ferris et al. (2005:53) propose a classification of the common ESL writing errors into five main types: morphological errors, lexical errors, syntactic errors, mechanical errors and miscellaneous errors. The following table illustrates the common ESL errors illustrated by Ferris (2000).

Morphological Errors ^ global / local errors

Verbs: Tense, From, Subject-verb agreement

Nouns: Articles/determiners, Noun endings (plural/possessive)

Lexical Errors ^ local errors

Word choice, Word form, Informal usage, Idiom error, Pronoun error

Syntactic Errors ^ global errors

Sentence structure, Run-ons, Fragments

Mechanical ^ local errors

Punctuation, Spelling, Capitalization

Figure 2.1 Common ESL Writing Errors based on Ferris’ (2005) Model

2.1.2.3. Explanation of errors

As Ellis (2008: 53) claims that the explanation of errors is an EA stage is related to the source of the error. Ellis (2008) suggests that the error sources can be psycholinguistic sources that concern the nature of L2 knowledge system and learner difficulties in target language production; sociolinguistic sources that explain the learner’s ability to adjust his/her language in accordance with the social context; epistemic sources that focus on the learners’ lack of world knowledge, and discourse sources that involve problems in the organization of information into a coherent ‘text’.

There are three explanations for the learner errors. They are:

• Firstly, errors could be caused by interference of the mother tongue. Language transfer from L1 causes interlingual errors. Saville-Troike (2005: 39) confirmed that the most likely causes of L2 errors were interlingual factors that resulted from negative transfer or interference from L1.
• Secondly, errors could be attributed to the learning process itself such the faulty generalization, and failure to learn conditions under which the rules apply. These are called Intralingual errors. They neither reflect the features of the L1, nor are they attributable to cross-linguistic influence (Saville-Troike, 2005). Corder (1975) and Saville-Troike (2005) argued that intralingual errors can be considered as developmental errors.
• Thirdly, faulty teaching techniques or materials. Corder (1975) states that when learner errors do not fall into inter-lingual or intra-lingual categories they can be assigned to this category of explanation.

Richards & Schmidt (2002:267) state that errors can be explained as being interlingual or intralingual. Interlingual errors can be grammatical, lexical or pragmatic errors which result from the learner’s first language features. On the other hand, intralingual errors are overgeneralizations in the target language.

Another taxonomy of sources of errors could be found in Keshavarz (2015). His taxonomy focuses on 1) interlingual 2) intralingual and developmental errors. Keshavarz hypothesized that “interlingual errors result from the transfer of phonological, morphological, grammatical, lexical, and sociocultural elements of the learner’s mother tongue to the learning of the target language” (Keshavarz 2015:111). On the other hand, he defined Intralingual and developmental errors as those which ‘are caused by the mutual interference of items in the target language, i.e., the influence of one TL item upon another’ (Keshavarz 2015:115). He classifies them into six subcategories: Overgeneralization Ignorance of rule restriction, false analogy, hyperextension, hypercorrection and faulty categorization.

2.4.2.4. Evaluation of error seriousness

The final stage of EA is concerned with error evaluation. In so doing, Burt (1975) distinguished global errors that affect overall sentence organization and local errors that affect single elements in a sentence (i.e. morphology or grammatical functors).

Ellis (2008: 53) raised many questions related to the importance of errors. The first issue was about the difficulty of errors. The second was about differences of NS’s errors and NNS’s errors. The third issue was about the criteria used to evaluate errors.

As a consequence of this analysis, Ellis (2008: 60) argued that errors per se did not have any hierarchy. Nevertheless, because teachers judged some errors more important than others and teaching became dependant on that classification of errors, they played a pedagogical role and gained some prioritization.

To justify this idea, Ellis (2008) reported that NSs considered lexical errors more serious than grammatical errors, whereas NNSs tended to be especially hard on morphological errors.

In agreement with the problematic nature of error evaluation (Ellis, 2008, p. 53) suggests three criteria in assessing error gravity. They are:

• Intelligibility; concerned with the extent to which sentences containing different kinds of error can be comprehended
• Acceptability; indicating the seriousness of the error
• Irritation; concerned with the emotional response of an addressee, and is related to the frequency of error occurrence.

In this study, the focus will be on the two last criteria, i.e., acceptability and irritation.

2.1.3. Benefits of error analysis

Corder (1967: 167) strongly emphasized error analysis benefits: “Errors tell the teacher how far the learner has progressed and what is left to learn; they tell the researcher how languages are learned; they provide the learner with a way of testing hypotheses about the language being learned”. Later, he reiterated this idea, explaining that errors made by second language learners are “normal and inevitable features of learning and inform the teacher about the learning process” (Corder, 1974:125). This view was supported by Stark (2001:19), who argued that students' errors should be seen positively and should be regarded as part and parcel of the learning process.

He added that Remedial exercises could draw more attention on the trouble spots. In fact, the learner determines input. What the teacher can present is the available input that should be learned (Corder, 1974:125).

Richards and Sampson, (1974:15) assert that “At the level of classroom experience, error analysis will continue to provide one means by which the teacher assesses learning and teaching and determines priorities for future effort.” In line with Richards and Sampson (1974), Svartvik (1973:13) has distinguished six ways in which EA might improve the language teaching situation:

1) To set up a hierarchy of difficulties

2) To set out the teaching priorities at different levels

3) To make the principles of grading more objective

4) To improve teaching materials

5) To revise syllabi in a non-ad hoc manner

6) To design tests according to the different levels and objectives

Other studies confirmed Corder’s observations. Kwok (1998:12) asserted that language errors provided insightful information about the language system, of the learner. Nation and Newton (2001:140-141) agreed that correcting errors was done to determine why the error occurred. Ultimately, EA served to reveal the processes behind their occurrence.

Kroll and Schafer'(1978:242-48) and Kwok (1998:11-22) proved how error analysis can be useful in improving the writing skill. They analyzed possible sources of errors in NNE writers’ work, and tried to provide a process approach to writing.

Olasehinde (2002) cited that errors are unavoidable and a necessary part of the “learning curve”. This view is supported by Stark (2001:19), who also explained that the teachers should not regard students' errors as a sign of failure but they need to view them positively as a process of learning. Vahdatinejad, (2008) maintains that error analyses can be used to determine what a learner still needs to be taught.

A number of studies have been devoted to written text analysis several authors such as Myles (2002), Chu (2005), Chen (2006), and Fang and Xue-Mei (2007) have undertaken studies on written text analysis. Their main common focus was on second language students’ written errors. They all agreed that their main objective was to enhance teaching and learning strategies.

This study totally agrees with this conception of EA. However, this study slightly differed with the quoted works since it will focus on the teachers’ reaction to errors in order to attempt an evaluation of their conception of Tunisian students’ errors.

2.1.4. Criticism of Error Analysis

James (1988) asserted that “it is not deemed legitimate ... to compare the child’s or the FL learner's idiosyncratic dialect (ID) to the dialect of adults or of native speakers respectively” (James, 1998: 16). The reason is that “the child or the FL learner are neither deliberately nor pathologically deviant in their language, so it would be wrong to refer to their repertoires as erroneous” James (1998: 16).

James (1998: 18) criticized EA as it does not take into consideration the strategy of avoidance, i.e. that learners tend to avoid certain language items which they are not sure about, and so they do not make errors in the areas where they would be expected to make them. More criticism comes from Dulay et al. (1982: 141-143) who point to the fact that EA confuses explanatory and descriptive aspects ,i.e, the process and the product.. However, despite all the criticisms, EA remains the most effective approach to L2 learners’ errors, and its advantages outweigh its drawbacks.

2.2. Error Treatment in writing

Although treating errors in writing has drawn the attention of many linguists, they differed in labelling the concept. For example, Hendrickson (1984) and Hammerly (1991) used ‘error correction’, whereas Cohen and Cavalcanti (1990) used the term ‘feedback’. According to Ellis (1994), the terms, ‘feedback’, ‘repair’ and ‘correction’ were generally used to refer to the same concept: ‘error treatment’.

2.2.1. Definition of Error Treatment

Researchers seemed to differ in the conception of the term ‘feedback’. Some have considered it an extensive scope encompassing different types of classroom interactions (Chaudron, 1988; Lalande, 1982). Indeed, Dulay et al. (1982) clearly stated that ‘feedback’ is the listener or reader’s responses provided to the learner’s spoken or written production. Similarly, Wajnryb (1992) expanded the meaning of ‘feedback’ to encompass the teacher’s responses given to any learners’ production in the classroom. Hence, they considered it as any kind of procedure used to inform whether a learner’s response was correct or wrong. However, some others, such as Keh (1990: 294) reduced the meaning of ‘feedback’ to “input from a reader to a writer with the effect to providing information to writer for revision”.

A third group of researchers, namely Nunan (1991) and Ayoun (2001), argued that teacher’s responses can be either negative feedback or positive feedback. Feedback for them was simply “information following an error produced by the language learner” (Ayoun, 2001: 226).

In the beginning researchers seemed to confuse the terms ‘correction’ and treatment of learners’ errors. Indeed, Hendrickson’s (1984) used the two terms interchangeably. However, later, Chaudron (1986: 66) defined the concept of correction as “any reaction by the teacher which transforms, disapprovingly refers to, or demands improvement of a students’ behavior or utterance”. This notion of disapproval was echoed by Ellis (1994) who asserted that ‘correction’ is defined as teachers’ attempts to provide negative evidence to deal specifically with learners’ linguistic errors.

For this reason, authors like Allwright and Bailey (1991) objected to the use of the term ‘correction’. In addition, for them the term correction implied “a permanent cure”. Nevertheless, when treating a learner’ error, the teacher is far from claiming to eradicate them permanently.

The ultimate objective of any error treatment or correction should be the establishment of beneficial pedagogical practices in classrooms that would advance the learning process. Authors such as Fregeau (1999), Koshik (2002) and Lewis (2002) recommended the establishment of a specific lesson. Harmer (1991) advised to use an error illustration, in which the teacher uses learners’ common errors for class explanations. In line with Ziv (1984), Moxley (1989) and Hyland (1990) recommended a direct method for the treatment of learners’ writing errors. The main stages of this method were the teacher’s tape-recording of their comments on each learner’s written errors on a cassette tape. Then, the teacher should end by writing feedback. These authors believed that the direct method would have a positive effect on the learners’ performance.

Many studies focus on the effectiveness of different types of teachers’ written responses to learner errors (Leki, 1990). Since the current study is meant to investigate and analyze teachers’ reactions to the gravity of written errors, it is necessary to restrict the meaning of the term. The next section will concentrate on the literature concerned with the impact of error treatment on second language learning and teaching.

2.2.2. Overview of Feedback in Learning Theory and Instructional Design

Error correction has been examined under different labels: as corrective feedback by second language teachers (Fanselow, 1977), as negative evidence by linguists (White, 1989), as negative feedback by psychologists (Annett, 1969), as repair by discourse analysts (Kasper, 1985) and as focus- on- form in more recent work by second language acquisition (SLA) researchers (Long, Inagaki & Oretega, 1998; Ammar & Spada 2006, and Ammar 2008).

Researchers have suggested that corrective feedback (CF) is directly related to L2 learning, because it helps learners to notice L2 forms (Bitchener & Knoch, 2010; Loewen & Erlam, 2006; Lyster &Mori, 2006; Varnosfadrani & Basturkmen, 2009). This noticing potential of CF paves the way to differentiate between “input-providing (CF) and output-pushing CF” (Ellis, 2006). For example, input-providing (CF) provides the correct reformulation through recasts; while, output-pushing CF promotes students’ self-repair through prompt (Lyster, 2002, 2007; Lyster &Mori, 2006, 2008; Ranta & Lyster, 2007).

Feedback as an instructional design model, helps in fostering cognitive learning (Reigeluth,1999). By the same token, Merrill (1994: 150) confirms the typical position of feedback in his Component Display Theory. In which he highlights the role of feedback in the linguistic forms which are “information added to the Primary Presentation Forms to enhance the learning that occurs” in an instructional situation.

The practice of instructional design has been influenced by major learning theories such as behavioral learning theory, cognitive information processing theory, and Gagne’s theory of instruction.

In the behavioral learning theory, two concepts seem to be basic to introduce: reinforcement and feedback as in Skinner’s (1958) Teaching Machines. Therefore, these two concepts can affect students’ learning behavior which is observed before and after an instruction to major its efficiency. By the same token, other researchers such as Lockee et al. (2007: 192) sustain the idea of

“reinforcement through evaluation and feedback” in the instructional design dates back to Skinnerian theory.

Cognitive learning theory emphasizes language learning as a cognitive process (Chomsky, 1965). Atkinson and Shriffin (1968) proposed a multistage theory of memory which categorizes human memory into three memory systems: sensory memory, short-term memory, and long-term memory. Building on this classification, learners process input from the environment through sensory memory and store it in short-term memory for a limited time. Through several processes including rehearsal, coding, and retrieval strategies, learner produces response output or store the information in long-term memory (Atkinson & Shriffin, 1968; 1971). Furthermore, Driscoll (2007) claim that feedback has a double function. The first function has to do with providing learners with accurate output, whereas the second function enables the learner to approximate the target linguistic system. As a matter of fact, learners get many information from feedback that try to store them in both short-term and long-term memory.

Feedback is part and parcel of Gagné et al (1992) systematic instructional design model. It is known as the “Events of Instruction”. This model includes nine steps: 1) gaining attention,2) informing learner of the objective, 3)stimulating recall of prerequisite learning,4) presenting the stimulus material, 5)providing learning guidance, 6)eliciting performance, 7) providing feedback, 8) assessing performance, 9)and enhancing retention and transfer. All in all this model is meant to show that feedback is not a nonsensical instructional moment because it evaluates the correctness of the learners’ performance.

2.2.3. Feedback in the Teaching of ESL/EFL Writing

Focusing on the impact of feedback in English as a Second Language (ESL) or English as a Foreign Language (EFL) represents a hot arena of debate in the teaching of writing (Paltridge, 2004; Reichelt, 1999). Cardelle and Corno (1981) claim that feedback on writing makes more effective learning because students become aware of their errors as they receive more feedback. In line with this thought, Carless (2006) and Brookhart (2003) confirm that students who receive feedback during the writing process have a clearer view of their actual performance and their future needs to improve. Furthermore, feedback may inform teachers about how well the students performance in a given task (Schwartz & White, 2000). Teachers are responsible for helping students develop their ability to reach their learning goals through teachers’ feedback.

Raising students’ awareness of the informational, rhetorical and linguistic expectations represents another role that feedback can play (Hedgcock & Lefkowitz, 1994). As Williams (2005) asserts that feedback in writing can stimulate explicit knowledge of student writers. This type of knowledge is about the language rules that students can apply. Through feedback students become able to check and balance their prior knowledge about writing rules that they have already learned. Furthermore, feedback represents an awareness-raising device which draws students’ attention to the writing aspects that need reinforcement, to improve their performance (Swain 1985: 290).

The investigation of the impact of feedback on students’ writing performance and accuracy have ranged from feedback source, function, focus, strategy, to feedback media. Studies on feedback source were conducted by Jacobs et al. (1998); Keh (1990); and Tsui and Ng (2000). Studies on feedback functions including informative and corrective feedback, were conducted by, Fazio, (2001); Ferris and Roberts (2001); and Hyland and Hyland (2001). Other studies were conducted on feedback strategy such as direct and indirect feedback; those conducted by Ashwell (2000); Bitchener (2008); Bitchener et al. (2005); Chandler (2003); Ferris and Roberts (2001). Studies on feedback media including written, oral, and electronic were conducted by Bitchener et al. (2005); Goldstein (2004); Buck (2008); Greenfield (2003); Honeycutt (2001); Liu and Sadler (2003); Matsumura and Hann (2004); and Tuzi (2004) among others.

These different forms of feedback mirror its importance in the teaching and learning of writing. However, there are clashing views about the validity of treating written productions. Through unveiling opponents’ and proponents’ viewpoints we can have a better understanding of error treatment in SLA.

2.2.4. Issues of Error Treatment in SLA

Second language teaching witnessed a shift from traditional methods to the communicative approach. This shift caused a drastic change in the attitudes towards learner errors. During the high days of audiolingualism, learners were expected to produce error-free utterances. Grammatical accuracy was recommanded; thus, errors were corrected immediately by the teacher after they occurred to avoid fossilization (Larsen-Freeman, 1986; Richards & Rodgers, 1986). This approach to language learning limits learners’ attention to only language forms (Major, 1988). However, by the advent of the communicative approach to language teaching was developed in the United Kingdom (Mings, 1993) to foster learners’ communicate competence. Unlike audio-lingualism, there is little error treatment in this method, because using the language communicatively is considered to be more important than using the language perfectly. Teachers are guided to help learners to enhance their communicative competence not only “the knowledge of grammar rules”, but also “sociolinguistic rules of use” and “paralinguistic phenomena such as body language” (Major, 1988:82).

Major (1988) points out that the trends in language teaching tend to go from one extreme to another and the attitudes toward error treatment have been swinging back and forth between these two extreme positions. It seems that final conclusions have not yet been reached, and researchers have approached the issue of error treatment in a variety of ways.

2.2.4.1. Negative Perspectives of Error Treatment

There are many debates concerning the effects of error treatment. However, despite the present controversy, many L2 teachers feel obligated to correct all errors to avoid the fossilization of learners’ errors (Kepner, 1991). Moreover, error treatment is a very demanding task because it requires time and energy. On the other hand, (Truscott, 1996: 239) states strong evidence against grammar correction in his article entitled ‘ The case against grammar correction in L2 writing classes ’. Truscott relates grammar correction with termed as “correction of grammatical errors for the purpose of improving a student’s ability to write accurately” is not only ineffective but also harmful. Consequently, Truscott’s argument has a great impact on theory and practice.

Truscott (1996) provides three issues of error treatment in the arena of SLA. They areas follows:

1) Teachers’ views of error treatment are stereotypical in nature. The correct knowledge from teachers to learners occurs unless the developmental systems of acquisition are clearly understood. Therefore, he explains the difficulty of an efficient error treatment by the Complexities of the interlanguage development system itself.

2) Truscott (1996) mentions that the second problem of grammar correction is related to the order of the acquired grammatical features in SLA. Therefore, some possible problems may arise when instructional sequences are inconsistent with the learning sequences. In the same vein, Pienemann (1985: 37) states that there should be a relationship between the natural acquisition sequence and the teaching sequence: “The teachability hypothesis predicts that instruction can only promote language acquisition if the interlanguage is close to the point when the structure to be taught is acquired in the natural setting”.

3) The final problem is related to the issue of the uncertainty of the interlanguage development processes. Thus, some types of teaching and learning practices may not be consistent with these processes.

Truscott (1996: 345) claims that when learners cannot perform the supposed acquired target language knowledge, the correction practice produces “pseudo­learning".

In addition to the theoretical problems mentioned above, Truscott (1996) insists that grammar correction has harmful effects because it demotivates learners. According to Krashen’s Monitor Theory (Krashen & Terrell, 1983:45), the focus on conscious learning may lead to an “undesirable result of encouraging over­use of the Monitor”. Therefore, the monitor of overusers will be in danger of becoming hesitant and their learning will be excessively slow (Stern, 1992).

Sheppard’s (1992) study demonstrated that learners who did not receive correction achieve more grammatical accuracy than those who received correction. Similary, Semke’s (1984) study with 141 university students who in German classes showed that the uncorrected learners perform better than the corrected ones. With this result, Semke (1984:195) concludes that “corrections do not increase writing accuracy, writing fluency, or general language proficiency and they may have a negative effect on student attitudes". So, correcting learner errors is both ineffective and counterproductive.

On the basis of both theoretical and empirical evidence, researchers’ approach error treatment as negative and critical. Truscott (1996) argues that it would be better for both language teachers and learners not to waste time on such ineffective and harmful corrective practice. Other tasks are more demanding for this time which can be spent more productively. Seemingly, Truscott’s argument seem to be valid to a certain extent; however, many researchers proved the necessity of error treatment in SLA.

2.2.4.2 Positive Perspectives of Error Treatment

It exists many theoretical assumptions that provide much research evidence to support the potential benefits of error treatment (Lyster et al., 1999). For example, Chaudron (1986: 82) made the following statement: “Despite the lack of evidence that feedback on the linguistic error in classrooms or outside them is consistently effective in stimulating learners' interlanguage progress, the possibility remains that certain learners, especially those with a formal learning style, can derive benefit from error correction”. Moreover, Hendrickson (1984:145) argues that “for those adults, whom Krashen calls ‘monitor-users’, error correction helps to discover the functions and limitations of the grammatical structures and lexical forms of the language they are studying”. In addition, some evidence from classroom studies prove the efficiency of error treatment. For instance, the results of (Carroll and Swain's, 1993: 358) study is in line with schachter’s claim that different types of feedback are helpful for L2 learners. They claim that negative feedback lets the learners "narrow the range of possible hypotheses that can account for the data"(Carroll and Swain, 1993: 358)

Moreover, Nunan and Lamb (1996:68) assert that making errors and subsequent teacher corrections “can provide the learners with valuable information in the target language”. In fact, Tomasello and Herron’s (1989) study found that learners who were given language rules in advance fail to achieve accuracy, whereas learners who were first allowed to make mistakes and were then corrected become better language achievers.

Some other studies also reveal the significance of negative feedback for L2 learners. White (1989,1991) and Rutherford (1987) argue that corrective feedback is needed when the learners cannot notice the differences between their interlanguage and the target language. The findings from studies by Cardelle and Corno (1981), Lalande (1982), and Robb et al. (1986) suggest that systematic and salient correction on grammatical errors is effective to improve the writing skills. Cardelle and Corno (1981:260) hypothesize that “specific feedback on errors draws attention to material not adequately learned, allowing the students to focus there and not be distracted by too much re-examination of work done well”.

Chenoweth et al. (1983), studied ESL learners’ reactions to the correction of their errors. He asserts that understanding the role of error treatment is related to the knowledge of learners’ attitudes and preferences. Leki’s (1991) survey of 100 ESL learners’ preferences for error correction shows that the learners’ main concern is the production of perfect writings. Leki (1991:206) notes that “many students had said that perfect grammar, spelling, vocabulary choice and punctuation were important”. More specifically, Ferris (1999) asserts that the lack of correction may demotivate learners; thus, the error correction is needed to enhance their motivation. Bartram and Walton (1991) point out that several problems will arise if teachers do not correct errors: teachers will feel guilty; and will be thought of as lazy, lacking responsibility or being incapable.

Consequently, there is a consensus among SLA researchers that error correction is necessary for adult language learners in sheltered settings. In order to acquire the target language structures effectively, the learners have to know whether what they are doing is right or wrong (Larson, 1985). Therefore, error treatment is vital.

The above-mentioned theoretical and empirical evidence showed that error correction should be continued. However, it is seen as an exhausting teaching task because teachers find themselves obliged to identify all kinds of errors. According to Moxley checking every single error is not a very productive activity for both teachers and learners. For example, Hairston (1986) recommends that teachers should set the priorities about learner errors. In fact, prioritizing errors and correcting them selectively has been recommended by many researchers. This field of investigation is labeled Error Gravity research. It is to this area that I turn next.

2.3. Error Gravity Research

Teachers have to choose whether to "treat" or "ignore" students' mistakes. Error gravity (EG) Teachers' manuals don't offer much guidance "on which err ors teachers should rectify and which ones they should ignore," according to E llis (2013:4).

• Which error to prioritize on the basis of which criterion?

• How to correct the errors?

• Who corrects the errors?

2.3.1. Which error to prioritize?

With the emergence of error evaluation research, the emphasis on errors has transitioned from simply describing and explaining them to assessing their significance and particularly determining which specific errors should be prioritized. Consequently, Lee (2003: 164) appears to recognize this issue by posing the following inquiries: “How can educators assess the severity of errors? Which errors should be corrected by teachers, and which ones can be ignored? These are crucial questions to investigate.”The efficiency of selective correction for textual errors appears to be well accepted in literature.

For instance, Hammerly (1991) asserts that providing error therapy in a meth odical and targeted manner increases its efficacy. Additionally, Celce- Murcia (1985) claims that one of the best teaching techniques is selective correction.

CelceMurcia's comparison of more and less successful teacher corrective techniques is shown in the following table.

Table 2.1 Celce-Murcia’s (1985) table of More Effective and Less Effective Teacher Correction Strategies

Illustrations are not included in the reading sample

The variability in evaluators’ assessments of the severity of L2 errors encourages many researchers and educators to consider the standards that underpin these judgments. James (1977: 116) appears to align with this perspective by stating that ESL instructors likely ‘do consistently refer to criteria of degrees of erroneousness when they assess, even though these criteria may not be explicitly articulated’. For instance, Hyland (2006) used violations of rules as his primary standard for evaluating the seriousness of errors. Similarly, Ferris (1999:6) indicated that written corrective feedback should focus on “treatable errors” (that is, errors related to elements that manifest in a “patterned, rule-governed way”).

Some researchers, on the other hand, use various criteria for addressing errors. Bartram and Walton (1991) argue that some types of errors hold greater significance than others. As a result, it is essential for educators to understand the rankings of these errors. While various criteria exist, the most significant errors frequently emphasized in the literature of English grammar by researchers and educators are associated with the following criteria:

(1) those that are relevant to the pedagogical focus, (2) those that occur frequently , (3) those that hinder intelligibility (4) those that affect teachers’ acceptability (5) those that cause irritation and (6) those that affect comprehensibility. On the basis of these criteria, the following section will examine what types of learner errors should be treated.

2.3.1.1 . Errors that are relevant to the pedagogical focus

It has been noted that the seriousness of learner errors and the kind of correction strategy used to deal with those errors depends on the objectives of a lesson (Nunan & Lamb, 1996; Wen, 1999). In Cohen’s (1975) article, he asserts that errors related to a specific pedagogic focus deserve higher attention than other less important errors. By the same token, Walz (1982) maintains that the features of the target language that have been recently taught in a class should be one of the criteria to decide which errors to correct. Hammerly (1991) claims that errors that learners make with what they have been taught are basically different from errors they make with what they have not yet been taught. Hence, (Hammerly, 1991: 31) terms these two dimensions of learner errors ‘ distortion ’ and ‘ fault ’ respectively and further classifies them into four types based on who contributes to the error.

Hammerly (1991:90) states that faults occur whenever the learners attempt the target language structure which are beyond what they have learned. He adds that “there is not much point in correcting faults, as there is no reason why the students should be able to correctly use structures they haven’t studied”. In this respect, his perception of correction is similar to Truscott’s (1996) argument that error correction has little value unless it is consistent with the developmental sequence of second language acquisition. So, the focus of correction must be on “what is being taught or has been taught” (Hammerly, 1991:90). In short, the teacher who chooses the pedagogical focus as the criteria of error correction, has to modify the preference of error correction not only on the objectives of a particular lesson but also on what individual learners have learnt in the target language.

2.3.1.2 Errors that occur frequently

Dresdner (1973) highlights the importance of prioritizing errors that happen frequently in correction. As a matter of fact, high-frequency errors, have been suggested for SLA evaluators when they respond to written output (Allwright, 1975, Mings, 1993). (Hendrickson,1980:161) hypothesized that it is necessary to examine the frequently committed errors at various stages of SLA, because it could bring insights to build “hierarchies of language learning features”. High frequency errors are considered by Dresdner (1973), and Allwright (1975) to deserve typical priority attention in error correction. Therefore, 'High frequency' indicates the repitition of the same error on the part of an individual student.

Some researchers have illustrated errors that are frequently produced by ESL learners. The following shows the most common errors made by ESL learners introduced by Dulay et al. (1982).

(1) Omitting grammatical morphemes, which are items that do not contribute much to the meaning of sentences, as in He hit car.

(2) Double marking, a semantic feature (e.g. past tense) when only one marker is required, as in She didn’t went back.

(3) Regularizing rules, as in womans for women.

(4) Using archiforms - one form in place of several - such as the use of her for both she and her, as in I see her yesterday. Her dance with my brother.

(5) Using two or more forms in random alternation even though the language requires the use of each only under certain conditions, as in the random use of he and she regardless of the gender of the person of interest.

(6) Misordering items in constructions that require a reversal of word-order rules that had been previously acquired, as in What you are doing?, or misplacing items that may be correctly placed in more than one place in the sentence, as in They are all the time late. (Dulay et al., 1982: 138-39)

Vann et al.’s (1984) study examined faculty’s opinion of ESL errors. Teachers agreed upon the following as common ESL writing errors:

• Spelling Subject-verb
• Article Relative clauses
• Comma splice Tense
• Prepositions It-deletion
• Pronoun agreement Word order

(Vann et al., 1984: 431)

In addition, Ferris and Roberts’s (2001) article introduces the following five categories that represent the five most frequent errors.

Table 2.2

Illustrations are not included in the reading sample

2.3.1.3 Errors that hinder intelligibility

In the arena of SLA, intelligibility has been defined as “the impact of errors upon the interlocutor’s ability to understand the meaning of the utterance” (Khalil, 1985: 344). Burt and Kisparsky (1972) build their distinction between global and local errors on the light of intelligibility or communicability. Accordingly, global errors may have a more serious impact on communication than local errors. Hicks (1983) claims that the distinction between global and local errors should be extended beyond the boundaries of the sentence to include errors in cohesion and coherence.

In general, intelligibility is related to the extent to which an utterance has been understood. Thus, correctness is required to guarantee intelligibility. Therefore, incorrectness could pose problems with different degrees. Errors that affect intelligibility need priority attention otherwise communication breakdown occurs.

Many researchers have agreed with the idea that the errors which hinder communication are considered to be the most important to correct. Hammerly (1991) suggests that teachers should only correct those errors that significantly affect intelligibility and ignore those that do not. With regard to this point, Burt and Kiparsky (1972) draw a clear-cut distinction between ‘ global ’ and ‘ local ’ errors based on the communicative importance of erros. In their study, global errors are errors that seriously obstruct communication and cause interlocutors to misunderstand a message. On the other hand, local errors are termed as errors that are isolated sentence elements, such as noun and verb inflections that make a structure in a sentence awkward, yet, do not hinder the comprehension of the message (Croft, 1980; Hammerly, 1991; Hendrickson, 1980; Lewis, 2002; Raimes, 1991; Walz, 1982). Dulay et al. (1982) introduced the following types of grammatical deviations as global errors in their study:

- Wrong order of major constituents
- Missing, wrong, or misplaced sentence connectors
- Missing cues to signal obligatory exceptions to pervasive syntactic rules
- Regularization of pervasive syntactic rules to exceptions
- Psychological predicate constructions
- Selectional restriction on certain types of verbs in sentential compliments (that-clauses, infinitive and gerunds)

(Dulay et al., 1982:191-7)

According to Dulay et al., in order to communicate with others successfully, learners must learn global aspects of grammar. Therefore, global errors must be prioritized over local errors. Olsson’s (1972) study reveals that semantic errors generally impede communication more than syntactic errors. Moreover, Politzer’s (1978) study of errors made by English speakers of German as well as Delidle’s (1982) study of written errors with native speakers of German found that vocabulary errors were considered to be the most serious errors (Chenoweth et al., 1983). Interestingly, the findings from the survey carried out by Medgyes and Reves (1994) show that vocabulary was chosen to be the most frequently perceived difficulty among non-native speakers of English. The possible reason for this is due to a wrong choice of words seriously hindering comprehension and causing miscommunication with other people.

Hendrickson (1977) suggested an error chart in order to record learners’ errors for diagnostic purposes. Hendrickson (1980) asserts that these types of charts are useful not only for developing teaching materials but also deciding a hierarchy of error treatment priorities. Furthermore, this error chart helps teachers to know more about the process of SLA.

Table 2.3

Error Chart (Hendrickson, 1977, extracted from Hendrickson, 1980, p.164)

Illustrations are not included in the reading sample

Apart from linguistic error categories, it is essential to raise students’ awareness about the socio-pragmatic differences between the learners’ native language and the target language since such differences may cause communication problems. Therefore, pragmatic errors should enjoy priority in corrective feedback because they affect the intelligibility of the utterance .Thus; Katayama’s (2007) study indicated that participants preferred their pragmatic errors to be corrected on all occasions, in contrast to linguistic errors such as grammatical, phonological, and lexical errors.

2.3.1.4. Errors that affect teachers’ acceptability

Bachman (1990) created a model of communicative competence that is fundamental to our understanding of acceptability. In reality, establishing certain norms to define acceptability is a very sensitive undertaking. Consequently, Corder (1973:101) advocates for “a sufficient degree of determinacy to identify some norms and make some statements about what is acceptable and what is not.” Indeed, without the validity of acceptability, it would be impossible to engage in meaningful language teaching.

2.3.1.5. Errors that cause irritation

Ludwig (1982:275) characterizes irritation as “a function of the speaker/writer’s erroneous employment of language evaluated against the characteristics and anticipations of the interlocutor.” Santos (1988) denotes irritation as the “bother” factor. In other terms, the reader experiences irritation in response to the utterance when he perceives frustration stemming from the produced utterance. Khalil (1985:336) articulates irritation in his own manner as “the element that prompts individuals to halt and pay attention to a textual feature.” Furthermore, Hendriksen and Faerch (1980: 394) propose that irritation, or the

emotional reaction to error, “is directly predictable from the quantity of errors present in an IL text, irrespective of error type or other linguistic dimensions of the text.” Irritation serves as a criterion in investigations conducted by Piazza (1979), Chastain (1980 a), Galloway (1980), Magnan (1981), and Ensz (1982).

2.3.1.6. Errors that affect comprehensibility

Comprehensibility is used as criterion in several studies such as by Burt and Kiparsky (1972, 1974), Piazza (1979), Olsson (1972), Guntermann (1978), Tomiyama (1980), and Chastain (1980). Piazza (1980:422) states that error gravity research determines errors that « interfere with comprehensibility and (which) may irritate native speakers ». Burt (1975) concludes that word order error was regarded as very serious to NS/NNS comprehensibility in English. However, in other studies (Chastain,1980 ; Johansson1978 ;Olsson1977 ; Politzer1978), lexical errors are found to be more serious than word order errors in impairing comprehensibility. For example, Chastain (1980 a: 212) concludes that "comprehension is most severely limited by word usage, the use of wrong word or the addition or omission of words". As far as comprehensibility is concerned, two studies (Albrechtsen, Henriksen, and Faerch, 1980; Guntermann ,1978)found that the number of errors affects comprehensibility more than the errors’ type. In examining the French tolerance of grammatical errors committed by Americans, Piazza (1980) concludes that "incorrect word order is relatively not irritating but can be a problem for comprehension" (p.424).

In many studies, irritation has been related to the lack of comprehensibility. But some errors could be irritating and comprehensible. However, Santos (1988) claims that teachers consider double negatives the most irritating errors, but the utterance is still comprehensible. Consequently, there is no positive correlation between comprehensibility and irritation.

To sum up, the study of the different possible criteria upon which we can judge the gravity of certain errors paves the way to establish some continua or scales to prioritize learning and teaching. For example, in her taxonomical study of common lexico-grammatical ESL errors made by Hong Kong Cantonese speakers, (Chan, 2010: 314) claims that "remedial efforts should, of course, be put on grievous errors, but prevalent errors, such as word class confusion, should also receive attention "and argues her taxonomy could be used to create "an error gravity scale and an error prevalence scale" ranging from "the most grievous or prevalent to the least grievous or prevalent". As a matter of fact, L2 instructors could use these scales to "sequence and prioritize their teaching focus according to the prevalence and gravity scales" (Chan, 2010: 315). In addition, Sheory’s research ends up with the following error gravity hierarchy (Sheory, 1986:310)

Deciding upon the seriousness of errors’ categories represents an unsettled matter in the context of EG. For example, some analysts such as Sheorey (1986), and McCretton & Rider (1993) found that according to NN teachers, lexical errors are least serious errors. In Hughes & Lascaratou’s study (1982) NSs marked semantic errors as more serious than other types. But in McCretton & Rider’s (1993) study NS teachers marked semantic errors as least serious in comparison to other categories. Khalil (1985) was of the view that semantically aberrant sentences were judged as less intelligible and hence, more serious than the syntactic ones. Thus, according to Khalil (1985), semantic errors are more serious, graver and irritating. These differences among linguists in deciding upon the most irritating error category represent an evidence that this point in literature deserves more research in different contexts and with different error categories to come up with solid conclusions.

2.3.2 How to correct errors?

Dealing with strategies of error correction is of a paramount importance. More specifically, treating written errors requires a certain specific corrective feedback. Drawing on the existing literature on written CF (Chandler, 2003; Ferris, 2002) as well as teacher handbooks written by methodologists (Hedge, 2000; Ur, 1996). Therefore, “focusing on patterns of error, allowing teachers and students to attend to, say, two or three major error types at a time, rather than dozens of disparate errors” (Ferris, 2002: 50). Ellis (2009: 98-103), comes up with a comprehensible typology of written CF. What follows is Ellis’s typology with some modifications illustrated by Keshavarz (2015).

2.3.2.1. Direct VS. Indirect Corrective Feedback

In direct corrective feedback, the teacher provides the correct form, highlights the wrong form or adds a missing word. Ellis (2011:40) claims that explicit instruction helps in “recruiting consciousness to overcome the implicit routines that are non-optimal for L2”.On the other hand, indirect corrective feedback is implicit in nature. Teachers do not provide direct correction, but they indicate the presence of an error. This is done in two ways, either through indicating and locating the error or indication only. Both observational studies (Ellis, Basturkmen, & Loewen, 2001; Loewen, 2004; Lyster, 1998; Oskoz & Liskin- Gasparro, 2001) and experimental studies (Ellis, Loewen, & Erlam, 2006; Nassaji, 2007) have shown the advantage of more explicit feedback over the more implicit feedback, due to the salience of their corrective force which promotes learner noticing. It seems that in adult SLA, teachers and researchers both agree that giving explicit explanations and corrections is more facilitative of learning (Nassaji, 2007, 2009).

2.3.2.2. Focused Vs. Unfocused Corrective Feedback

Whether the teacher corrects all students’ errors or selects some of them, creates the so- called focused- unfocused dichotomy. It should be noted that unlike the focused CF which is intensive, the unfocused CF is extensive and constitutes a relatively unsystematic way of correcting errors. So, focused CF seems to be of greater benefits because it enables learners to notice errors and correct them efficiently. Therefore, recent studies (Bitchener, Young, & Cameron, 2005; Sheen 2007; Ellis et al., 2008) have shown that when written CF is “focused” it is effective in promoting acquisition. As Ferris (1999:7) argues, “it is vitally important for teachers to commit themselves to selective error feedback and to a strategy for building students’ awareness and knowledge of their most serious and frequent grammar problems”.

2.3.2.3. Reformulation

Reformulation is defined as a native speaker's reformulation of student's text ‘to preserve as many of the writer's ideas as possible while expressing them in his/her own words so as to make the piece sound native-like'(Cohen 1989: 4). Ellis's mentioning of the native speaker is problematic for two reasons. First, in foreign language contexts, we do not have access to native speakers. Second, a native speaker is not that perfect to correct the work of non- native writers. Therefore, Jenkins (2007: 129) disregards the notion of native speakerism to come up with an alternative approach that highlights the role of "the expert speaker of English". Keshavarz (2015:144) proposed the use of "experienced and proficient English teacher, instead” of "expert". In fact, resorting to this dichotomy between native and nonnative evaluators paves the way to answer the following question.

2.3.3 Who should correct learners’ errors?

Error gravity research focuses on the existing discrepancies between different groups of language users, focusing mainly on those between (a) native and non­native speakers, (b) native- and nonnative-speaking evaluators, and (c) teaching and non-teaching judges (Johansson, 1978: McCretton & Rider, 1993; Roberts & Cimasko, 2008). In addition, the linguistic background and the teaching experience represent two important factors in shaping judges’ variation. Vann, Meyer and Lorenz' (1984) study on error gravity revealed that teachers’ reactions varies according to age, academic discipline and experience.

1. Native versus non-native speaker

2) Experienced versus novice teachers

2.3.3.1 Native versus non-native speaker

Rifkin (1995) claims that priorities in determining error gravity vary according to cultural norms and native speaker expectations. Johansson conducted a number of experiments which focus on error judgment of native and nonnative­speaking teachers of English. The findings show that NNSTs are more tolerant than NSTs in evaluating the gravity of the learners written errors. However, when it comes to orally produced errors, NST become more flexible. (1978: 121­23).

In Vann, Meyer, and Lorenz's (1984) study 164 participants were asked to rank the relative gravity of 12 typical ESL written errors. The results show that errors were not judged equally grievous. Therefore, they came up with a hierarchical order of errors and the results showed that "those respondents who were most consistent likely to be less tolerant in their judgment" (p. 433).By the same token, Kresovich's (1988) study addresses 43 teachers of ESL (one British, sixteen American, and twenty-six Japanese) who were asked to judge the acceptability of certain errors in Japenese students' productions. As a matter of fact, the results of this study did not present any differences between native and non-native speaking teachers in their error perceptions. By the same token Khalil (1985), who conducted a study by Arab EFL learners, concluded that semantically deviant forms are judged by native speakers of English as less intelligible and less accurate than are grammatically deviant forms.

Therefore, the cultural background of the evaluator plays a predominant role in shaping the differences between English instructors in evaluating the gravity of written errors. In addition to the cultural background, teaching experience seems to have a great impact on teachers’ variations in error seriousness evaluation. It is to this point that we will turn next.

2.3.3.2 Experienced versus novice teachers

The concept of "experienced" can be described as a common-sense idea, referring to people who have a special control over a language, insider knowledge about 'their' language (Davies, 1991:1). Many research findings have shown significant differences between experienced and inexperienced teachers' reaction to error both quantitatively and qualitatively. Experienced and inexperienced teachers have differing perceptions of error gravity with respect to various error categories (Allwright, l998).

In his study, James (1977) compared the responses of seventeen experienced teachers and seventeen inexperienced teachers to fifty sentences demonstrating a range of syntactic and lexical errors. He concluded that the inexperienced teachers were more severe than their experienced counterparts. James concluded that the inexperienced teachers tended to fall into two groups; one, relatively tolerant of error in line with the experienced teachers, the other, intolerant. As far as category analysis of errors is concerned, James found broad similarities in the way groups ranked different categories of error, and noted that both groups rated lexical error low in significance. Sheorey (1986) maintains that experienced teacher sensitivity to errors should be a guiding principle for teachers. As Hughes & Lascaratou stress, there is a need for a range of studies based on clearly defined and consistent categories for the relationship between types of error and levels of intelligibility (Hughes & Lascaratou, 1982:180).

Another vexed question is why inexperienced teachers appear to be less tolerant of error. Hughes & Lascaratou (1982) attribute it to differential language proficiency between experienced teachers and inexperienced ones, experienced teacher tolerance being based on 'better knowledge of the language, particularly of the wide variety of acceptable structures' (Hughes & Lascaratou, 1982:180).

In a survey of over two hundred inexperienced teachers, Medgyes& Reves (1994) Comment that inexperienced teachers are usually preoccupied with accuracy, the formal features of English, the nuts and bolts of grammar, the printed word, and formal registers. Many lack fluency, have a limited insight into the intricacies of meaning, are often in doubt about appropriate language use, have poor listening and speaking skills, and are not familiar with colloquial English. It is only logical to deduce that they place the emphasis on those aspects of the language they have a better grasp of (Medgyes& Reves, 1994: 59) .They see this as explaining the results of error treatment studies where inexperienced teachers lay great stress on grammatical errors and priority accuracy over intelligibility. He suggests that while experienced teachers generally regard language as a means of achieving a communicative goal, inexperienced teachers, on the other hand, perceive language as a set of rules that should be learned. In fact, Sheorey (1986:311) himself ponders in his conclusion whether learners are 'short-changed' by experienced teacher tolerance of error, and failure to alert students to some of their language deficiencies.

2.3.4. Criticism of error gravity research

Like any other type of research, error gravity research has its criticism. In fact, this criticism is related to two sub-areas. The first is related to protocols and the second is linked to constract.

2.3.4.1. Problems in the Protocols

First, common criticisms of typical error analysis studies tend to rely on isolated sentences to represent authentic discourse (Albrechtsen et al, 1980; Rifkin & Roberts, 1995; Chastain, 1980; Ellis, 2008). Ellis (2008) asserts, “error evaluation is influenced by the context in which the errors occurred”. In fact, he criticizes error gravity studies’ results for being aloof from the communicative contexts despite the “appearance of rigor given by the use of descriptive and qualitative statistics” (p. 60). So, the problem with error gravity stems from its failure in focusing on communicative contexts. As Chastain (1980: 212) argues, “an isolated sentence often may not supply the sufficient information”. Thus, the same error may be evaluated very differently depending on who made it and when, where, and how it was made” (Chastain, 1980:60).

Second, researchers in error gravity studies frequently predetermine which errors will be considered. Santos (1988) notes this approach does not “allow the NS judges to decide for themselves which errors are most glaring” and that “selectively inserted errors give equal weight to each error type by representing them only once each, an unrealistic condition that ignores the frequent recurrence of certain error types and the relatively infrequent occurrence of others” (p. 74). Studies involving both limited discourse and pre-selected errors include Burt (1975), Janopoulos (1992), Tomiyana (1980), McCretton and Rider (1993), and Vann, Meyer, and Lorenz (1984).

2.3.4.2. Problems in the Construct

Error gravity study typically limits error evaluation to intelligibility, comprehensibility, irritation, and acceptability (Khalil, 1985). However, these concepts are impossible to separate entirely (Ludwig, 1982; Johansson, 1975; Santos, 1988).

In effect, researchers are not sure where the separation between comprehensibility and irritation or acceptability and irritation lies. Ludwig (1982: 275), for instance, concludes, “in general, higher comprehensibility implies lower irritation”. Piazza (1980) makes a similar claim when she asserts “the more comprehensible an error type, the less irritating it was” (Piazza, 1980: 424).

Additionally, the line between irritation and acceptability is also elusive. Gynan (1995:161) posits that an error stems from an attitude, or “a learned predisposition to respond in a consistently favorable or unfavorable way with regard to an object, entity, person, or state. He asserts that “attitude toward language may be measured by determining whether the listener feels good or bad about the language or whether the listener thinks the language is good or bad”, which indicates that defining an error includes both the evaluative and affective responses coexisting simultaneously. The gravity of an error is a matter of personal opinion. Vann, Meyer, and Lorenz (1984), for example, found that some teachers view all errors as equally serious—“an error is an error.” There is no accepted theory of grammatical complexity to help teachers decide which features are marked. Hence, this argument debunks all error gravity hierarchies because all errors are seen as equal and there is no obvious need to prioritize an error over another.

Another problem in error gravity is proposed by James (1977) who speculates that the majority of raters do not “explicitly formulate...criteria” when considering error seriousness (James, 1977: 116). In addition, through his study James notes that judging the seriousness of error types is subjective in nature. Williams (1981) questions the ferocity with which teachers perceive errors and he notes that “.it is all very puzzling: (There is) great variation in our definition of error, great variation in our emotional investment in defining and condemning error, great variation in the perceived seriousness of individual errors” (Williams, 1981: 155).So, the variation among teachers may cause fragmentation and weak decision making. Also, Rifkin and Roberts (1995: 512) came up with findings that debunk the conclusions of 28 studies in error gravity. These findings reveal “inconsistent findings” and striking contradictions in evaluation “which make it difficult to point confidently in any one direction and proclaim it the route for improving native/nonnative interaction”. Rifkin (1995) claims that priorities in determining error gravity vary according to cultural norms and native speaker expectations.

Albrechtsen et al (1980: 395) claim that “one should not expect to establish a hierarchy of errors with respect to irritation: All errors are equally irritating, provided they are in fact errors ,i.e., violations of a target language norm”. But, the notion of a universal “target language norm” seems to be problematic. Norms depend on culture and standards. Therefore, they vary from community to another. This makes it impossible to state conclusively whether a particular utterance will still be erroneous when detached from its native community (Hultfors, 1986).

The next part will be about the methodological section, which will present the research methodological tools and instruments.

CHAPTER 3

METHODOLOGY

3.0. Introduction

This chapter seeks to examine the methodology of this study in order to get a deep insight into judgmental differences in teachers' evaluation of students' written errors. The first section introduces the corpus of the study, consisting of the final exams of second-year major English university students; the second, third, and fourth sections describe the subjects and their profiles as well as the reasons behind selecting this particular population. The following point will be about the instruments and the procedures used in this research. The last point describes the methodological tools employed in this study.

3.1. Corpus of the study

The corpus of the study is made up of 50 exam papers of third-year business students from the Faculty of Economic Sciences and management. The exam topic was "How do you perceive the Tunisian business landscape and what are the possible solutions to ensure a better future?”. The corpus will be analyzed according to a modified version of Keshavarz (2012) model of error analysis. The rationale behind this corpus selection is because we need a large number of errors. Therefore, we have relied on the production of the third-year to get as lengthy writings as possible. The identification of errors was done by the researcher and then negotiated by another experienced teacher. The latter agreed on the researcher’s identification of errors according to James (1998:129-160) and Ferris (2005:53) Analysis Model. However, the researcher neglected about twelve errors to identify which have been identified by the teacher. Apart from this, there is no other remark to draw concerning the validation of error identification process.

3.2. Participants

30 university English teachers contributed in this experimental research. These participants belong to Faculty of Economics and Management of Sfax (FSEGS), Higher Institute of Arts And Crafts, Sfax (ISAMS) and Private Polytechnic Institute of Advanced Science of Sfax (IPSAS), and they are male and female teachers.

3.2.1. Profile of the participants

The participants under scrutiny are all native speakers of Tunisian Arabic. They are aged between 29 and 60.

Table 3.1

Profile of long- teaching- experience teachers

Illustrations are not included in the reading sample

Table 3.2

Profile of short- teaching-experience teachers

Illustrations are not included in the reading sample

3.2.2. Population selection

The rationale behind the choice of this very population was that the teachers are the researcher’s colleagues. It is also worth noting that the choice of long teaching experience and short teaching experience dichotomy stems from the literature of error gravity. Therefore, answering the question which error to tolerate most? necessitates different viewpoints from different teachers with different experiences. The choice of teaching experience criterion is as follows: Long-teaching-experience-teachers have at least fifteen years of experience or more. However, Short-teaching-experience-teachers have less than ten years of teaching experience.

3.3 Methodology

This section will introduce the type of the study as well as the different procedures and instruments used in order to collect data.

3.3.1 Type of the study

This study aims to investigate teachers’ reactions towards the gravity of written errors on the basis of acceptability and irritation criteria on the one hand, and teaching experience on the other hand. As far as the nature of the investigation is concerned, this study is classified as experimental. Indeed, the relationship between variables is framed within the so-called factorial design because we have two independent variables. In this study, teachers’ variability in perceiving error seriousness depends on two variables: (1) criteria selection and (2) teaching experience. Therefore, we have one dependent variable which is teachers’ tolerance towards students’ errors that depends on two independent variables which are criteria and experience.

3.3.2 Procedures

The data collection procedure lasted one month in the second semester of the year2015/ 2016. To start with, the data collection of this study is based on learner’s error samples because they represent the ground upon which teachers are going to set their judgments. So an Error analysis research is carried out following Keshavarz’s model (2012:168) which “involves collecting samples of learner language identifying errors, classifying them according to their nature and causes and evaluating their seriousness”. The choice of this model is not at random, thus, this model is one of the most recent models in error analysis field of research. Moreover, Keshavarz’s mentioning of the evaluation of error seriousness makes his model a good fit to my research requirements.

The researcher changed the model to fit the research objectives and to craft a tentative model that goes with EG research. As a result, the two first steps in Keshavarz’s model remain in situ. Error seriousness explanation, on the other hand, will follow error seriousness evaluation. The latter contains two phases: phase 1 (evaluation on the basis of acceptability criterion), and phase 2 (evaluation on the basis of irritation criterion). Teachers' judgments are situated in the third step of the research model.

This is Keshavarz’s (2012) original model:

1- Error identification 2- classification 3- explanation 4- error seriousness evaluation

Table 3.3

Error Analysis research procedures

Illustrations are not included in the reading sample

3.3.2.1 Instruments of data collection

Wilkinson and Birmingham (2003:3) claim that "research instruments are simply devices for obtaining information relevant to your research project, and there are many alternatives from which to choose". The instruments influence both the information gathered and its interpretation. Accordingly, there are two approaches to use in gathering information. The first approach has to do with secondary sources, or second-hand data. In this approach, the information is available and needs only to be extracted. For example, government or semi­government publications, earlier research, mass media, books, and periodicals. However, the second approach deals with the collection of information from primary sources which provide ‘first hand' information. For example, tests, observation, questionnaire, interview, etc.

In fact, the tools used in this piece of research are gathered from primary sources. As this research is experimental in nature, it resorts to three techniques. They consist in analyzing teachers’ reactions towards gauging written errors. Each method contributes differently in answering the research objectives:

1. To have information about the teachers’ personal background.

2. To find out about the teachers ‘understanding of learners’ errors.

3. To understand how teachers behave in a no-contact context with students’ errors.

4. To get information about teachers’ preferences in tolerating or penalizing errors.

5. To look for teachers’ variability in approaching error gravities vis-à-vis error types and sources.

6. To assess the extent to which prioritizing errors in education can have pedagogical gains in learning and teaching. So, prioritizing errors helps the teacher know which error to tackle first and this informs teaching because it affects syllabus and curriculum design, teaching and evaluating methods (feed-forward instead of feedback) and decision making.

7. To come up with a continuum of error gravity that ranges from the substantial to marginal errors via-à-vis error types and error sources.

In this study, there is a triangulation of quantitative and qualitative methods. This triangulation is required to ensure validity and reliability.

Therefore, the triangulation in the present research is established through: Quantitative tool, which consists in the scale item questionnaire with teachers. Qualitative tool, which consists in the semi-structured interview with teachers.

3.3.2.2 Teachers’ questionnaire

This section will tackle the rationale behind the adoption of the questionnaire in this research, the objective, questionnaire design consideration and the description of the questionnaire.

3.3.2.3. Rationale and Objectives

The use of the questionnaire is considered as one of the most important and reliable tools for data collection. Therefore, using questionnaires provides an opportunity for the researcher to obtain the following types of information from the respondents:

a) Biographical details (age, educational qualifications, income and so on) b) Typical behavior (which television programs they favor and so on) c) Opinions, beliefs, and convictions about any topic or issue.

d) Attitudes (for example towards Outcomes Based Education). (Welman&Kruger, 2005)

In the present study, the questionnaire is administered to:

1. Provide insights about their differences in dealing with students’written error seriousness.

2. Investigate teachers’ reactions towards errors in using different criteria.

3.3.2.4. Questionnaire design considerations

• The incentive:

In this part of the questionnaire the researcher introduces the topic of the inquiry. Also, he/she describes the objective of the questionnaire to inform the participants. It is necessary to highlight the importance of their participation in the research findings. As a matter of fact, the researcher should convince the participants to contribute in this study.

3.3.2.5. Description of the questionnaire

Teachers at higher education levels are given the questionnaire.The scale item s in the questionnaire range from 1 to 5.Therefore, the questionnaire consists o f statements with errors that need to beassessed.Regarding the mistake criterio n, the questionnaire was given out twiceTo put it another way, mistakes are ini tially assessed using the acceptability criterion, and then they are reassessed us ing the annoyance criterion.While the second number showed the error's serio usness based on annoyance, the first number indicated the error's seriousness b ased on the acceptability criterion.Teachers therefore allocated two numbers to each error kind.

The first number represented the error's seriousness based on the acceptability criterion, while the second number represented the error's seriousness based onthe annoyance criterion.For instance, if the first error had a score of 2 and th e second received a score of 5 in the acceptability criterion, this indicates that t he second error is more ignificant than the first since it is more unacceptable. Because teachers are supposed to evaluate the importance of errors rather than identify or classify them, the flaws in the questionnaire are underlined.

After the second semester examinations are corrected, teachers receive the que stionnaire so they have time to respond.The administration of the questionnair e takes roughly four hours.This kind of questionnaire was used to assess pupil' mistakes.The results of the questionnaire will be discussed in the findings and discussion chapter, but first the teachers' interviews the second data gathering tool—should be clarified.

3.3.2.6. Teachers’ interview

The following section will zero in on the rationale and the objectives when designing the interview.

3.3.2.7. Rationale and objectives

The interview is considered as one of the most important and useful data collection instruments. Thanks to the interview the researcher can get insightful and in -depth information from the real sources: the teachers themselves.

For the sake of triangulation, this interview tends to:

1- Investigate how university teachers account for writing students errors.

2- Know about teachers’ attitudes towards error gravity considerations in the realm of teaching writing

3- Get insights about the importance of evaluating errors on the basis of two separate criteria

4- Know the importance of weighing errors in teaching pedagogy.

3.3.2.8. Description of the interview

The semi-structured interview is opted for in this study to address four teachers to see how they differ in their error priority judgment. The teachers are two LTETs and two STETs. They belong to the questionnaire population. This interview is composed of five questions. They focus on the three main research questions. Therefore, Wilkinson and Birmingham (2003) claim that the semi-structured interview takes place when the interviewer directs the interview more closely than the informal unstructured interview. The questions are predetermined to the interviewee with some flexibility “to shape the flow of the information” Wilkinson and Birmingham (2003:45)

The statistics in this study are based on the (SPSS) statistical tool. It is applied on the data of the questionnaire to extract teachers’ variation in responding to the study errors in the light of teaching experience and criteria selection. A two­tailed T-test was used to account for teachers’ variability in judging the error seriousness. T-test indicates that there is a significant independence for the equality of means if (p) value is less or equal to (0.05). However, if it exceeds (0.05), we say that T-test indicates no significant independence for the equality of means. Moreover, the correlation between the study variables is calculated through Pearson’s correlation coefficient.

The last part of the analysis will focus on the explanation of the most serious error category which is the lexical one in this context. Therefore the researcher takes care of this phase through accounting for the psycholinguistic soures of errors: interlingual and intralingual.

3.4. Summary

This chapter presented the different procedures that were adopted to investigate teachers’ variability in tolerating students written errors. This chapter gave a description of the context of the study, the subjects who took part in the experiment, the reasons behind selecting them, their profiles, the different tools of data collection in terms of rationale, objective, design consideration and description.

The next part is devoted to the analysis of data and its interpretation. Therefore, we will focus on the research findings and discussion.

CHAPTER 4

FINDINGS AND DISCUSSION

4.1 . Introduction

This chapter is divided into two main sections. The first section will analyze the findings obtained from the different tools of data collection. The second one, on the other hand, will be concerned with discussing these results. However, before embarking on data analysis, error frequencies should be presented. The errors in this study are classified into five main categories. The frequency distributions of these errors are as follows: morphological errors 28.26%, lexical errors 30.07%, syntactic errors 22.23%, mechanical errors 16.83%, and miscellaneous errors 2.61%.Concerning the morphological error category, seven error types are illustrated and their frequencies is as follows: vt 0.82%, vf 11.44%, sv 3.43%, Art 4.08%, n 4.25%, Adj 3.76%, Adv 0.49%.

Table 4.1

Examples of morphological errors

Illustrations are not included in the reading sample

Lexical errors aremade up of four types, and their frequencies are as follows: wc 20.26%, wf 5,39%, PR 2,45%, and col 1,96%.

Table 4.2

Examples of Lexical errors

Illustrations are not included in the reading sample

As far as syntactic errors are concerned, there are three error types. Their frequencies are illustrated as follows: 13,73%, Ro 0,98% and Frag 8,01%.

Table 4.3

Examples of syntactic errors

Illustrations are not included in the reading sample

For mechanical errors, there are two error types. Their frequencies are as follows: Pun 1.63% and Sp 14, 71%.

Table 4.4

Examples of Mechanical errors

Illustrations are not included in the reading sample

The frequency of miscellaneous errors is 2.61%.

Table 4.5

Examples of Miscellaneous errors

Illustrations are not included in the reading sample

4.2 . Findings

This chapter analyses and interprets the data obtained from quantitative and qualitative data. Therefore, the findings were divided into four main sections. The first subheading will focus on the impact of teachers’ experience on their error gravity judgment of the five error categories. The second subheading will analyze the impact of criteria selection on teachers’ error gravity judgment of the five error categories. The third subheading will focus on the impact of teaching experience and criteria selection in shaping teaching hierarchies of the five error categories. The fourth subheading will provide a detailed analysis of the data provided by teachers’ semi- structured interview.

4.2.1. The impact of teachers’ experience on error gravity judgment variation of the five error categories

To analyze the impact of teachers’ experience on their severity score variation of the five error categories, we should focus on two main concerns. First, the analysis of Short Teaching Experience Teachers (STETs) and Long Teaching Experience Teachers (LTETs) judgment of error gravity using the Acceptability criterion. Second, the analysis of STETs and LTETs judgment of error gravity using the Irritation criterion. The analysis of their judgment is based on descriptive statistics based on SPSS statistical software. Pearson’s correlation coefficient is significant for STET (0.26 >0.01). However, it is not significant for LTET (0.00<0.01).Therefore there is a correlation between teaching experience and the five error categories evaluation of gravity in using both criteria for STETs. But, concerning LTETs, the correlation is null.

4.2.1.1. Analysis of STETs and LTETs judgment of error gravity using Acceptability criterion

This part compares the severity scores of STETs and LTETs responses to five error categories using Acceptability criterion.

An independent Samples T-test indicates that there is a significant independence for equality of means between two groups of teachers in judging the gravity of the five error categories ( p=0.009 <0.05). Scores were significantly higher for STETs (m=2345, SD= 361, 27) that for LTETs (m=1653, SD=654,63). In addition, table 4.6 presents the mean differences of scoring severity between both groups of teachers in each category. Therefore, the analysis will zoom on each error category per se.

Table 4.6 Descriptive Statistics for Errors Severity scoring of error categories in Acceptability criterion

Illustrations are not included in the reading sample

• Independence between different Morphological error types and both groups of teachers

An independent Samples T-test indicates that there is no significant independence for equality of means between two groups of teachers in judging the gravity of Morphological errors (p= 0.606>0.05). Therefore, Levene’s Test for equality of variances is valid (sign=0,062> 0.05).So, there is no independence and the scores are close between both groups of teachers (see Appendix 1). For STETs (M=447, SD=106) and for LTETs (M=409, SD=205).Although there is no significant independence of equality of means, the group of STETs is less tolerant to Morphological error types than their

LTETs counterpart. Therefore, the latter group is more lenient in judging the gravity of the Morphological error types.

We will fine tune the analysis by tackling each Morphological error type on its own. Therefore, the means of each category is used.

The figure below translates the severity judgment of each group of teacher vis- à-vis the Morphological error category which is composed of seven error types. STETs means of severity of these errors are: verb form (M=150,6), noun (M=82,1) adjective (M=69,7) , article (M=59,8), subject-verb agreement (M=52) verb tense (M=17,4) and adverb (M=15,8). However, LTET's means of severity of these errors are: verb form ( M= 159,4), noun (M=61,8), article (M= 59,4), subject-verb agreement (M=55,3), adjective( M=52,9), verb tense (M=11,8), and adverb (M=8,4). These figures show teachers' differences in scoring the morphological error types severity. It seems that both teacher groups give the same error priority to verb form, noun, and adverb. But they differ in prioritizing the remaining four categories. The latter are judged by STETs as follows: adjective, article, subject-verb agreement, and verb tense. However, LTETs judge them as follows: verb tense, adjective, subject-verb agreement, article. The analysis of the means of these seven error categories shows the lack of independence between both groups of teachers.

• Independence between different Lexical error types and both groups of teachers

An independent Samples T-test indicates that there is a significant independence for equality of means between two groups of teachers in judging the gravity of Lexical errors (p= 0.00 <0.05). Scores were significantly higher for STETs (M=880, SD=48, 77) than for LTETs (M=464, SD=206, 92).There is a striking difference between both groups of teachers. Thus, the results reflect the high rate of severity of the group of STETs. The group of LTETs, however, has been interpreted as more tolerant of all Lexical error types.

The results show the severity judgment of each group of teacher vis-à-vis the Lexical error category which is composed of four error types. STET’s judge the severity of these errors as follows: word choice (M=679,8), word form (M=141,2), pronoun (M=59,1) and collocation (M=44,3) However, LTET’s judge lexical error types as follows word choice (M=346,4), word form (M=83,3), pronoun(M=34,6) and collocation (M=26,6). There is a significant independence for equality of means between both groups. Therefore there is variability between both groups.

• Independence between different Syntactic error types and both groups of teachers

An independent Samples T-test indicates that there is no significant independence for equality of means between two groups of teachers in judging the gravity of Syntactic errors (p= 0.86 >0.05).So, there is no independence and the scores are close between both groups of teachers. For STETs (M=431, SD=228, 36) and for LTETs (M=416, SD=139, 75).The results show the dependence of equality of means between two groups. However, the group of STETs is slightly less tolerant than their counterpart.

The figure below illustrates the severity judgment of each group of teachers vis- à-vis the Syntactic error category which is composed of three error types. A comparison of means show that STETs judge the severity of errors as follows: (M=224, 7) sentence structure, (M=129, 2), fragmentation (M=32,8) run-ons. LTETs have the same priority to these errors as their counterpart, (M=228, 1) sentence structure,(M=5.25%), (M=118,7) fragmentation and (M=19,2) run-ons.

• Independence between different Mechanical error types and both groups of teachers

An independent Samples T-test indicates that there is a significant independence for equality of means between two groups of teachers in judging the gravity of Mechanical errors (p= 0.00 <0.05). Scores were significantly higher for STETs (M=515, SD=64, 80) than for LTETs (M=320, SD=120, 65). Again, while the STETs group shows a great deal of severity in their judgment of gravity, the group of LTETs is much more lenient in responding to the seriousness of Mechanical error types.

The pie charts below illustrate the severity judgment of each group of teachers vis-à-vis the Mechanical error category which is composed of two error types. STETs judge the severity of these errors as follows (M=48.7) punctuation and (M=467) spelling. LTETs have the same priority to these errors as their counterpart (M=34, 6) punctuation and (M=304, 8) spelling. T-test indicates a significant independence for equality of means in these two error types (p= 0.00 <0.05).Thus there is an independence between both groups of teachers.

• Independence between Miscellaneous errors and both groups of teachers

An independent Samples T-test indicates that there is a significant independence for equality of means between two groups of teachers in judging the gravity of Miscellaneous errors (p= 0.00 <0.05). Scores were significantly higher for STETs (M=87, SD=6.01) than for LTETs (M=43, 30 SD=26, 36).As a result, the scores record a high gravity evaluation variability between both groups. In addition, the group of STETs is less tolerant in the gravity evaluation of Miscellaneous error than the other group of LTETs.

4.2.1.2. Analysis of STETs and LTETs judgment of error gravity using Irritation criterion

This part compares the severity scores of STETs and LTETs response to five error categories using the Irritation criterion.

An independent Samples T-test indicates that there is no significant independence for equality of means between two groups of teachers in judging the gravity of the five error categories (p=0,083>0.05). So, both groups Scores are close. So, Levene’s Test for equality of variances seems to be valid (sig= 0.016<0.05) But, Scores were slightly higher for STETs (M=2141, 30 SD=257, 31) than for LTETs (M=1767, 80, SD=570, 54).Therefore, the analysis will zoom on each error category.

Table 4.7 Descriptive Statistics for scoring severity of error categories in Irritation criterion

Illustrations are not included in the reading sample

• Independence between different Morphological error types and both groups of teachers

An independent Samples T-test indicates that there is a significant independence for equality of means between two groups of teachers in judging the gravity of Morphological errors (p= 0.025<0.05). For STETs (M=304, 10, SD=47) and for LTETs (M=426, 90, SD=152).T-test results show that the group of STETs is more tolerant to Morphological error types irritation judgment than their LTETs counterpart. Therefore, the latter group is less lenient in judging the gravity of Morphological error types.

We will fine tune the analysis by tackling each Morphological error type per se. Therefore, the means of each category are used.

The results translate the severity judgment of each group of teacher vis-à-vis the Morphological error category which is composed of seven error types. STET's means of severity of these errors are: verb form (M=121,8), noun (M=51,8) adjective (M=40,5) , article (M=37,8), subject-verb agreement (M=36,2) verb tense (M=15,1) and adverb (M=9,6). However, LTETs’ means of severity of these errors are: verb form( M= 167,9), adjective( M=71,3), noun (M=70), subject-verb agreement (M=48,1), article (M= 45,3), and adverb (M=9,2) verb tense (M=6,4) . These figures show teachers' differences in scoring the morphological error types severity. It seems that both teacher groups give the same error priority to verb form, noun, and adverb. But, they differ in prioritizing the remaining four categories. The latter, are judged by STETs as follows: adjective, article, subject-verb agreement, and verb tense. However, LTETs judge them as follows: verb tense, adjective, subject-verb agreement, and article. The analysis of the means of these seven error categories shows the existence of a difference between both groups of teachers.

• Independence between different Lexical error types and both groups of teachers

An independent Samples T-test indicates that there is a significant independence for equality of means between two groups of teachers in judging the gravity of Lexical errors (p= 0.00 <0.05). Scores were significantly higher for STETs (M=834, SD=66, 39) than for LTETs (M=518, 20, SD=188,90).There is a striking difference between both groups of teachers in the judgment of the gravity of lexical errors in terms of irritation. Thus, the results reflect the high rate of severity of the group of STETs. However, the LTETs severity scores results reflect the leniency of the latter group comparing to the other group’s results.

The figure below illustrates the severity judgment of each group of teacher vis- à-vis the Lexical /7error category which is composed of four error types. STETs judge the severity of these errors as follows: word choice (M=650,5), word form (M=121,7), pronoun (M=61,8) and collocation (M=46,1). However, LTETs judge lexical error types as follows: word choice (M=387,5), word form (M=90,1), pronoun(M=40,6) and collocation (M=30,2). There is a significant independence for equality of means between both groups. Therefore there is variability between both groups.

• Independence between different Syntactic error types and both groups of teachers

An independent Samples T-test indicates that there is no significant independence for equality of means between two groups of teachers in judging the gravity of Syntactic errors (p= 0.84 >0.05).So, there is no independence and the scores are close between both groups of teachers. For STETs (M=419, SD=170,16) and for LTETs (M=432,60,SD=127, 63).The results show the dependence of equality of means between the two groups. However, the groups of STETs are slightly less tolerant than their LTETs counterpart.

The results depicts the severity judgment of each group of teacher vis-à-vis the Syntactic error category which is composed of three error types. A comparison of means show that STETs judge the severity of errors as follows: (M=230, 9) sentence structure, (M=119,3), fragmentation (M=48,9) run-ons. LTETs have the same priority to these errors as their counterpart, (M=252,5) sentence structure (M=101) run-ons and (M=48,9) fragmentation.

• Independence between different Mechanical error types and both groups of teachers

An independent Samples T-test indicates that there is a significant independence for equality of means between two groups of teachers in judging the gravity of Mechanical errors (p= 0.00 <0.05). Scores were significantly higher for STETs (M=512,70 , SD=101,62) than for LTETs (M=342,30 ,SD=115,46).Again, STETs group shows a great deal of severity in their judgment of gravity. However, the group of LTETs is much more lenient in responding to the seriousness of Mechanical error types.

The result illustrates the severity judgment of each group of teachers vis-à-vis the Mechanical error category which is composed of two error types. STETs judge the severity of these errors as follows (M=46.9) punctuation and (M=465, 8) spelling. LTETs prioritize these errors as (M=268,9) punctuation and (M=37,4) spelling. T-test indicates a significant independence for equality of means in these two error types (p= 0.00 <0.05).Thus there is an independence between both groups of teachers. This independence is obvious through the figures’ representation in the chart. Therefore we have two (L) forms that translate this independence.

• Independence between Miscellaneous errors and both groups of teachers

An independent Samples T-test indicates that there is a significant independence for equality of means between two groups of teachers in judging the gravity of Miscellaneous errors (p= 0.00 <0.05). Scores were significantly higher for STETs (M=88, 30, SD=6.46) than for LTETs (M=62, SD=15,96). As a result, the scores record high gravity evaluation variability between both groups. In addition, the group of STETs is less tolerant in the gravity evaluation of miscellaneous error than the other group of LTETs.

4.2.2. The impact of criteria selection on teachers’ error gravity judgment variation of the five error categories

To analyze the impact of criteria selection on teachers' severity score variation of the five error categories, we should focus on two main concerns. First, the analysis of STET's judgment of error gravity using acceptability and irritation criteria and second, the analysis of LTET's judgment of error gravity using acceptability and irritation criteria. In other words, this part of the analysis will diagnose the effect of two different criteria on each group of teachers. To investigate the existing correlation between these two criteria we resort to Pearson's correlation coefficient which is (0.00<0.01). Thus, the correlation is not significant between the two criteria and the judgment of the gravity of the five error categories.

4.2.2.1. Analysis of STET’s judgment of error gravity using acceptability and irritation criteria

This part compares the severity scores of STETs response to five error categories using Acceptability and irritation criteria.

An independent Samples T-test indicates that there is no significant independence for equality of means between two criteria in judging the gravity of the five error categories (p=0.083 >0.05). Scores were significantly higher for acceptability criterion (M=2345, SD= 361,27) than for irritation criterion (M=2141,30,SD=257,31).In addition, table 1 presents the mean differences of scoring severity between two criteria in the five categories. Therefore, the analysis will zoom on each error category per se.

Table 4.8 Descriptive Statistics for STETs judgment of Acceptability and irritation criteria

Illustrations are not included in the reading sample

• Independence between different Morphological error types and the two criteria for STETs

An independent Samples T-test indicates that there is no significant independence for equality of means between the two criteria in judging the gravity of Morphological errors (p= 0.083>0.05).For acceptability criterion (M=447, 10, SD=106) and for irritation criterion (M=304, SD=47). T-test results show that the group of STETs is more tolerant to Morphological error types irritation judgment than acceptability judgment. In addition, the pie chart illustrates the total scores as 2345 for acceptability and 2141.3 for irritation.

We will fine tune the analysis by tackling each Morphological error type per se. Therefore, the means of each error type are used.

The results present the severity judgment means of STETs vis-à-vis the morphological error types using the two criterion. In using acceptability criterion the errors are judged as follows: verb form (M=150,6), noun (M=82,1) adjective (M=69,7) , article (M=59,8), subject-verb agreement (M=52) verb tense (M=17,4) and adverb (M=15,8). Whereas, in using irritation criterion the severity means of these errors are: verb form( M= 167,9), adjective( M=71,3), noun (M=70), subject-verb agreement (M=48,1), article (M= 45,3), and adverb (M=9,2) verb tense (M=6,4) . These figures show teachers' differences in scoring the morphological error types in adapting two sets of criteria. The analysis of the means of these seven error categories shows the existence of a slight difference between both criteria.

• Independence between different Lexical error types and the two criteria for STETs

An independent Samples T-test indicates that there is no significant independence for equality of means among STETs in adapting two different sets of criteria in judging the gravity of Lexical errors (p= 0.00<0.05). For acceptability criterion (M=880, SD=48,77) and for irritation criterion (M= 834,SD=66,39).

The results illustrate the severity judgment of STET’s vis-à-vis the Lexical error category which is composed of four error types. STET’s judge the severity of these errors according to acceptability as follows: word choice (M=679,8), word form (M=141,2), pronoun (M=59,1) and collocation (M=44,3), whereas lexical error types are judged in the light of irritation criterion as follows: word choice (M=346,4), word form (M=83,3), pronoun (M=34,6) and collocation (M=26,6). There is no significant independence for equality of means between both criteria.

• Independence between different Syntactic error types and the two criteria for STETs

An independent Samples T-test indicates that there is no significant independence for equality of means between two criteria in judging the gravity of Syntactic errors (p= 0.083>0.05). For acceptability criterion (M=431 SD=228, 36) and for irritation criterion (M=419, SD=170,16).

A comparison of means shows that STETs judge the severity of errors in using acceptability as follows: (M=224,7) sentence structure, (M=129, 2), fragmentation (M=32,8) run on. However, in adapting irritation criterion, (M=230.9) sentence structure, (M=119.3) fragmentation and (M=22.9) run on.

• Independence between di fferent Mechanical error types and the two criteria for STETs

An independent Samples T-test indicates that there is a significant independence for equality of means between two criteria in judging the gravity of Mechanical errors (p= 0.00 <0.05). Scores were significantly higher for irritation (M=101,62, SD=88,30) and for acceptability (M=64,80 ,SD=87).The figure below illustrates the severity judgment of STETs vis-à-vis the Mechanical error category in implementing two sets of criteria. STET's judge the severity of these errors as follows: For acceptability criterion (M=48,7) punctuation and (M=467,8) spelling. However, for irritation, criterion errors are prioritized as follows: (M=268,9) punctuation and (M=73,4) spelling. T-test indicates a significant independence for equality of means in these two error types between the two criteria (p= 0.00 <0.05).

• Independence between different error Miscellaneous types and the two criteria for STETs

An independent Samples T-test indicates that there is no significant independence for equality of means between two groups of teachers in judging the gravity of Miscellaneous errors (p= 0.083 <0.05). Scores were higher for irritation criterion (M=88,30 SD=6,46) and for acceptability (M=87 SD=6,01). As a result, the scores record no gravity evaluation variability between two criteria.

4.2.2.2 Analysis of LTETs judgment of error gravity using acceptability and irritation criteria

This part compares the severity scores of LTETs response to five error categories using Acceptability and irritation criteria.

An independent Samples T-test indicates that there is no significant independence for equality of means between two criteria in judging the gravity of the five error categories ( p=0.083 >0.05). Therefore Levene’s test for equality of variance is valid. Scores were significantly lower for acceptability criterion (M=1653, SD= 654) than for irritation criterion (M=1767,80, SD=570,54). In addition, table 17 presents mean differences of scoring severity of the two criteria in the five categories. Therefore, the analysis will zoom on each error category per se.

Table 4.9 Descriptive Statistics for LTET’s judgment of Acceptability and irritation criteria

Illustrations are not included in the reading sample

• Independence between di fferent Morphological error types and the two criteria for LTETs

An independent Samples T-test indicates that there is no significant independence for equality of means between the two criteria in judging the gravity of Morphological errors (p= 0.083>0.05). For acceptability criterion (M=406, SD=205) and for irritation criterion (M=426.90, SD=152).T-test results show that the group of LTETs is more tolerant to Morphological error types irritation judgment than acceptability judgment.

We will fine tune the analysis by tackling each Morphological error type per se. Therefore, the means of each error type is used. The figure below presents the severity judgment means of LTETs vis-à-vis the morphological error types using the two criteria. In using acceptability criterion the errors are judged as follows: verb form (M=159,4), noun (M=61,8) adjective (M=69,7) , article (M=59,4), subject-verb agreement (M=55.3) verb tense (M=11.8) and adverb (M=8.4). Whereas, in using irritation criterion the severity means of these errors are: verb form( M= 167,9), adjective( M=71,3), noun (M=70), subject-verb agreement (M=48,1), article (M= 45,3), and adverb (M=9,2) verb tense (M=6,4) These figures show teachers' differences in scoring the morphological error types in adapting two sets of criteria. The analysis of the means of these seven error categories shows the existence of a slight difference between both criteria.

• Independence between different Lexical error types and the two criteria for LTETs

An independent Samples T-test indicates that there is no significant independence for equality of means among LTETs in adapting two different sets of criteria in judging the gravity of Lexical errors (p= 0.083>0.05). For acceptability criterion (M=464 SD=206,92) and for irritation criterion (M= 518, SD=188,90).

The results illustrate the severity judgment of LTET’s vis-à-vis the Lexical error category which is composed of four error types. They judge the severity of these errors according to acceptability as follows: word choice (M=346.4), word form (M=83.3), pronoun (M=34.6) and collocation (M=26.6) whereas, lexical error types are judged in the light of irritation criterion as follows: word choice (M=387.5), word form (M=90.1), pronoun (M=40.6) and collocation (M=30.2). The means are higher for irritation criterion than for the acceptability one.

• Independence between different Syntactic error types and the two criteria for LTETs

An independent Samples T-test indicates that there is a significant independence for equality of means between two criteria in judging the gravity of Syntactic errors (p= 0.00<0.05). For acceptability criterion (M= 416 SD=139.75) and for irritation criterion (M=432, SD=127.63).

A comparison of means shows that LTETs judge the severity of errors in using acceptability as follows: (M=228.1) sentence structure, (M=118.7), fragmentation (M=19.2) run-ons. Whereas, in adapting the irritation criterion (M=252.5) sentence structure, (M=48.9) fragmentation and (M=101) run-ons. The figures show that there is a discrepancy of means in run-ons and fragmentation from one criterion to another.

• Independence between different Mechanical error types and the two criteria for LTETs

An independent Samples T-test indicates that there is a significant independence for equality of means between two criteria in judging the gravity of Mechanical errors (p= 0.00 <0.05). Scores were significantly higher for irritation (M=342.30, SD=115.46) and for acceptability (M=320, SD=120.65).The figure below illustrates the severity judgment of LTETs vis-à-vis the Mechanical error category in implementing two sets of criteria. They judge the severity of these errors as follows: For acceptability criterion (M=34.6) punctuation and (M=304.8) spelling. However, for irritation, criterion errors are prioritized as follows: (M=268,9) punctuation and (M=73.4) spelling. T-test indicates a significant independence for equality of means in these two error types between the two criteria (p= 0.00 <0.05).

• Independence between different error Miscellaneous types and the two criteria for LTETs

An independent Samples T-test indicates that there is no significant independence for equality of means between two groups of teachers in judging the gravity of Miscellaneous errors (p= 0.083 <0.05). Scores were higher for irritation criterion (M=62, SD=15.96) and for acceptability (M=43.30 SD=26.36). As a result, the scores record no gravity evaluation variability between two criteria.

4.2.3. The impact of teaching experience and criteria selection in shaping severity hierarchies of the five error categories

Shaping error seriousness hierarchies is the ultimate concern of our study in particular and of any error gravity research in general. Therefore, this part of analysis will focus on ranking errors in the light of teaching experience and criteria selection. The results show the existence of six different error hierarchies and they are presented as follows.

4.2.3.1. Analysis of the hierarchical order of errors according to acceptability criterion for the two groups.

STETs group is less tolerant of the evaluation of the five error categories than the LTETs. Therefore, we can draw two hierarchies of errors ranging from the most serious error to the least serious error.

Table 4.10 Error gravity hierarchy of STETs group in Acceptability criterion

Illustrations are not included in the reading sample

For example:

Tunisians themselves must develop their vivion for the country.(lexical errors)

... the most important solution tustravelagainsvoilance. (mechanical errors)

Encourage Investissersto invester (morphological errors)

The number of the viseters from the athercontreydeclin as a result the declin of devises (syntactic errors)

Investment étrange « Forren ». (miscellaneous error)

Table 4.11 Error gravity hierarchy of LTETs group in Acceptability criterion

Illustrations are not included in the reading sample

For example these sentences could reflect the above error hierarchy :.

il result chomage ...(Lexical error)

After revolution, The Tunisian business very bad, because the situation economic change. (syntactic error)

many people etrangerd’ont travel, or go in Tunisia (morphological errors)

before révolution many things are changed . (mechanical error)

Tunisi is creation plusieur travail thè a tuni (miscellaneous error)

These two hierarchies show that STETs and LTETs groups have similar error gravity preferences of lexical, Morphological and Miscellaneous error categories. However, we can draw significant differences. So, while Mechanical errors have the second rank with STETs group, it is ranked the fourth with LTETs group. Also, Syntactic errors are ranked the last but one with the first group of teachers, but, it is prioritized with LTETs group to be ranked the second.

Apart from error categories, both groups of teachers vary in prioritizing error types in the Morphological error category. Thus, STETs give the following error priority: verb form, noun, adjective, article, subject-verb agreement, verb tense and adverb. Whereas, LTETs prioritize errors as follows: verb form, noun, verb tense, adjective, subject-verb agreement, article; and adverb.

4.2.3.2 Analysis of the hierarchical order of errors as judged by two groups of teachers according to irritation criterion

STETs group is less tolerant in the gravity evaluation of the lexical, Mechanical and Miscellaneous error categories than the LTETs that performmore leniency in judging the gravity of these three error categories. However, Morphological and Syntactic error categories are judged as more serious by STETs group than the LTETs group. Therefore, we can draw a hierarchy of errors ranging from the most serious error to the least serious error.

Table 4.12 Error gravity hierarchy of STETs group in Irritation criterion

Illustrations are not included in the reading sample

First, on economie, the croissant is negative betwin revolution and after car PIB is bad and decroissant.(lexical errors)

The recentage of the homeless, jobless and the poor pupels is so haier than the other countries.(mechanical errors)

... and so much problems OR difficults has been studies. (Syntactic errors)

The situation in Tunisia become more difficult and resultthe chomage.(morphological

errors)

(But without actrau vision to guide it on this jourme.(Miscellaneous errors)

Table 4.12 Error gravity hierarchy of LTETs group in Irritation criterion

Illustrations are not included in the reading sample

First, In fact the Picture of political conflict, Business landscape have an important degradation.(lexical errors)

.. .advice business and customers to products and advice the in them in this fenominal to save different possibility of jobs with jobless peoples diplomats causes it’s call inflation.(Syntactic errors)

Tunisian business have many problem, this problem handicape the economie in Tunisia. (Morphological errors)

The number of the viseters from the ather contrey declin as a result the declin of devises.(Mechanical errors)

Tunisia nedie influse sience revolution . (Miscellaneous errors)

These two Hierarchies show that STETs group and LTETs group have similar error gravity judgment preferences of lexical and miscellaneous errors. However, we can draw significant differences with the three remaining error categories. So, while Mechanical errors have the second rank with STETs group, it is ranked the fourth with LTETs group. Also, Syntactic errors have the third rank with the first group of teachers, but, it is prioritized with LTETs group to be ranked the second.Finally, Mechanical errors are ranked the second by STETs group and the fourth in gravity by LTETs group.

Concerning the different error types in each category, both groups of teachers differ in prioritizing them in Syntactic, Morphological, and Mechanical error categories. First, with syntactic error types STETs prioritize fragmentation over run-ons.On the contrary, LTETs, prioritize run-ons over fragmentation. Second, both groups disagree in prioritizing the morphological error types. Thus, STETs give the following priority: verb form, noun, adjective, article, subject-verb agreement, verb tense and, adverb. However, LTETs assign the following

priority: verb form, noun, verb tense, adjective, subject-verb agreement, article, and adverb. Finally, mechanical errors are evaluated differently which implies differences in error hierarchies. For example, STETs perceive spelling as more irritating than punctuation. LTETs consider punctuation as more irritating than spelling. The results clarify how two different groups of teachers with different teaching experience differ in the judgment of certain error categories

4.2.3.3. Analysis of the hierarchical order of errors according to acceptability and irritation criteria of STETs

Table 4.12 Comparative error gravity Hierarchies of STETs group in both criteria

Illustrations are not included in the reading sample

Within the same group, we notice some differences in error gravity judgment of the five error categories. The ranking of errors is different from one criterion to another. Therefore, STETs judge Morphological errors as the third most serious error category in acceptability criterion and Syntactic errors as less serious in affecting acceptability judgment. However, the later error category is ranked the third in Irritation criterion followed by Morphological errors.

To conclude, STETs find lexical errors as the most unacceptable error category followed by Mechanical, Morphological, and Syntactic and Miscellaneous errors. Whereas, they find Lexical errors the most irritating error category followed by, Mechanical, Syntactic, Morphological and Syntactic errors.

In addition, we find a difference among STETs in prioritizing mechanical errors. With acceptability criterion, punctuation errors are more unacceptable than spelling ones. But, with irritation criterion, spelling errors are more irritating than punctuation errors. As a result, we can draw some conclusions about the correlation between Acceptability and Irritation criteria vis-à-vis error categories.

4.2.3.4. Analysis of the hierarchical order of errors according to acceptability and irritation criteria of LTETs

Table 4.13 Comparative error gravity Hierarchies of LTETs group in both criteria

Illustrations are not included in the reading sample

The ranking of errors is the same from one criterion to another for LTETs. Therefore, they are very consistent in their error gravity evaluation of the five error categories. To conclude, STETs find lexical errors the most unacceptable error category followed by Syntactic, Morphological, Mechanical and Miscellaneous errors. Similarly, they find Lexical errors the most irritating error category followed by Syntactic, Morphological, Mechanical and Miscellaneous errors.

In an attempt to fine-tune our analysis, we can zero on the different error types under each category. As a matter of fact, LTETs have the same priorities of lexical, mechanical and miscellaneous error types except for syntactic and morphological ones. As far as syntactic errors are concerned, the hierarchy in acceptability criterion is as follows: fragmentation, run-ons, and sentence structure. Whereas in irritation criterion becomes as follows: sentence structure ,run-ons, and fragmentation. In addition, the hierarchies present a difference in prioritizing morphological error types .In acceptability criterion, we have the following order: verb form, noun, article, subject-verb agreement, adjective, adverb, and verb tense. And in irritation criterion errors ranking becomes verb form, adjective, noun, subject-verb agreement, article, adverb, verb tense. As a result, we can draw some conclusions about the correlation between Acceptability and Irritation criteria vis-à-vis error categories.

In this research, lexical errors are the most serious. Therefore, the explanation will focus only on this category. In fact, 30% of lexical errors are intralingual and 70% of lexical errors are interlingual. So, Arabic and French languages are the main source of lexical errors.

Apart from T-test results, a discriminant Analyses is used primarily to predict membership in two or more mutually exclusive groups. In order to achieve the validity of the discriminant Analysis, we should check the hypothesis of equality of covariance matrices to determine whether two or more covariance matrices are equal. Therefore, according to Box’s M test, the Null Hypotheses of equality of matrices (variance-covariance) is (sign> 5%). So, the results of Fisher’s discriminant function are valid.

Table 4.14

Box’s M test Results

Illustrations are not included in the reading sample

Table 4.15 Discriminant Analyses: Classification Results

Illustrations are not included in the reading sample

The classification results show that 75% of original observations and 75% of Cross-validated observations are correctly classified.

According to acceptability and irritation criteria, we remark that there are five observations which are not well classified. For example, the results show that two LTETs are supposed to be in the group of STETs. And, three STETs are supposed to belong to LTETs.

4.2.4. Analysis of the data provided by teachers’ semi- structured interview

Four teachers participate in the semi-structured interview, two teachers are short experienced and two are long experienced. They will respond to five questions.

The first and the second questions are about the impact of teaching experience on judgment. In answering the first question, the four teachers agree that not all errors are equal. For short experience teachers, “written errors are critical by themselves and teachers should prioritize his feedback by focusing on particular areas of seriousness”. For long experience teachers, «In fact, errors are errors, but when it comes to practice we can shed more light on some errors rather than others.” As far as the second question is concerned, the four teachers agree that experience can affect teachers' judgment of error seriousness. For example,

SET1 said, "more experienced teachers may be more tolerant of students' errors than less experienced teachers because their experience helps them to be more familiar with students' deviant linguistic forms". For LET1 added that "teaching experience has a great impact in shaping teachers' judgment of error seriousness. In fact, majoring the seriousness of an error is a tuff task and therefore it deserves experience".

The third interview question sheds some light on the role of using criteria when treating students’ written errors. SET1 answered," I ask myself this question frequently before embarking on any correction in fact deciding upon the criteria depends on the students and the topic." SET2 said," For me, frequent errors should be treated first and later we can focus on errors that affect communication and understanding". Whereas, The two long experienced teachers show no interest in using criteria when treating students' written errors.LET1 argued, " the criteria does not matter for me because it is the error that matters most" . LET2 claimed, "I use to focus on criteria but now I feel that I have a set of integrated criteria in my mind which enable me to be more automatic in my practice. The findings of this question match those of the questionnaire in the point that more experienced teachers are more automatic in their evaluation.

The two remaining questions put the emphasis on the impact of the selected criteria in creating The two short experienced teachers believe that the criteria selection does not affect the judgment of error seriousness judgment. Whereas, long experience teachers argued that the judgment of the error gravity differs from one criterion to another. For example, LET1 thought that" the shift in criteria perception causes a difference in the judgment of the seriousness of some errors". Also, LET2 said, "apart from experience, criteria selection creates many differences in the evaluation because adapting a new criterion means having a new vantage point of perception."The four teachers agree upon the fact that changing the criterion engenders a change in the hierarchy of errors. As we mention error hierarchies, we may talk about the category which is on the top seriousness. Therefore, the fifth question addresses the issue of explaining the most serious error.Three teachers find it unimportant to explain the errors. However, just one long experience teacher seem to be motivated"to provide an explanation of the most serious errors:" it is important, if not compulsory to think about the sources of our learners' serious errors. So, understanding the origin of these errors enables teachers to provide better treatment and therefore makes them give priorities in teaching and testing".

4.2. Discussion

This part will interpret the findings obtained from the methods of data collection made use of in the present research, namely the teachers’ questionnaire and interview.

4.2.1. Teachers’ questionnaire

In the previous part, the findings of the T-test scores indicate the existence of differences in evaluating the gravity of written errors. These differences are due to two main variables: teaching experience and criteria selection.

To interpret the findings the following section will be divided into three main parts. The first subheading will focus on the effect of teachers' experience on their error gravity judgment of the five error categories. The second subheading will shed light on the impact of criteria choice on teachers’ error gravity judgment of the five error categories. Finally, the last subheading will examine the role of teaching experience and criteria selection in shaping severity hierarchies of the five error categories.

4.2.1.1 . The effect of teachers' experience on their error gravity judgment of the five error categories

In judging the five error categories in the light of acceptability and irritation criteria, T-test results reflect a significant independence in the equality of means between both groups of teachers. This proves that the teaching experience has an effect on the judgment of error seriousness. Examining the role of experience in this study matches with other studies that highlights raters’ professional experience effect in teachers’ rating of students’ writing (Barkaoui, 2010 and Vaughan, 1991). Furthermore, the concern with the impact of experience in error gravity judgment in this study represents a response to Myford and Wolfe’s (2009) call for more research to be conducted on the impact of rater experience on rater effects.

As far as STETs are concerned, the study results show that there is a positive correlation between teachers’ experience and error gravity judgment in accounting for acceptability and irritation criteria. Thus, this research finding correlates with the findings of Cascio and Valenzi (1977) who found a significant positive job-experience effect on rating errors. Hence, Teachers’ professional experience represents a dynamic variable in shaping the judgments of error gravity. So, experience makes the difference in determining STETs beliefs and practice in evaluation because it has to do with the so-called 'deficient knowledge of English' which is the main factor in determining error treatment practice (Medgyes, 1994, p.63).The findings show that STETs are more severe in responding to students' written errors than LTETs. Medgyes (1994) says that the picture of inexperienced teachers which emerges is one where they are usually preoccupied with accuracy, the formal features of English, the nuts and bolts of grammar, the printed word and formal registers. Many lack fluency, have a limited insight into the intricacies of meaning, are often in doubt about appropriate language use, have poor listening and speaking skills, and are not familiar with colloquial English. It is only logical to deduce that they place the emphasis on those aspects of the language they have better grasp of (Medgyes, 1994, p.59).Other studies match the findings of the current research, for example, James (1977) concluded that the inexperienced tended to mark more severely. The severity of STETs could be explained by their tendency to focus on correctness. In addition, the lack of experience presupposes lack of teachers’ exposure to students’ written errors. In addition, STETs lack of leniency could be explained by their prior learning experience. Thus, their performance is shapes by the remnant of their learning rather than actual classroom experience. Therefore, their pre-service experience may inform their cognition and act as a filter in teacher education decisions, and affect their subsequent practices as evaluators and judges (Ariogul, 2007; Borg, 2006).

However, there is no positive correlation for the more experienced teachers in accounting for acceptability and irritation criteria. Therefore, this research findings confirm Klores’s (1966) results which showed no effect for rater job experience. Furthermore, the results of the current research show that LTETs are more tolerant than STETs in the judgment of written errors.

Song and Caruso (1996) reported that the more years of experience a rater had in teaching; the more lenient he or she would be in rating ESL essays. The LTETs are more tolerant than STETs. Therfore, this variation in tolerance dates back to the role of differential language proficiency impact on teachers’ judgment. This goes hand in hand with Hughes & Lascaratou (1982) research findings. For them, experienced teacher tolerance being based on 'better knowledge of the language, particularly of the wide variety of acceptable structures' (Hughes &Lascaratou, 1982, p.180). By the same token, Sheorey (1985) concluded that tolerance of errors increases as language proficiency increases. By way of contrast, the research findings contradicts the conclusion of Birdsong and Kassen (1988) which shows that as people increase in language proficiency they become harsher in their judgments of error seriousness. They found that French-speaking teachers of French were harsher judges than English-speaking ones. The leniency of LTETs is the result of their familiarity with the erroneous linguistic forms because experienced teachers generally regard language as a means of achieving a communicative goal. Unlike inexperienced teachers who regard English primarily as a school subject to be learnt and only secondarily as a communicative medium to be used. This result correlates with the claim of Schwartz and white (2000). Hughes and Lascaratou (1982) suggest that the greater leniency may be attributed to the superior knowledge of the wide-ranging norms of English. Hence, teachers’ sharpened linguistic awareness, and their conscious knowledge about grammar and other aspects of language, potentially provide a sound basis for well - informed error evaluation, and professionally grounded

feedback on students’ errors. Another explanation of LTETs leniency could be expectation. It is so because teachers’ experience affects their repertoire of linguistic familiarity and therefore evaluative perception. This echoes the findings by Hauenstein (1992) who believed that increased experience leads to increased familiarity with organizational performance dimensions and expectations, and subsequent increases in ratings accuracy.

Teachers’ experience affects both groups differently. It is in the light of teaching experience that we decide upon the severity of teachers’ judgment. Therefore, we can conclude that like any other type of competence, teachers’ have a typical competence shaped through teaching experience. Teachers shape their practice from a gradual acquisition of cognitive, social, linguistic, and physical skills through experience (Bandura, 1982). These skills craft the so-called the Teachers’ Evaluation Competence. This newly coined competence explains the differences among teachers and invites more research on this particular area. Furthermore, the findings extracted from the discriminant Analyses paves the way to think about the reason behind the misclassification of the five observations.

4.2.1.2 The impact of teaching experience and criteria selection in shaping severity hierarchies of the five error categories

The rationale behind choosing teaching experience and criteria selection as two main variables in this piece of research is to come up with different error hierarchies. These hierarchies highlight the importance of perspectives. Therefore, Davies (1983:310) asserts that” Any error evaluation will be colored by the particular viewpoint from which it is carried out, and thus may not be consistent with evaluations made from other viewpoints”. Davies’ claim seems to be valid in this part of discussion because both groups of teachers differ in gravity judgment in the light of teaching experience and criterion choice. Therefore, we can get different error gravity hierarchies based on these two variables.

The need to establish error hierarchies in this research is rooted in many error gravity studies. Therefore, error gravity hierarchies have been investigated for Secong and foreign language classes in English (Johansson, 1978, Khalil, 1985, Sheorey, 1986, Santos, 1987, Tamiyama, 1980, MC Cretton and Rider 1993, James 1977, Hughes and Lascaratou 1982, and Davies 1983), French (Magnan, 1983, Piazza,1980),German (Delisle, 1982, Politzer, 1978); and Spanish (Chastin 1980, Guntermann, 1978, Gynan, 1985). These studies tend to establish hierarchies of errors that can be used as an aid for teachers to assess the students written work.

It appears to be a universal concern to prioritize errors. It could be explained by the human cognitive potential that tends to create hierarchies of difficulties which make the processing of knowledge easier. Therefore, human error processing appears to be hierarchically organized regarding the different complex neural systems (Krigolson and Holroyd, 2006; Doya, 2000; Doya et al., 2001).

When considering acceptability criterion, the teaching experience appears to have an impact on teachers ranking of errors. Therefore, the STETs consider lexical errors as the most unacceptable error category which is followed by mechanical, morphological, syntactic and miscellaneous errors. However, the LTETs consider lexical errors as the most unacceptable error category followed by syntactic, morphological, mechanical and miscellaneous errors. The two extremes remain situ whereas; the three other categories have witnessed a total reshuffle which mirrors teachers’ evaluation preferences. These findings seem to stand in stark contrast to a host of research findings which hold the mainstream assumption that consider “...all errors to be equally serious” (Van Meyer and Lorenz, 1984)

In irritation criterion, STETs rank lexical errors as the most irritating error category followed by mechanical, syntactic, morphological and miscellaneous error categories. However, LTETs have a different priority of these errors. They perceive lexical errors as the most irritating which are followed by syntactic, morphological, mechanical and miscellaneous errors. In adapting irritation criterion both groups prioritize errors differently. We come up with two irritation continua in which error categories could be classified as Ludwig (1982:275) described from an “unconcerned, distracted awareness of a communicative trait to a conscious, constant preoccupation with form, to the point that the message is totally obscured or lost”. As a matter of fact, we have different classifications. Therefore, the difference in the hierarchical order of the five error categories is explained by the effect of teaching experience in the seriousness judgment of teachers. This cements the impact of experience in shaping severity differences between teachers.

STETs differ in judging the gravity of errors while adapting acceptability and irritation criteria. The errors are prioritized as follows in acceptability criterion: lexical errors, mechanical errors, morphological errors, syntactic errors and miscellaneous errors. While, in adapting irritation criterion the hierarchy is as follows: lexical errors, mechanical errors, syntactic errors, morphological errors and miscellaneous errors. We notice a change in the ranking of two error categories from one criterion to another.

LTETs do not differ in judging the gravity of errors while adapting acceptability and irritation criteria. The errors are prioritized as follows in both criteria: lexical errors, syntactic errors, morphological errors, mechanical errors and miscellaneous errors. This consistency in adapting both criteria without any change in the priority of errors reflects the role of experience in achieving consistent evaluation among teachers. Myford and Wolfe (2009) called for research on the impact of rater experience on the stability of rater severity. Their findings indicated that there was more instability when raters are less experienced. Experienced teachers can automate well-learned routines, they achieve this because their cognitive skills become automatic with extensive practice (Chi et al, 1981). They achieve this because their cognitive skills become automatic with extensive practice (Chase & Simon, 1973).

The analysis of the study findings shows that there is a consensus upon the priority of lexical error category in acceptability and irritation criterion and by both groups of teachers. The findings corroborate with those of Hamdi (2016) which suggest that Tunisian EFL learners face more difficulties with lexical formal errors. Also, the findings echo those of Sheory's research in which lexical errors are perceived as the most serious error category in his error gravity hierarchy (p:310). In addition, Ellis (1994:63) confirms that “Researchers who have delt with the evaluation of errors have found out that lexical errors are considered the most serious, above all ..”

Therefore, the most serious error category should be explained to understand the sources of lexical errors Abott puts it:” The aim of an EA is to provide a psychological explanation” (1980:124). Recent studies (Bitchener,Youth and Cameron, 2005, Sheen 2007, Ellis et al 2008) have shown that when written corrective feedback is focused it is effective in promoting acquisition.

In this piece of research, the sentences including lexical errors are presented separately. Thus, the findings are in line with other results that use isolated sentences. For example, Ludwig (1982) and Rifkin and Roberts (1995) find that lexical errors are only irritating in these studies because word issues are more likely to stand out in isolated sentences than in paragraphs or essays.

Focusing on Lexical errors correlates with the findings of (Porte, 2008; Roberts & Cimasko, 2008; Johansson, 1978; Khalil, 1985) who focus on the importance of lexical errors. Thus, in our research the explanation of errors is postponed after evaluation phase, because we are concerned with the examination of the most serious error category. This focused explanation enriches the findings of this error gravity research because we focus on the most serious errors not only on evaluation but also in the psycholinguistic explanation and therefore in remediation. This finding seems to stand in stark contrast to a host of research findings which hold the mainstream assumption that NN teachers judge lexical errors as the least serious errors (Sheorey 1986, and McCretton & Rider 1993). In addition, the study results differ from those of James' study which noted that both groups of experienced and non- experienced teachers rated lexical error low in significance.

Coming up with error gravity hierarchies is common in second and foreign language classes in English: (Johansson, 1978; Khalil, 1985; Sheorey, 1986; Santos, 1987; Tomiyana, 1980); French: (Magnan, 1983; Piazza, 1980); German: (Delisle, 1982; Politzer, 1978); and Spanish: (Chastain, 1980; Guntermann, 1978; Gynan, 1985). Bartram and Walton (1991) assert that certain types of errors are more important than others. Therefore, it would be necessary for teachers to know the hierarchies of those errors. Moreover, Svartvik (1973:13) asserts that Error Analysis may help the teacher to set up a hierarchy of difficulties and to achieve a realistic ranking of teaching priorities at different levels. In fact prioritizing errors has a pragmatic dimension because error analysis helps teachers build their teaching priorities in an informed way. For example, according to Richards and Sampson (1974: 15), “At the level of pragmatic classroom experience, error analysis will continue to provide one means by which the teacher assesses learning and teaching and determines priorities for future effort.”Making a focused written corrective feedback promotes acquisition (Bitchener,Youth and Cameron, 2005, Sheen 2007, Ellis et al 2008).

Another finding presented by the Discriminant Analyses, predicts the existence of a membership in STETs and LTETs. In other words, we come up with a classificatory system that enables us to estimate teachers classification. For example, in our study, only 75% of teachers are well classified. Thus, three teachers from the STET's should be in the LTETs group and two teachers from the LTETs group should be in the STETs group. The Discriminant Analyses findings enables the researcher and the SLA practitioner to classify teachers in the light of teaching experience and criteria selection. In fact, by explicitly creating a hierarchy of error seriousness classification, we implicitly come up with a model of teacher’s classification. This explains that teachers’ evaluation output mirrors their perception, experience and teaching philosophy. Therefore, the teacher evaluates the error and the error opens the door to evaluate the evaluator himself. This analysis may create a continuum based on teaching experience through which, we judge the impact of teaching experience on error gravity seriousness. So, we come up with a dialogic relationship that sustains the applied potential of the error analysis.

4.2.2. Teachers’ interview

The data obtained from the interview reflect teachers differences in approaching error gravity. The findings of the interview correspond the claims proposed by (Vann, Meyer and Lorenz 1984) that faculty teachers respond differently to various kinds of errors ESL students make in their writing

In responding to the first research question, teaching experience seem to affect teachers’ error treatment practice. The third interview question shows that teachers agree on the impact of experience on the judgment of error seriousness. There is an agreement that more experienced teachers are more tolerant of students’errors.

In responding to the second research question, STETs show their interest in treating students' errors using criteria. However, LTETs' error treatment is not based on pre-selected criteria. This confirms Eckes (2008) findings of experienced raters who tended to see the scoring criteria as less important than younger raters. The finding that older and more experienced raters may make judgments about essay quality using factors not recognized by the scoring rubric is a serious threat to both the validity and reliability of assessment procedures. The findings of this question contradict two other studies in which expert raters used a wider range of criteria when making judgments compared with inexperienced raters (Cumming, 1990; Wolfe, Kao, & Ranney, 1998).The fourth question creates a disagreement among teachers. Therefore, STETs believe that the criteria selection has no impact on teachers’ seriousness evaluation. Whereas, LTETs claim that the choice of the criteria may affect teachers’ judgment. In fact, it is considered as another variable apart from experience that may color teachers’ judgment.

In responding to the third research question, respondents agree that all errors are not equally serious and therefore they have different priorities. The teaching experience enables LTETs to be more systematic in the evaluation of students' errors. The findings of this question confirm those of the questionnaire. Thus, the consistency of LTETs in prioritizing errors goes in harmony with the systematicity in the questionnaire.

In responding to the last question, teachers are not motivated to explain the most serious errors, except one LTET who finds it urgent to go back to error sources to explain the gravity of the most serious errors. The de-motivation of the majority of respondents reflects the lack of sensitivity on their part in explaining their students’ errors. In fact, explaining all errors is a very tiring task. It is because of this reason that I opted for error seriousness explanation to encourage teachers explain the errors in an informed way. To sum up, the results obtained from the interview confirm the ones obtained from the questionnaire to a certain extent.

CHAPTER 5

CONCLUSION

This concluding chapter will be devoted to the summary of the main findings of this research. It also aims at suggesting the implications of the study for teaching. This chapter will ultimately discuss the experimental study’s limitations and provide suggestions for future research.

5.1. The main findings

The research studied the impact of teaching experience and criteria selection on teachers’ seriousness evaluation of five error categories notably morphological, lexical, syntactic, mechanical and miscellaneous.

As far as the effect of teaching experience on teachers’ evaluation is concerned, the findings of the questionnaire show that there is a correlation between teaching experience and the evaluation of the five error categories in using acceptability and irritation criteria for STETs. However, there is no correlation for LTETs between acceptability and irritation criterion in considering the gravity of the five error categories.

T-test results reflect a significant independence in the equality of means between both groups of teachers. Therefore, the teaching experience has an effect on teachers’ reactions to errors. In terms of severity/tolerance dichotomy STETs are more severe in judging the acceptability of the five error categories than the LTETs. So, the more experienced the teacher is, the more tolerant he will be in accepting errors. In considering irritation criterion, T-test results show no significant independence in the equality of means between both groups of teachers. The results show that STETs are more severe than LTETs in judging the gravity of lexical, syntactic, mechanical and miscellaneous error categories. However, LTETs are more severe in judging the gravity of the morphological error category.

As far as the effect of criteria selection on teachers’ evaluation is concerned, the findings of the questionnaire show that there is a correlation between criteria selection and the evaluation of the five error categories in using acceptability and irritation criteria for LTETs. However, there is no correlation for STETs between acceptability and irritation criterion in considering the gravity of the five error categories. The questionnaire analyses show no independence in the judgment of STETs of the five error categories on the light of acceptability and irritation criteria.

For STETs the scores of acceptability criterion are higher than those of irritation one for all error categories except the miscellaneous error category. Therefore STETs are more tolerant in judging the gravity of written errors in using irritation criterion than acceptability criterion.

However, there exists no independence in the judgment of the LTETs of the five error categories. Therefore, the results show that LTETs are less serious in judging the gravity of written errors in using acceptability criterion than the irritation one in all error categories. Although there is no independence between both criteria among both groups, the results show that both groups of teachers differ in their evaluation gravity from one criterion to another.

The questionnaire results show six different error gravity hierarchies based on teachers’ experience and criteria selection. These hierarchies reflect teachers’ preferences in responding to the seriousness of the five error categories. Also, we come up with an arrangement of the substantial and marginal errors.

On the basis of the SPSS analysis, the discriminant Analyses test allows us to predict membership in STETs and LTETs. Therefore, the results show that two LTETs are supposed to be in the group of STETs. And, three STETs are supposed to belong to LTETs.

The findings coming from the semi-structured interview reflect that STETs and LTETs differ in their response to the interview questions. Accordingly, the findings of the interview confirm those of the questionnaires to a certain extent.

Finally, the study gives us an insight about the most serious error category as evaluated by both groups of teachers and provides us with its explanation. Therefore in this research error explanation comes after error evaluation.

5.2. Research limitations

Any research has limitations. In this regard, the present experimental study does have a number of limitations which should be acknowledged.

One limitation concerns the limited number of participants. Although this number can be a representative sample of teachers' population, it places some limitations on the generalization of the results in the study. A future study with more participants and a different setting could offer more substantial evidence to better understand this experimental study.

Another limitation is related to teachers' time constraint. In fact, they took much more time than it is required. Moreover, others refused to answer the questionnaire because of its length. Furthermore, the research focuses just on describing teachers’ reactions to written errors without examining the effectiveness of these comments on students’ performance.

This piece of research could be more practical when measuring the effect of error gravity judgment evaluation on learners’ uptake. But neither space nor time would allow for this.

5.3. Implications for teaching

The findings in the present study might offer some implications for the Tunisian teachers since the major objective of this research is to provide insights about teachers’ error treatment judgment of the seriousness of students’ written errors through testing the impact of teaching experience and criteria selection.

The first implication that emerges from the findings is to teach with a hierarchy in mind. That is to say, teachers’ prioritization of errors facilitates their job and enables them to prioritize their teaching objectives because they become aware of their students’ hierarchies of difficulties. Also, accounting for teachers’ areas of severity and tolerance helps learners notice the gap in their learning and know more about their serious errors that bother them in their learning process.

The second implication centers on the importance of criteria in error judgment. Therefore, Tunisian teachers should be aware that errors are not serious by themselves, but the way we approach them through the choice of our criteria makes the difference in accounting for their gravity. Therefore, teachers' awareness should be drawn to the impact of criteria selection on their teaching practice and self-reflection because "reflecting on [corrective feedback] serves as a basis both for evaluating and perhaps changing existing [corrective feedback] practices and, more broadly, for developing teachers' understanding of teaching and of themselves" (Ellis, 2009: 15).

The third implication emanates from our concern with the teaching experience; the discriminant test reflects a fusion in the boundaries between the experienced and the non-experienced teachers. So, it is preferable to blur these boundaries through exchanging experiences and enriching their experiences dialogically.

The forth implication holds the label ‘errors we live by”. Therefore teachers can build their teaching pedagogies on the most serious errors and hardships extracted from their learners either in a classroom context or in a no-contact context.

Finally, this research has an impact on curriculum design for Tunisian university teachers. For example, instead of having a syllabus to each faculty, we can make a national curriculum based on the most serious errors committed by our students on the basis of different criteria. Therefore, the objectives of the curriculum become tailored around the learners’ errors. As Lee (2011:9) notes, “teacher education holds the key to helping teachers understand the need to change and improve their feedback practices, as well as equiping them with the knowledge and skills to implement change”. Most recently, Troudi and Mazandarani (2025) recommended refocusing EFL teachers’ professional development to help them cope with the possible pedagogical challenges as an act of practical empowerment and critical remediation of the Educational system.

5.4. Conclusion

This piece of research provided an opportunity “to share insights, experiences, and tools across universities, particularly as many are solely responsible for teachers’ professional development in their department” (Allen & Negueruela- Azarola, 2010, p. 389). In addition, focusing on students’ written errors raises the awareness of teachers to be more humanistic in their treatment. Striking a bridge between the ESL teacher and the SLA researcher represents the ultimate goal of this research to shape a good theory of teaching because “there is nothing more practical than a good theory (Lewin, 1952: 169)

References

ABBOT, Gerry. 1980. “Towards a more rigorous error analysis.” International Review of Applied Linguistics 18.2: 121-133.

AbiSamra, N. (2003).An analysis of errors in Arabic speakers’ English writing. In Mourtaga, K. (Ed.),Investigating writing problems among Palestinian students studying English as a foreign language. Unpublished doctoral dissertation.

Albrechtsen, D., B, H., & Faerch, C. (1980). Native speaker reactions to learners‘ spoken interlanguage. Language Learning, 30, 394.

Allen, H. W., & Negueruela-Azarola,E. (2010). The professional development of future professors of foreign languages: Looking back, looking forward. Modern Language Journal, 94, 377-395.

Allwright, D., & Bailey, K. M. (1991). Focus on the language classroom: An introduction to classroom research for language teachers. Cambridge: Cambridge University Press.

Allwright, R. L. (1975). Problems in the study of the language teachers' treatment of learner error. In M. K. Burt & H. Dulay (Eds.), New directions in second language learning, teaching and bilingual education: 9th TESOL convention (pp. 96-106). Washington, DC: TESOL.

Allwright, R. L. (1998). Observation in the language classroom. London: Longman.

Ammar, A. (2008). Prompts and recasts: Differential effects on second language morphosyntax. Language Teaching Research, 12(2), 183-210

Ammar, A., & Spada, N. (2006). One size fits all? Recasts, prompts, and L2 learning. Studies in Second Language Acquiis,iti(nt, 28, 543-574.

Andrews, D. H., & Goodson, L. A. (1980). A comparative analysis of models of instructional design. Journal of Instructional Development, 3, 2-16.

Annett, J. (1969). Feedback and human behaviour. Harmondsworth, UK: Penguin.

Ariogul, S. (2007).Understanding foreign language teachers’ practical knowledge: What’s the role of prior anguage learning experience? Journal of Language and Linguistic Studies, 3(1), 168-181.

Ashwell, T. (2000). Patterns of teacher response to student writing in a multiple-draft composition classroom: Is content feedback followed by form feedback the best method? Journal of Second Language Writing, 9(3), 227-257.

Atkinson, R. C., & Shriffin, R. M. (1968). Human memory: A proposed system and its control processes. In K. Spence & J. Spence (Eds.), The psychology of learning and motivation (Vol. 2). New York, NY: Academic Press.

Ayoun, D. (2001). The role of negative and positive feedback in the second language acquisition of the passé composé and imparfait. The Modern Language Journal, 85 (2), 226-243.

Bachman, L. F. (1990). Fundamental Consideration in Language Testing. Oxford: OUP.

Bandura A. 1982. The psychology of chance encounters and life paths. Am. Psychol. 37:747-55

Barkaoui, K. (2010). Do ESL essay raters‘ evaluation criteria change with experience? A mixed methods, cross­sectional study. TESOL Quarterly, 44 (1), 31-55.

Bartram, M., & Walton, R. (1991). Correction. Hove, England: Language Teaching Publications.

Bell, N. (1992). The role of spoken error correction in second language acquisition: Issues in corrective technique. ORTESOL Journal, 13, 21-32. Bond University Course description. (2002). Gold Coast, Australia: Bond University.

Bitchener, J. (2008). Evidence in support of written corrective feedback. Journal of Second Language Writing, 17, 102-118.

Bitchener, J., & Knoch, U. (2010). The Contribution of Written Corrective Feedback to Language Development: A Ten Month Investigation. Applied Linguistics, 31(2), 193-214

Bitchener, J., Young, S., & Cameron, D. (2005). The effect of different types of corrective feedback on ESL student writing. Journal of Second Language Writing, 14, 91-205.

Borg, S. (2006). Teacher cognition and language education: Research and practice. London: Continuum.

Brookhart, S. M. (2003). Developing measurement theory for classroom assessment purposes and uses. Educational Measurement: Issues and Practice, 22(4), pp. 5-12.

Brown, J.D. (2001). Using Surveys in Language Programs. Cambridge, UK: Cambridge University Press.

Burt, M. & Kiparsky, C. (1972). 'Global and local mistake', in J.H. Schuman & N. Stenson(eds). New Frontiers in Second Language Learning, 71-80. Rowley, Mass: Newbury House.

Burt, M. K. (1975). Error analysis in the adult EFL classroom. TESOL Quarterly, 9 (1), 53-63.

Cardelle, M., & L. Corno (1981). "Effects on second language learning of variations in written feedback on homework assignments". TESOL Quarterly, 15: 25l-261.

Carless, D. (2006). Differing perceptions in the feedback process. Studies in Higher Education, 31(2), 219-233.

Carroll, S., & Swain, M. (1993). Explicit and implicit negative feedback: An empirical study of the learning of linguistic generalizations. Studies in Second Language Acquisition, 15, 357-386.

Carroll, S., Swain, M., & Roberge, Y. (1992). The role of feedback in adult second language acquisition: Error correction and morphological generalizations. Applied Psycholinguistics, 13 (2), 173-198.

Cascio, W. F., & Valenzi, E. R.( 1977). Behaviorally anchored rating scores: Effects of education and job experience of raters and ratees. Journal of Applied Psychology, 1977, 62, 278-282.

Celce-Murcia, M. (1985). Making informed decisions about the role of grammar in language teaching. Foreign Language Annals, 18, 297-301.

Chan, A. Y. (2010). Toward a taxonomy of written errors: Investigation into the written errors of Hong Kong Cantonese ESL learners. TESOL Quarterly, 44 (2), 295-319.

Chandler, J. (2003) The Efficacy of Various Error Feedback for Improvement in the Accuracy and Fluency of L2 Student Writing. Journal of Second Language Writing. 12: 267-296.

Chase, W.G., and H.A. Simon (1973). Perception in chess. Cognitive Psychology 1:33-81.

Chastain, K. (1980). Native speaker reaction to instructor- identified student second-language errors. The Modern Language Journal, 64 (2), 210-215.

Chastain, K. 1980(a). "Native Speaker Reaction to Instructor-Identified Student Second Language Errors,"Modern Language Journal 64(2):210-215.

Chaudron, C. (1986). Teachers' priorities in correcting learners' errors in French immersion classes. In R. R. Day (Ed.), Talking to learn: Conversation in second language acquisition (pp. 64-84). Cambridge: Newbury House Publishers.

Chaudron, C. (1988). Second language classroom: Research on teaching and learning. Cambridge: Cambridge University Press.

Chenoweth, N. A., Day, R. D., Chun, A. E., & Luppescu, S. (1983). Attitudes and preferences of ESL students to error correction. Studies in Second Language Acquisition, 6, 79-87.

Chomsky, N. (1965). Aspects of the theory of syntax.

Cambridge, MA: The M.I.T. Press.

Cohen, A. D. (1975). Error Correction and the Training of Language Teachers. The Modern Language Journal, 59, 414-

Cohen, A. D. (1987). Student processing of feedback in their compositions. In A. L. Wenden & J. Rubin (Eds.), Learner strategies in language learning (pp. 57-69).

Englewood Cliffs NJ: Prentice-Hall International.

Cohen, A. D. (1989). ‘Reformulation: A technique for providing advanced feedback in writing’. Guidelines 11/2: 1­9.

Cohen, A. D., & Cavalcanti, M. C. (1990). Feedback on compositions: Teacher and student verbal reports. In B. Kroll (Ed.), Second language writing (pp. 155-177). Cambridge: Cambridge University Press.

Corder, S. (1967). The significance of learners‘ errors. International Review of Applied Linguistics, 5 (4), 161-169.

Corder, S. (1971). Idiosyncratic dialects and error analysis. IRAL, 9 (2), 147-159 (Reprinted in J. H. Schumann, & N. Stenson (eds.), New frontiers of second language learning, (1975, pp. 100-113).

Corder, S. (1973). Introducing Applied Linguistics. Harmondsworth, U.K.: Penguin Books.

Corder, S. (1974). Error Analysis, In Allen, J.L.P. and Corder, S.P. (1974). Techniques in Applied Linguistics. Oxford: Oxford University Press.

Corder, S. (1981). Error analysis and interlanguage. Oxford: Oxford University Press.

Corder, S. 1975. 'Applied Linguistics and Language Teaching'. In Allen, J.P .B . and Corder S. Pit (eds) Papers in Applied Linguistics: Edinburgh Course in Applied Linguistics, Vol 2:1-15. London: Oxford University Press.

Croft, K. (1980). Reading in English as a second language: For teachers and teacher trainees (2nd ed.). Cambridge: Wintheop Publishers, Inc.

Crystal, D. (2003). A dictionary of linguistics and phonetics (5th ed.). London: Blackwell.

Cumming, A. (2001). Learning to write in a second language: Two decades of research. International Journal of English Studies, 1(2), 1-23.

Cunningsworth, A. (1987). ‘Coursebooks and conversational skills’ in L. E. Sheldon (ed.): ELT Textbooks and Materials: Problems in Evaluation and Development (ELT Documents 126). Oxford: Modern English Publications/The British Council.

Davies, A. (1991) The Native Speaker in AppliedLinguistics. Edinburgh: Edinburgh University Press

Davies, E.E., (1983). Error evaluation: the importance of viewpoint. English Language Teaching Journal 37 (4),304­311.

Delisle, H. (1982). Native speaker judgment and the evaluation of errors in German. The Modern Language Journal, 66 (1), 39-48.

Dingwall, S. (1984). Focus on the teacher: Some EFL/ESL teachers' views on teaching, with special reference to the treatment of error. Neueren Sprachen, 83 (4), 386-399.

Doya, K. (2000). Complementary roles of basal ganglia and cerebellum in learning and motor control. Current Opinion in Neurobiology, 10(6), 732-739.

Doya, K., Kimura, H., & Kawato, M. (2001). Computational approaches to neural mechanismsm of learning and control. IEEE Control Systems Magazine, 21(4), 42-54.

Driscoll, M. P. (2002). Psychological foundations of instructional design. In R. A. Reiser & J. V. Dempsey (Eds.), Trends and issues in instructional design and technology (57­69). Upper Saddle River, NJ: Pearson Education Inc.

Driscoll, M. P. (2007). Psychological foundations of instructional design. In R. A. Reiser & J. V. Dempsey (Eds.), Trends and issues in instructional design and Technology (2nd ed.), (36-44). Upper Saddle River, NJ: Pearson Education Inc.

Dulay, H., Burt, M., & Krashen, S. (1982). Language two. New York: Oxford University Press.

Eckes, T. (2008). Rater types in writing performance assessments: A classification approach to rater variability. Language Testing 25.2, 155-185.

Ellis, N. (2011). Implicit and explicit SLA and their interface. In C. Sanz, & R. Leow (Eds.), Implicit and explicit language learning (pp. 35-47). Washington DC: Georgetown University Press.

Ellis, N. C. (2013). Second language acquisition. In Oxford Handbook of Construction Grammar (pp. 365-378), G. Trousdale & T. Hoffmann (Eds.). Oxford: Oxford University Press. SecondLanguageAcquistionCxGPreprint.pdf

Ellis, N.C. (1996). Sequencing in SLA: Phonological memory, chunking, and points of order. Studies in Second Language Acquisition, 18, 91-126.

Ellis, R. (1994). The studies in second language acquisition. Oxford: Oxford University Press.

Ellis, R. (2001). Introduction: Investigating form-focused instruction. Language learning, 51(Supplement 1), 1-46.

Ellis, R. (2003). Task-based language learning and teaching. Oxford: Oxford University Press

Ellis, R. (2006). Researching the effects of form-focused instruction on L2 acquisition. AILA, 19: 18-41.

Ellis, R. (2008). The study of second language acquisition (2nd ed.). New York: Oxford University Press.

Ellis, R. (2009). A typology of written corrective feedback types. ELT 63 (2),97-107.

Ellis, R. (2013). Corrective feedback in teacher guides and SLA. IJLTR, 1 (3), 1-18.

Ellis, R., & Sheen, Y. (2006). Re-examining the role of recasts in SLA. Studies in Second Language Acquisition, 28, 575-600.

Ellis, R., Basturkmen, H., & Loewen, S. (2001). Learner uptake in communicative ESL lessons. Language Learning, 51, 281-318.

Ellis, R., Loewen, S., & Erlam, R. (2006). Implicit and explicit corrective feedback and the acquisition of l2 grammar. Studies in Second Language Acquisition, 28, 339­368.

Ellis, R., Sheen, Y., Murakami, M., & Takashima, H. (2008). The effects of focused and unfocused written corrective feedback in an English as a foreign language context. System, 36, 353-371.

Ellis, R., Sheen, Y., Murakami, M., & Takashima, H. (2008). The effects of focused and unfocused written corrective feedback in an English as a foreign language context. System, 36, 353-371.

Ensz, K. Y. (1982). French attitudes toward typical speech errors of American speakers of French. The Modern Language Journal, 66 (2), 133-139.

Eun-pyo, L. (2002). Error analysis on medical students" writing.

Fanselow, J. (1977). The treatment of error in oral work.

Foreign Language Annals, 10, 583-593

Fazio, L. L. (2001). The effect of corrections and commentaries on the journal writing accuracy of minority- and majority-language students. Journal of Second Language Writing, 10,235-249.

Ferreira, K. (2011). Brazilian English as a Foreign Language Teachers’ Beliefs about Grammar-Based Feedback on L2 Writing. Retrieved from http://lib.dr.iastate.edu/cgi/viewcontent.cgi?article=1038&con text=etd

Ferris, D. (1999). The case for grammar correction in L2 writing classes: A response to Truscott (1996). Journal of Second Language Writing, 8 (1), 1-11.

Ferris, D. R. (1995a). Student reactions to teacher response in multiple-draft composition classroom. TESOL Quarterly, 29 (1), 33-53.

Ferris, D. R. (1999). The case for grammar correction in L2 writing classes: A response to Truscott (1996). Journal of Second Language Writing, 8 (1), 1-11.

Ferris, D. R. (2002). Treatment of error in second language student writing. Ann Arbor, MI:The University of Michigan Press.

Ferris, D. R., & Hedgcock, J. S. (2005). Teaching ESL composition: Purpose, process, and

Ferris, D. R., & Roberts, B. (2001). Error feedback in L2 writing classes: How explicit does it need to be? Journal of Second Language Writing, 10 (3), 161-184.

Ferris, D., & Roberts, B. (2001). Error feedback in L2 writing classes: How explicit does it need to be?. Journal of Second language Writing, 10, 161-184.

Frantzen, D. (1995) The Effects of Grammar Supplementation on Written Accuracy in an Intermediate Spanish Content Course. Modern Language Journal, 79: 244­329.

Fregeau, L. A. (1999). Preparing ESL students for college writing: Two case studies. The Internet TESL Journal, 5(10) [On line]. http://iteslj.org/Articles/Fregeau-CollegeWriting.html

Gagné, R. M., Briggs, L. J., & Wager, W. W. (1992). Principles in instructional design. Fortworth, TX: Harcourt Brace College Publisher.

Galloway, V. B. (1980). Perceptions of the communicative efforts of American students of Spanish. The Modern Language Journal, 64 (4), 428-433.

Gass, S.M., & Selinker, L. (2008). Second Language Acqusition: An Introductory Course: Taylor & Francis Group

Glew, P. J. (1998). Verbal interaction and English second language acquisition in classroom contexts. Issues in Educational Research, 8 (2), 83-94.

Goldstein, L. M. (2004). Questions and answers about teacher written commentary and student revision: Teachers and students working together. Journal of Second Language Writing, 13, 63-80.

Greenfield, R. (2003). Collaborative e-mail exchange for teaching secondary ESL: A case study in Hong Kong. Language Teaching and Technology, 7 (1), 46-70.

Guntermann, G. (1978). A study of the frequency and communicative efforts of errors in Spanish. Modern Language Journal

Gynan, S. N. (1985). Comprehension, irritation, and error hierarchies. Hispania, 68 (1), 160-165.

Hairston, M. (1986). On not being composition slave. In C.W.Bridges (Ed.), Training the new teacher of college composition (pp. 117-124). Urbana, Ill: National Council of Teachers of English.

Hamdi, S. (2016). An analysis of lexical errors in the English compositions of EFL Tunisian learners. International Journal of Humanities and Cultural Studies, vol 2, (4).

Hammerly, H. (1991). Fluency and accuracy: Toward balance in language teaching and learning. Clevedon: Multilingual Matters LTD.

Hamp-Lyons, L. (1996). Ethical test preparation: The case of TOEFL. Paper presented at the Language Testing Research Colloquium, Chicago, IL.

Harmer, J. (1991). The practice of English language teaching (New ed.). London: Longman.

Hauenstein, N.M.A. (1992). An information-processing approach to leniency in performance judgments. Journal of Applied Psychology. 77(4): 485-493.

Hedgcock, J., & Lefkowitz, N. (1994). Feedback on feedback: Assessing learner receptivity to teacher response in L2 composing. Journal of Second Language Writing, 3, 141-163.

Hedge, T. (2000). Teaching and learning in the language classroom. Oxford:Oxford University Press.

Hendrickson, J. (1978). Error correction in foreign language teaching: Recent theory, research, and practice. Modern Language Journal, 62, 387-398

Hendrickson, J. M. (1977). Error analysis and selective correction in the adult ESL classroom: An experiment. ERIC Document Reproduction Services, ED 135260.

Hendrickson, J. M. (1980). Error correction in foreign language teaching: Recent theory, research and practice. In K. Croff (Ed.), Reading in English as a second language: For teachers and teacher trainees (2nd ed., pp. 153-173).

Cambridge: Winthrop Publishers, Inc.

Hendrickson, J. M. (1984). The treatment of error in writing work. In S. McKay (Ed.), Composing in a second language (pp. 145-159). Rowley MA: Newbury House Publishers.

Hicks, R.B. (1983 ). A Study of the Classroom Language of Tanzanian Primary School Teachers of English with Special Reference to their Interactive Discourse Ability and Linguistic Errors. Unpublished Ph.D. thesis. University of London, Institute of Education.

Honeycutt, L. (2001). Comparing e-mail and synchronous conferencing in online peer response. Written Communication, 18 (1) , 26-60. https://www.ijhcs.com/index.php/ijhcs/article/view/209/219

Hughes, A., & Lascaratou, C. (1982). Competing criteria for error gravity. ELT Journal, 36 (3),175-182.

Hultfors, P. (1986). Reactions to non-native English: Native English-speakers‘ assessments of errors in the use of English made by non-native users of the language. Stockholm: Almqvist & Wiksell International.

Hyland, F. (1998). The impact of teacher written feedback on individual writers. Journal of Second Language Writing, 7, 255-286.

Hyland, F. and Hyland, K. (2006) Feedback on Second Language Students’ Writing. Lang. Tech., 39: 83-101. Cambridge: Cambridge University Press.

Hyland, F., & Hyland, K. (2001). Sugaring the pill: Praise and criticism in writing feedback. Journal of Second Language Writing, 10, 185-212.

Hyland, K. (1990). Providing productive feedback. ELT Journal, 44 (4), 279-285.

Hyland, K., & Anan, E. (2006). Teachers’ Perceptions of Error: The Effects of First Language and Experience. System, 34, 509-519.

Jacobs, G. M., Curtis, A., Braine, G., & Huang, S. Y. (1998). Feedback on student writing: Taking the middle path. Journal of Second Language Writing, 7 (3), 307-317.

James, C. (1972). The diagnosis and analysis of error--some insights from linguistics. Audiovisual Language Journal: Organ of the British Association for Language Teaching,10 (2), 75-79.

James, C. (1998). Errors in language learning and use. New York: Longman.

James, C. (1977). Judgments of error gravities. English L anguage Teaching Journal, 2, 116 124.

Janopoulos, M. (1992). University Faculty Tolerance of NS and NNS Writing Errors: A comparison. Journal of Second Language Writing, 1(2), 109-121.

Jenkins, J. (2007). English as a lingua franca: Attitudes and identity. Oxford: Oxford University Press.

Johansson, S. (1975). Papers in contrastive linguistics and language testing. Lund, Sweden: Gleerup

Johansson, S. (1978). Studies of error gravity: Native reactions to errors produced by Swedish learners of English. Goteborg, Sweden: Acta Universitatis Gothobergensis.

Kasanga, L. A. (2006). Requests in a South African variety of English, pp 1; 65-89. Accessed on: 17/05/2010. Retrieved from: www.nie.edu.sg/profile/kasanga-luangaadrien-0.

Kasper, G. (Ed.) (1986). Learning, teaching and communication in the foreign language classroom. Arhus: Aarhus University Press.

Katayama, A. (2007). Learners’ perceptions toward oral error correction. In K. Bradford-Watts (Ed.) JALT2006 Conference Proceedings, Tokyo: JALT

Keh, C. L. (1990). Feedback in the writing process: A moral and methods for implementation. ELT Journal, 44 (4), 294-304.

Kepner, C. G. (1991). An Experiment in the Relationship of Types of Written Feedback to the Development of Second- Language Writing Skills. Modern Language Journal, 75 (3), 305-313.

Keshavarz, M. (2012). Contrastive analysis and error analysis (2nd ed.). Tehran: Rahamana Press.

Keshavarz, M. H.( 2015). Contrastive Analysis, and Interlanguage (Revisited Edition). Tehran: Rahnama Press.

Khalil, A. (1985). Communicative error evaluation: Native speakers‘ evaluation and interpretation of written errors of Arab EFL learners. TESOL Quarterly, 19 (2), 335-351.

Khalil, A. (1985). Communicative error evaluation: Native speakers‘ evaluation and interpretation of written errors of Arab EFL learners. TESOL Quarterly, 19(2), 335-351.

Klores, M. S.(1966) Rater bias in forced-distribution ratings. Personnel Psychology, 19, 411-421

Koshik, I. (2002). Designedly incomplete utterances: A pedagogical practice for eliciting knowledge displays in error correction sequences. Research on Language & Social Interaction, 35(3), 277.

Krashen, S., & Terrell, T. (1983). The natural approach: Language acquisition in the classroom. Oxford: Pergamon.

Kresovich, B. M. (1988). Error gravity: Perceptions of native-speaking and nonnative speaking faculty in EFL. (ERIC Document Reproduction Service No. ED 311732)

Krigolson, O. E., and Holroyd, C. B. (2006). Evidence for hierarchical error processing in the human brain. Neuroscience

Kwok, H. L. (1998). Why and when do we correct learner errors? An error correction project for an English composition class.

Accessed on: 16/04/2009. Retrieved from: http://sunzi1.lib.hku.hk/hkjo/view/45/4500101.pdf.

Lalande, J. F. (1982). Reducing composition errors: An experiment. Modern Language Journal, 66 (2), 140-149.

Lamberg, W. (1980). Self-provided and peer-provided feedback. College Composition and Communication, 314, 63­69.

language learning experience? Journal of Language and Linguistic Studies, 3 (1), 168-181.

Larsen-Freeman, D. (1986). Techniques and principles in language teaching. Oxford: Oxford University Press.

Language Teaching Research, 12, 183-210.

Larson, R. (1985). Emotional scenarios in the writing process: An examination of young writers' affective experiences. In M. Rose (Ed.), When a writer can't write: Studies in writers' block and other composing-process problems (pp. 19-42). New York: The Guilford Press.

Lee, I. (1997). ESL learners' performance in error correction in writing: Some implications for teaching. System, 25 (4), 465-477.

Lee, I. (2003). L2 writing teachers’ perspectives, practices and problems regarding error feedback. Assessment Writing, 8, 216-237.

Lee, Icy (2011). Feedback revolution: what gets in the way?. ELT Journal 65: 1-12.

Leki, I. (1990). Coaching from the margins: Issues in writing response. In B. Kroll (Ed.), Second Language Writing (pp. 57-68). Cambridge: Cambridge University Press.

Leki, I. (1991). The Preferences of ESL Students for Error Correction in College-Level Writing Classes. Foreign Language Annals, 24 (3), 203-218.

Lewin, K (1952). "Group decision and social change." In G. E. Swanson, T. M. Newcomb, and E. L. Hartley (eds.), Readings in So- cial Psychology: 459-473. New York: Holt

Lewis, M. (2002). Giving feedback in language classes. Singapore: SEAMEO Regional Language Centre.

Lightbown, P. M., & Spada, N. (1990). Focus-on-form and corrective feedback in communicative language teaching: Effects in second language learning. Studies in Second Language Acquisition, 12, 429-448.

Liu, J., & Sadler, R. W. (2003). The effect and affect of peer review in electronic versus traditional modes on L2 writing. Journal of English for Academic Purposes, 2, 193­227.

Lockee, B. B., Larson, M. B., Burton, J. K., & Moore, D. M. (2007). Programmed technologies. In J. M. Spector, M. D. Merrill, J. V. Merrienboer., & M. P. Driscoll (Eds.), Handbook of research for educational communications and technology (187-197). New York, NY: Lawrence Erlbaum Associates.

Loewen, S. (2004). Uptake in incidental focus on form in meaning-focused ESL lessons. Language Learning, 54, 153­188.

Loewen, S., & Erlam, R. (2006). Corrective feedback in the chatroom: An experimental study. Computer Assisted Language Learning 19(1), 1-14.

Long, M. H. (1991). Focus on form: A design feature in language teaching methodology. In K. D. Bot & R. B. Ginsberg & C. Kramsch (Eds.), Foreign language research in cross-cultural perspective (pp. 39-52). Amsterdam: Benjamins.

Long, M., Inagaki, S., & Ortega, L. (1998). The role of implicit feedback in SLA: Models and recasts in Japanese and Spanish. The Modern Language Journal, 82(3), 357-371.

Ludwig, J. (1982). Native-speaker judgments of second- language learners‘efforts at Communi cation: A review. The Modern Language Journal, 66 (3), 274-283.

Lunz, M. E., Stahl, J. A., & Wright, B. D. (1994). Interjudge reliability and decision reproducibility. Educational and Psychological Measurement. 54(4), 913-925.

Lyster, R. (1998). Negotiation of form, recast, and explicit correction. Language Learning, 48 (2), 183-218.

Lyster, R. (2007). Learning and teaching languages through content: A counterbalanced approach. Amsterdam: Benjamins.

Lyster, R., & Mori, H. (2006). Interactional feedback and instructional counterbalance. Studies in Second Language Acquisition, 28, 269-300.

Lyster, R., & Mori, H. (2008). Instructional counterbalance in immersion pedagogy. In R. Fortune & D. Tedick (Eds.), Pathways to bilingualism and multilingualism: Evolving perspectives on immersion education (pp. 133-151). Clevedon, UK: Multilingual Matters.

Lyster, R., 2002, Negotiation in immersion teacher-student interaction. International Journal of Education

Lyster, R., Lightbown, P. M., & Spada, N. (1999). A response to Truscott's 'What's wrong with oral grammar correction'. The Canadian Modern Language Review, 55 (4), 457-467.

Magnan, S. S (1983). Age and sensitivity to gender in French. Studies in Second Language Acquisition, 5, 194-212.

Major, R. C. (1988). Balancing form and function. Iral, 26 (2), 81-100.

Matsumura, S., & Hann, G. (2004). Computer anxiety and students’ preferred feedback methods in EFL writing. The Modern Language Journal, 88 (3), 403-415.

Maazoun Z, E (2018). In Response to Hendrickson’s (1978) Third Question: Which Learner Errors Should Be Corrected? TAYR Quarterly

McCretton, E., & Rider, N. (1993). Error gravity and error hierarchies. International Review of Applied Linguistics in Language Testing, 31 (3), 177-188.

Medgyes, P., & Reves, T. (1994). The non-native English speaking EFL/ESL teacher's self image: An international survey. System, 22 (3), 353-367.

Merrill, D. M. (1994). Instructional design theory. Englewood Cliffs, NJ: Educational Technology Publications.

Michaelides, N.N. 1990. Error Analysis: An Aid to teaching. English Teaching Forum Vol.XXXVIII/ 4: 28:30.

Mings, R. C. (1993). Changing perspectives on the utility of error correction in second language acquisition. Foreign Language Annals, 26 (2), 171-179.

Moxley, J. M. (1989). Responding to student writing:

Goals, methods, alternatives. Freshman English News, 17 (2), 3-11.

Muncie, J. (2000). Using written teacher feedback in EFL composition classes. ELT Journal, 54 (1), 47-53.

Myford, C. M., and Wolfe, E. M. (2009). Monitoring rater performance over time: A framework for detecting differencial accuracy and differencial scale use. Journal of Educational Measurement, 46, 371-389.

Myles, J. (2002). Second language writing and research: The writing process and error analysis in student texts. TESL-EJ, 6(2) A-

Nassaji, H. (2007a). Elicitation and reformulation and their relationship with learner repair in dyadic interaction. Language Learning, 57, 511-548.

Nassaji, H. (2009). Effects of recasts and elicitations in dyadic interaction and the role of feedback explicitness. Language Learning, 59 (2), 411-452.

Nassaji, H., & Swain, M. (2000). A Vygotskian perspective on corrective feedback in L2: The effect on random versus negotiated help on the learning of English articles. Language Awareness, 9 (1), 34-51.

Nation, I. S. P., and Newton J. (2001). Teaching EFL listening and speaking. New York: Routledge/Taylor and Francis Group.

Norrish, J. (1983). Language Learners and Their Errors. Macmillion Publishers.

Norrish, J. (1987). Language Learning and their Errors. London: Macmillan Publisher Ltd.

Nunan, D. (1991). Language teaching methodology: A textbook for teachers. New York: Prentice-Hall International.

Nunan, D., & Lamb, C. (1996). The self-directed teacher: Managing the learner process. Cambridge: Cambridge University Press.

Olasehinde, M.O. (2002). Error analysis and remedial pedagogy. In Babatunde S. T. and D.S. Adeyanju (eds.). Language, Meaning and Society. Ilorin : Itaytee Press and Publishing Co.

Olsson, M. (1972). Intelligibility: A Study of Errors and Their Importance. Research Bulletin 12, Dept of Educational Research, School of Education, Gothenburg, Sweden.

ORTEGA, L. (2007) Understanding Second Language Acquisition. London: Hodder Arnold (Understanding Language).

Oskoz, A. & Liskin-Gasparro, J.E. (2001). Corrective feedback, learner uptake, and teacher beliefs: A pilot study. In X. Bonch-Bruevich, W.J. Crawford, J. Hellermann, C.

Higgins, and H. Nguyen (Eds.), The past, present, and future of second language research (pp. 209- 228). Somerville, MA: Cascadilla Press

Paltridge, B. (2004). Academic writing. Language Teaching, 37, 87-107.

Piazza, L. G. (1979). Communicative Effects of Grammatical Errors Made by American Learners of French. Ph.D. Dissertation,University of Rochester.

Piazza, L. G. (1980). French tolerance for grammatical errors made by Americans. Modern Language Journal, 64, 425.

Pienemann, M. (1985). Learnability and syllabus construction. In K. Hyltenstam & M. Pienemann (Eds.), Modelling and assessing second language acquisition (pp. 23-76). Clevedon: Multilingual Matters.

Politzer, R. L. (1978). Error of English speakers of German as perceived and evaluated by German natives. Modern Language Jounnal, 62,253-261.

Raimes, A. (1991). Errors: Windows into the Mind. College ESL, 1 (2), 55-64.

Ranta, L., & Lyster, R. (2007). A cognitive approach to improving immersion students’ oral language abilities: The awareness-practice-feedback sequence. In R. DeKeyser (Ed.), Practice in a second language: Perspectives from applied linguistics and cognitive psychology (pp. 141-160). New York: Cambridge University Press.

Reichelt, M. (1999). Toward a more comprehensive view of L2 writing: foreign language writing in the U.S. Journal of Second Language Writing, 8 (2), 181-204.

Reigeluth, C. M. (1999). Instructional design theories and models: A new paradigm of instructional theory (Vol. 2). Mahwah, NJ: Lawrence Erlbaum Associates.

Richard, J.C. & Schmidt, R. (Eds.). (2002). Longman dictionary of language teaching and applied linguistics (3rd ed.). London: Longman.

Richards, J. C. (1984). A non-contrastive approach to error analysis, English Language Teaching. 25, pp 204-219.

Richards, J. C., & Sampson, G. P. (1974). The Study of Learner English. In J. C. Richards (Ed.), Error Analysis: Perspectives on Second Language Acquisition (pp. 3_18). London: Longman

Richards, J., & Farrell, T. (2005). Professional development for language teachers. Cambridge: Cambridge University Press.

Richards, J.C. (ed). (1973). Error Analysis. London: Longman.

Richards, Jack C. and Theodore S. Rodgers (1986). Approaches and methods in language teaching: A description and analysis. Cambridge: Cambridge University Press. 167pp.

Rifkin, B. (1995). Error gravity in learner’s spoken Russian: A preliminary study. Modern Language Journal, 79, 477-490.

Rifkin, B., & Roberts, F. (1995). Error gravity: A critical review of research design. Language Learning, 45 (3), 511­537.

Rivers, W and Temperley, M (1978). A practical Guide to the Teaching of English as a Foreign or Second Language. New York: Oxford University Press. ISBN0-19-502210-6

Robb, T., Ross, S., & Shorteed, I. (1986). Salience of feedback on error and its effect on EFL writing quality. TESOL Quarterly, 20 (1), 85-95.

Roberts, F., & Cimasko, T. (2008). Evaluating ESL: Making sense of university professors‘responses to second language writing. Journal of Second Language Writing, 17, 125-143

Rutherford, W. E. 1987. Second Language Grammar: Learning and Teaching, Longman.

Santos, T. (1988). Professors‘ reactions to the academic writing of nonnative-speaking students. TESOL Quarterly, 22 (1), 69-90.

Santos, T.(1987). Markedness theory and error evaluation : an experimental study. Applied Linguistics, 8,207-218.

Sari, M., & Mulyani, R. (2020). An Analysis of Students’ Errors in Writing Narrative Text. PROJECT (Professional Journal of English Education), 3(2), 190-196. https://doi.org/10.22460/project.v3i2.p190-196

Schachter, 1. (1974). An error in error analysis. Language Learning, 24, 205-214.

Schwartz, F., & White, K. (2000). Making sense of it all: Giving and getting online course feedback. In White, K. W., & Weight, B. H. (Eds.), The Online Teaching Guide: A handbook of attitudes, strategies, and techniques for virtual classroom (pp. 167-182).

Scovel, T. (1997). Psycholinguistics. Oxford: Oxford University Press

Selinker, L (1972). "Interlanguage". International Review of Applied Linguistics. 10: 209-241.

Seddik, M. (2023). Error Analysis in EFL Students’ Writing Skill. Journal of English Language Teaching and Applied Linguistics , 5 (4), 163­ 172. https://doi.Org/10.32996/jeltal.2023.5.4.16

Semke, H. D. (1984). Effects of the red pen. Foreign Language Annals, 17 (3), 195-202.

Sheen, Y. (2007). The effect of focused written corrective feedback and language aptitude on ELS learners’ acquisition of articles. TESOL Quarterly, 41 (2), 255-284.

Sheorey, R. (1986). Error perceptions of native-speaking and non-native-speaking teachers of ESL. ELT Journal, 40, Oxford University Press. 306-312.

Sheppard, K. (1992). Two feedback types: Do they make a difference? RELC Journal, 23 (1), 103-110.

Sheppard, K. (1992). Two feedback types: do they make a difference? RELC Journal, 23, 103-110.

Skinner, B. F. (1958). Teaching machines. Science, 128, 969-977.

Song, B., & Caruso, I. (1996). Do English and ESL faculty differ in evaluating the essays of native English-speaking and ESL students? Journal of Second Language Writing, 5(2), 163-182.

Sridhar, S. N. (1981). Contrastive analysis, error analysis, and interlanguage: Three phases of one goal. In J. Fisiak, (ed.), Contrastive linguistics and the language teacher (pp. 207-241). Oxford: Pergamon Press.

Stark, L. 2001. Analysing the interlanguage of ASL natives. Newark: University of Delaware.

Stern, H. H. (1992). Issues and options in language teaching. Oxford: Oxford University Press.

Svartvik. J. (1973). Errata; Papers in error artalysis. Lund, Sweden; Gleemp.

Swain, M. (1985). Communicative competence: Some roles of comprehensible input and comprehensible output in its development. In S. Gass & C. Madden (Eds.), Input in second language acquisition. Cambridge, MA: Newbury House.

Tomasello, M., & Herron, C. (1989). Feedback for language transfer errors: The garden path technique. Studies in Second Language Acquisition, 11, 385-395.

Tomiyana, M. (1980). Grammatical errors communication breakdown. TESOL Quarterly, 14 (1),71-79.

Troudi, S. & Mazandarani, O. (2025) (Eds.). Confronting challenges in English language teacher education: Global innovations and opportunities. Routledge.

Truscott, J. (1996). The case against grammar correction in L2 writing classes. Language Learning, 46 (2), 327-369.

Tuzi, F. (2004). The impact of e-feedback on the revisions of L2 writers in an academic writing course. Computers and Composition, 21, 217-235.

Tuzi, F. (2004). The impact of e-feedback on the revisions of L2 writers in an academic writing course. Computers and Composition, 21, 217-235.

Ur, P. (1996). A course in language teaching. Cambridge: Cambridge University Press.

Vahdatinejad, S. 2008. Students’ error analysis and attitude towards teacher feedback using a selected software : a case study. Unpublished Masters thesis. Univesiti Kabangsaan Malaysia, Bangi

Valdman, Albert . 1975. 'Learner Systems and Error Analysis' . In Jarvis, Gilbert A. (ed), Perspective : A New Freedom. Skokie IL: National Textbook Co.

Valette, R.1977: Modern language testing. Second edition. San Diego, CA: Harcourt Brace Jovanovich.

Vann, R. J., Meyer, D. E., & Lorenz, F. O. (1984). Error gravity: A study of faculty opinion of ESL errors. TESOL Quarterly, 18 (3), 427-440.

Varnosfadrani, A. D., & Basturkmen, H. (2009). The effectiveness of implicit and explicit error correction on learners’ performance. System, 37, 82-98.

Vaughan, C. (1991). Holistic assessment: what goes on in the rater's mind? In Hamp-Lyons, L. (ed.), Assessing Second Language Writing in Academic Contexts, 111-25. Norwood, NJ: Ablex

Wajnryb, R. (1992). Classroom observation tasks: Cambridge: Cambridge University

Walz, J. C. (1982). Error correction techniques for the foreign language classroom. Washington DC: Center for Applied Linguistics.

Welman, C. , Mitchell, B. & Kruger, F. (2005). Research Methodology. 3rd Edition. Cape Town: Oxford University Press

Wen, X. (1999). Error and corrective feedback. Journal of the Chinese Language Teachers Association, 34 (3), 1-22.

White, L. 1989. Universal grammar and second language acquisition. Amsterdam; John Benjamins.

White, L. 1991. "Adverb placement in second language acquisition: Some effects of positive and negative evidence in the classroom."Second Language Research 7: 133-161

Wilkson, D.,& Birmingham, P.(2003). Using Research Instruments: A guide for researchers. London and New York: Routledge Famler

Willcott, P. J. (1972): An analysis of the written English of native speakers of Arabic as found in American history final examinations at The University of Texas at Austin. Doctoral dissertation, The University of Texas at Austin.

Williams, J. (2005). Teaching writing in a second and foreign language classrooms. Boston, MA:McGraw-Hill.

Williams, J. (1981). The phenomenology of error. College Composition and Communication, 32, 152 - 168.

Wolfe, E.W., Kao, C.W., & Ranney, M. (1998). Cognitive differences in proficient and non proficient essay scorers. Written Communication, 15, 465-492.

Xie F and Jiang X (2007). Error analysis and the EFL classroom teaching. US-China Education Review,4(9), pp.10­14.

Yule, G. (2010). The study of language. (4th ed.).Cambridge: Cambridge

Zhang, S. (1995) Reexamining the Affective Advantage of Peer Feedback in the ESL Writing Class. Journal of Second Language Writing, 4(3): 209-222.

Ziv, N. D. (1984). The effect of teacher comments on the writing of four college freshmen. In R. Bach and L. S.

Bridwell (Eds.), New directions in composition research (pp. 362-280). New York: Guilford Press.

[...]

---

• Author: Emna Maazoun (Author)
• Publication Year: 2026
• Pages: 97
• Catalog Number: V1696694
• ISBN (PDF): 9783389184042
• ISBN (Book): 9783389184059
• Language: English
• Tags: error gravity in EFL, written error assessment, foreign language teaching, teacher experience, error analysis, EFL writing evaluation, Tunisian education