Probing Iranian EFL Students’ Interlanguage: Fossilized and Developmental Grammatical Errors in English
الموضوعات : نشریه مطالعات آموزش زبان انگلیسی
1 -
الکلمات المفتاحية: Developmental Errors, EFL, Fossilized errors, Interlanguage,
ملخص المقالة :
This study wanted to understand the fossilized and developmental grammatical errors in Persian EFL learners’ compositions at three levels of proficiency. The researchers divided the participants into lower-intermediate, upper-intermediate, and advanced levels. The researchers collected three compositions from each participant. After identifying the grammatical errors for each level, the researchers compared their frequency and mean differences across these levels to identify developmental and fossilized errors. Finally, using SPSS, five grammatical error types were identified, of which the most frequently made was the wrong selection of verb forms. Out of 5 error types, the researchers identified one as developmental (verb forms) and four as fossilized (tenses, prepositions, articles, and active and passive voices). Findings indicate that most students committed fossilized errors, showing the permanent retention of grammatical errors in their compositions. Accordingly, serious pedagogical considerations need to be given to the teaching of grammar to improve the writing competence of Iranian EFL learners.
Abbot, G. (1980). Towards a more Rigorous Error Analysis. International review of applied linguistics, 18, 121-134.
Allan, D. (1992). Oxford placement test 2. Oxford, UK: Oxford University Press.
Ancker, W. (2000). Errors and corrective feedback: Updated theory and classroom practice. English teaching forum, 38(4), 20-24.
Aslam, F. A. F., Shafique, N. S. N., & Nasrullah, S. N. S. (2023). A Study to Analyze the Fossilized Spelling Errors in Compositions of Second Language Learners at Intermediate Level. Journal of Policy Research, 9(3), 178-183.
Brown, H. Douglas (2000). Principles of language learning and teaching, San Francisco University, Pearson Longman.
Buzdar, H. Q. (2024). Identifying the grammatical errors in ESL students at graduate level: A Comparative analysis of multimedia and traditional modes of pedagogy. Pakistan Journal of Social Sciences, 44(1), 1-8.
Candling, R. B. (2001). Vocabulary and language teaching. New York: Longman Inc.
Chan, A.Y.W (2004). Syntactic transfer: Evidence from the Interlanguage of Hong Kong Chinese ESL learners, Modern Language Journal, 88, 56-74.
Do, P. T. V. (2023). Common written error analysis committed by EFL graders. A case study at a secondary school in Vietnam. Pegem Journal of Education and Instruction, 13(1), 1-12.
Dulay, H., Burt, M., & Krashen, S. (1982). Language Two. Oxford: Oxford University Press.
Elkilic, G. and Akca, C. (2008). Attitudes of the Students Studying at Kafkas University Private Primary EFL Classroom towards Storytelling and Motivation. Journal of Language and Linguistic Studies, 4(1), 1-22.
Ellis, R. (1985). Understanding second language acquisition. Oxford University Press.
Ellis, R. (1996). Second language acquisition research and language teaching. Oxford: Oxford University Press.
Gass, S. and Selinker, L. (2001). Second language acquisition: An Introductory Course Mahwah, NJ: LEA, chapter 3.2.
James, C. (2001). Errors in language learning and use: Exploring error analysis. Harlow, Essex: Addison Wesley Longman Limited.
Kharma, N. (1981). Analysis of errors committed by Arab university students in the use of the English definite and indefinite articles. Oral, 4, 333-345.
Larsen-Freeman, D. (2002). The grammar of choice. In E. Hinkel and S. Fotos (Eds.). New perspectives on grammar teaching in second language classrooms. Mahwah, New Jersey: Lawrence Erlbaum Associates.
Michaelides, N. N. (1990). Error Analysis: A teaching aid. English Teaching Forum Vol. XXVIII / 4: 28-30.
Mitchell, R. and F. Myles. (1998). Second language learning theories. Arnold.
Mitchell, R. and Myles, M. (2004). Second language learning theories. New York: Hodder Arnold.
Morelli, J. A. (2003). Ninth Graders’ Attitudes toward different approaches to Grammar Instruction. Unpublished Dissertation. The Graduate School of Education, Fordham University, New York.
Nasim, S. M., & Mujeeba, S. (2024). Arab EFL Students’ and Instructors’ Perceptions of Errors in Mechanics in Second Language Paragraph Writing. FWU Journal of Social Sciences, 18(1).
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., Nigeria.
Richards, J. C. Ed. (1974). Error analysis. Longman.
Selinker, L. (1972). Interlanguage. In Richards, J. C.
Selinker, L., & Lakshmanan, U. (1992). Language transfer and fossilization: The multiple effects principle. In S. Gass & L. Selinker (Eds.), Language transfer in language learning (pp. 197-216). Amsterdam: John Benjamins.
Sercombe, P. G. (2000). Learner language and the consideration of idiosyncracies by students of English as a second or foreign language in the context of Brunei Darussalam. In A.M. Noor et al. (eds.) Strategizing teaching and learning in the 21st century. Proceedings of the International Conference on Teaching and Learning. Faculty of Education: Universiti Kebangsaan Malaysia, Bangi.
Skehan, P. (1998). A Cognitive approach to language learning. Oxford University Press.
Stark, L. (2001). Analyzing the interlanguage of ASL natives. Newark: University of Delaware.
Vahdatinejad, S. (2008). Students’ error analysis and attitude towards teacher feedback using a selected software: a case study. Unpublished Master’s thesis. Universiti Kebangsaan Malaysia, Bangi.
Widdowson, H. G. (1990). Grammar and nonsense and learning. In H. G.
Willcot, P. (1978) Problems of definiteness in the Written English of Arabic speakers. E.L. T. Journal, 33(1), 67-83.
Volume 9, Issue 2 Spring, 2024 | |
| |
Accepted: January 2024 Published: June 2024 |
Research Article |
Probing Iranian EFL Students’ Interlanguage: Fossilized and Developmental Grammatical Errors in English Nafiseh Lashgari: Ph.D. Candidate, English Language Department, Shiraz Branch, Islamic Azad University, Shiraz, Iran. Email: lashgari.nasim@yahoo.com
|
ABSTRACT
This study wanted to understand the fossilized and developmental grammatical errors in Persian EFL learners’ compositions at three levels of proficiency. The researchers divided the participants into lower-intermediate, upper-intermediate, and advanced levels. The researchers collected three compositions from each participant. After identifying the grammatical errors for each level, the researchers compared their frequency and mean differences across these levels to identify developmental and fossilized errors. Finally, using SPSS, five grammatical error types were identified, of which the most frequently made was the wrong selection of verb forms. Out of 5 error types, the researchers identified one as developmental (verb forms) and four as fossilized (tenses, prepositions, articles, and active and passive voices). Findings indicate that most students committed fossilized errors, showing the permanent retention of grammatical errors in their compositions. Accordingly, serious pedagogical considerations need to be given to the teaching of grammar to improve the writing competence of Iranian EFL learners.
Keywords: Developmental Errors; EFL; Fossilized errors; Interlanguage
|
1. INTRODUCTION
Based on James (2001), “In the 1950s and 1960s, the main paradigm for improving second/foreign language learning and arranging its teaching was the Contrastive Analysis” (p. 4). The Contrastive Analysis Hypothesis implied that the main barrier to second language acquisition was the interference of the first language system with the second language system. It was arising from the shortcomings of the Contrastive Analysis Hypothesis to adequately account for second Language learners’ errors. Error Analysis was suggested by researchers as a new approach to investigate student errors. “Error Analysis assigned learner errors to different feasible sources, not just the effect and interference from learners’ first language” (Brown, 2000, p. 218). In Error Analysis, errors play an essential role in describing the learners’ language. Interlanguage (Selinker, 1972), is a system of rules which included both L1 and L2 elements. The developmental nature of interlanguage accounts for a learner’s ability to create and test hypotheses of the L2 grammar, a process that leads to internalizing the L2 rules. This process cannot improve without making errors. Based on the research of Ellis (2011), analyzing learners’ errors is not undesirable but rather can be a guide for the analyst to understand the nature of the learning process.
Errors need to be classified according to whether they have been gradually developed during the second language learning process, which is called developmental errors, or those errors which have been stable, which are called fossilized errors (Aslam et al., 2023). Developmental errors are those errors that reflect the learners’ competence at a particular state and illustrate some of the general characteristics of language acquisition (Do, 2023). At first, fossilization was introduced by Selinker (1972, p. 197), “the durability of the non-target-like structures in the interlanguage of non-native speakers”.
Many ESL instructors and learners alike regard the teaching and learning of sentence-level grammar as tiring or somewhat ineffectual, and consequently, grammar does not attract a lot of attention in many ESL classes. It is tempting to think that just as long as students have a large enough vocabulary, then the substantive words in their written sentences will carry enough so that any minor slips in the function words and grammatical relationships will not seriously take from the basic message (Buzdar, 2024). Unfortunately, this relaxed approach to the teaching and learning of grammar and sentence structures can have a different result when we consider the long-term effects on our ESL students' success. If the readers become instructors of college courses that our students take after they have finished ESL classes, then minor slips in grammar or sentence structures could bring disastrous for the ESL students. If those teachers do not completely understand what the students are writing, then the students' grades and G.P.A.'s could suffer, and the students' long-term success could be jeopardized.
Thus, it seemed reasonable to conduct a study to discover the types and the rate of grammatical errors and to determine the grammatical error types which disappear developmentally across different levels of language proficiency, and those which insist on the interlanguage of Persian EFL learners. This thesis addressed this issue.
2. REVIEW OF LITERATURE
In the 20th century, when Behaviorist Psychology and Structuralism were very popular, CA was very widely accepted in language teaching. CA was a kind of treatment for language teaching problems. After the popularity of this approach, a series of contrastive studies started to appear. They were usually pedagogical and had the aim of analyzing learners' errors. Brown (2000: p. 208) claimed that "the effect of the first language system on the second language system is the main barrier of the second language system. This effect and interference are the result of CA and it causes errors”.
Interfering first language knowledge in learning a second language is transfer theory. Transferring can be positive when a first language system identical to a second language system is transferred or it can be negative when a first language system different from the second language system is transferred. In the latter case, L1 induced errors occur (Nasim & Mujeeba, 2024). Chan (2004), in his article, represented some proofs of syntactic transfer from Chinese to English. It was based on data obtained from 710 Hong Kong Chinese ESL learners. They had different proficiency levels. The focus of the study was on five error types: (a) wrong placement of adverbs (b) lack of control of copula (c) failure to use the “there be” structure for expressing the existential function (d) inability to use the relative clause and (e) problem in verb transitivity. The result presented that many Chinese ESL learners in Hong Kong were attracted to think in Chinese first before they wrote in English. The surface structures of many of the interlanguage strings created by the participants were equal or very similar to the normative sentence structures of their first language (L1).
EA is a branch of applied linguistics. It has two functions which are theoretical and practical. The theoretical area of EA is methodological. It explains the learner's knowledge of the target language. It helps the researcher find out the nature of the psychological processes, and find the relation between the knowledge and the teaching the learner has been achieving. Gass and Selinker (2001) pretended that errors are systematic because they occur repeatedly also are not recognized by the learner. So, only the teacher or researcher could determine them. According to Ancker (2000), making mistakes or errors have to be considered as part of cognition and is a natural process of learning.
Sercombe (2000) explains that EA follows three purposes. First, to find out the language proficiency level that the learner has achieved. Second, to obtain some information about common difficulties in language learning, third, to find out the way that people learn a language. Candling (2001: p. 69) considers EA as “the monitoring and analysis of learner’s language. Candling adds that the L2 learner’s errors are essential for the understanding SLA’s processes”.
Olasehinde (2002) cited that errors are inevitable and an essential part of learning. Stark (2001: p. 19) in his study, explained that “the teachers need to view students’ errors positively. The need to view errors as a process of learning and should not look at them as the learners’ fracture to understand the rules and structures”. He supports the view that errors are normal, they are inevitable features of learning. He added that errors are essential to understand learning.
Vahdatinejad (2008) continues that error analyses are used to assign learners’ needs to be taught. He also makes a distinction between errors and simple mistakes. According to him, even native speakers produce lapses or simple mistakes, they can correct those mistakes. They call for on-the-spot correction rather than useful, which is needed for errors.
Three areas have to be described in teaching grammar: grammar as rules, form, and also resource. For many L2 learners, learning grammar means learning the rules and structures of grammar. They also think that learning grammar means achieving an intellectual knowledge of grammar. According to Widdowson (1990: p. 86), "grammar is not a constraining imposition. It is a liberating force. Grammar helps us to be free from a dependency on context. It also helps us to be free from a completely lexical categorization of reality." Given that many students and teachers like to see grammar as a set of limitations that they allowed or disallowed to use. In Larsen Freeman’s language, grammar is a linguistic straitjacket (2002: p. 103). “Grammar is something that makes you free not oppress you”.
According to Morelli (2003), students thought that they would have a better attitude towards grammar instruction in context. After having experienced the traditional grammar instruction, they perform slightly better. Elkilic and Akca (2008) reported an attraction of students studying English grammar at a private EFL classroom. In particular, however, over 50% of their subjects claimed to enjoy grammar very much. About 10% reported finding some difficulty in learning and recalling grammar. Selinker (1972: p. 41-47) assumes that fossilized linguistic structures are still present in the brain, they stored by a fossilization mechanism in an interlanguage.”
More recent interpretations on fossilization could be in two groups: the communicative or naturalistic group on the one hand and the form or instruction-led group on the other. Mitchell and Myles (1998, p. 13) call the two groups sociolinguistic and psycholinguistic description of fossilization.
Skehan’s interpretation of fossilization (1998, p. 61) is that “a rule-based system generates errors. It becomes an exemplar; although it can be changed later if the underlying rule-based system evolves sufficiently. Or, in other words, fossilization is learners’ misinterpretation rule that the accumulation of well-formed formulaic items has not eradicated it. This short review of the literature went over grammatical errors and error analysis. As stated earlier, almost no research has been done on fossilized grammatical errors and the English grammar development in Iran. Thus, this study will address the issue and occupy the gap.
3. METHODOLOGY
3.1. Design
This study investigated the kinds and frequencies of grammatical errors (prepositions, articles, active and passive voices, tenses, and verb forms) in the English writings of Persian EFL learners. Furthermore, the research investigated three levels of English proficiency errors to follow the students’ interlanguage development. Then, it identified the grammatical errors that disappeared developmentally across the three levels, as well as the type and the frequency of fossilized errors. Then, the design is descriptive in nature.
3.2. Participants
Seventy-five homogeneously selected learners of English as a foreign language participated in this study (N = 75). The participants were male (n= 44) and female (n=31) English as a Foreign Language (EFL) students from different levels. Their ages were between 20 and 30 years. They were all from Shiraz Azad University and their first language was Persian. They were sophomore, junior, and senior B.A (Bachelor of Arts) student of English (Translation and Teaching).
According to scores from a Standard English proficiency test called the Oxford Placement Test (Allan, 1992), the students were divided into three levels of lower-intermediate (LI), upper-intermediate (UI), and advanced (AD) groups. The feasible scores on this test are arranged from zero to one hundred.
Those students who scored below 35 were assigned to the LI (Lower Intermediate) group (24 students), the students whose scores were between 35 and 65 were assigned to the UI (Upper Intermediate) group (29 students), and the students who scored above 65 were assigned to the AD (Advanced) group (22 students).
3.3. Instruments and Materials
The Oxford Placement Test has been firmed to be a highly forcible initial-placement instrument. It is a reliable means of grading students at all levels from elementary upwards, with a consistent record of predictive validity regarding examination entry. The test includes two sections, listening, and grammar. It takes about an hour to complete. For this study, the grammar section of the Oxford Placement Test 2 was used. It consists of 100 multiple-choice items, in which the test takers were asked to read the stem with a blank, and then choose one of the three options. They were allowed 50 minutes to answer the questions. The researcher used an Oxford Placement Test for gaining a homogenous group.
The corpus of data for this study was the learners’ compositions during the first semester of the 2012-2013 academic years. For each participant, three compositions were collected, which formed a total sample of 225 compositions. Each composition was administered at the university by the teacher and the researcher. The participants were given three topics and were asked to write their compositions within a 50-minute class session. The selected topics were: (1) Features of a good teacher, (2) Your problems at this university, (3) Do you like your major? Why?
To get more reliable and uniform data, the participants were requested to write their compositions in about three paragraphs and to write on their own. They were asked to write without consulting their classmates, teacher, or researcher. Moreover, they were not told that their use of grammar would be studied.
3.4. Data Collection and Analysis Procedures
After the participants were placed in their groups, the compositions of the participants were selected. The compositions were analyzed in terms of grammatical errors, and the errors were assessed by an English professor and researcher to ensure that the errors were accurately spotted. They were asked to identify the grammatical errors.
Dulay, Burt, and Krashen (1982: p. 75) indicate that there are four major linguistic categories of errors:
1. Orthography (spelling)
2. Lexicon and semantics (vocabulary and meaning)
3. Syntax and morphology (grammar)
4. Discourse (style)
For this study, only students’ errors in the following grammatical areas were identified and categorized: tenses; prepositions; articles; verb forms; active and passive voices.
After classifying the types of grammatical errors, the most frequently occurring grammatical errors were identified. The grammatical error types for each proficiency level were then identified from the corpus. Next, the types and frequency of these errors were compared across the three levels to find out whether the rate of each grammatical error type significantly reduced across the three levels (developmental errors) or stayed the same (fossilized errors). To analyze the data, SPSS software will be used to analyze the data regarding the frequency and percentage and Chi-Square, one-way ANOVAs of each error type in each level.
4. RESULTS AND DISCUSSIONS
The first research question addressed the major grammatical errors of Persian EFL learners at three levels of language proficiency on using English grammars in their compositions. Five major types of grammatical errors were identified in the corpus, which totaled 445 errors. The error categories were: (1) verb forms, (2) prepositions, (3) tenses, (4) articles, (5) active and passive voices. The SPSS program was used to analyze the data regarding the frequency of each error type in each level. Table 5.1 shows the percentage and frequency of the occurrence of each error type in each level. As can be seen, the majority of students in the lower-intermediate had difficulty in the selection of the correct verb forms. In other words, the most frequently recurring error in the corpus was the wrong selection of verb forms.
Table 1
Frequency and percentage of each error type in each level
No | Error Type | Lower-Intermediate | Upper-Intermediate | Advanced | Total | ||||
|
| Frequency | Percentage | Frequency | Percentage | Frequency | Percentage | Frequency | percentage |
1 | Prepositions | 46 | 23.116 | 29 | 20 | 32 | 31.683 | 107 | 24.045 |
2 | Tenses | 37 | 18.593 | 48 | 33.103 | 18 | 17.822 | 103 | 23.146 |
3 | Articles | 22 | 11.055 | 23 | 15.862 | 12 | 11.881 | 57 | 12.809 |
4 | Verb forms | 88 | 44.221 | 42 | 28.966 | 31 | 30.693 | 161 | 36.18 |
5 | Active and Passive voices | 6 | 3.015 | 3 | 2.096 | 8 | 7.921 | 17 | 3.82 |
| Total | 199 | 100 | 145 | 100 | 101 | 100 | 445 | 100 |
To identify the differences between probabilities, Chi-square was run for each level.
Analysis of errors in lower-intermediate students
HO1: the probability of errors in lower-intermediate students is the same.
Table 2
Chi-square results of the lower-intermediate
| Observed N | Expected N | Residual |
Tenses | 37 | 39.8 | -2.8 |
Prepositions | 46 | 39.8 | 6.2 |
Articles | 22 | 39.8 | -17.8 |
Verb forms | 88 | 39.8 | 48.2 |
Active and passive voices | 6 | 39.8 | -33.8 |
Total | 199 |
|
|
Table3
Chi-square results of the lower-intermediate
Chi-square | 96.201 |
df | 4 |
Asymp. sig | .000 |
According to Table 2, the numbers of errors in the lower-intermediate are not the same (verb form > preposition > tense > article > active and passive voice). Moreover, based on the Chi-square results in Table 5.3, the probability of errors in lower-intermediate students was significant (p = .000), which is lower than .05. This means that the null hypothesis of the study was rejected.
Analysis of errors in Upper-Intermediate students
HO2: the probability of errors in upper intermediate students is the same.
Table 4
Chi-square table of the Upper-Intermediate
| Observed N | Expected N | Residual |
Tenses | 48 | 29.0 | 19.0 |
Prepositions | 29 | 29.0 | .0 |
Articles | 23 | 29.0 | -6.0 |
Verb forms | 42 | 29.0 | 13.0 |
Active and passive voice | 3 | 29.0 | -26.0 |
Total | 145 |
|
|
Table 5
Chi-square results of the Upper-Intermediate
Chi-square | 42.828 |
df | 4 |
Asymp. sig | .000 |
According to Table 4, the numbers of errors in upper-intermediate are not the same (tense > verb form > preposition > article > active and passive voice). Moreover, based on the Chi-square results in Table 5, the probability of errors in upper-intermediate students was significant (p = .000), which is lower than .05. This means that the null hypothesis of the study was rejected.
Analysis of errors in advanced students
HO3: the probability of errors in advanced students is the same.
Table 6
Chi-square table of the Advanced
| Observed N | Expected N | Residual |
Tenses | 18 | 20.2 | -2.2 |
Prepositions | 32 | 20.2 | 11.8 |
Articles | 12 | 20.2 | -8.2 |
Verb forms | 31 | 20.2 | 10.8 |
Active and passive voices | 8 | 20.2 | -12.2 |
Total | 101 |
|
|
Table 7
Chi-square results of the Advanced
Chi-square | 23.604 |
df | 4 |
Asymp. sig | .000 |
According to Table 6, the numbers of errors in the upper-intermediate are not the same (preposition > verb form > tense > article > active and passive voice). Moreover, based on the Chi-square results in Table 5.7, the probability of errors in upper-intermediate students was significant (p = .000), which is lower than .05. This means that the null hypothesis of the study was rejected.
Based on the rate of error occurrence across different proficiency levels which of these errors are developmental and which are fossilized?
The second research question investigated those grammatical errors which gradually (developmentally) disappeared from the learners’ writings in the higher levels of proficiency, and those grammatical errors which persisted across the three levels of proficiency (fossilized errors). To identify them, one-way ANOVAs were run for each error type, and the mean differences were identified. If the mean difference on each error type was significant across the three groups, and there was a decrease in means from the LI (Lower-Intermediate) to the UI (Upper-Intermediate) to the AD (Advanced) Level, the error was regarded as developmental. On the other hand, if there was no significant difference in the mean of each error type across the three groups, and the error remained nearly the same for all of them, the error was categorized as fossilized.
In this category, students didn’t use various types of tenses in their writing.
Some of these errors are listed below:
· I have a good job in the future (Simple present instead of future).
· My parents encouraged me and finally, I accept it (Simple present instead of simple past).
· When I was 12 years old, I don’t like English (Simple present instead of simple past).
HO4: the mean differences on each error type in tenses in different levels of LI, UI, and AD are the same.
Table 8
Descriptive Statistics for Tenses
| N | Mean | Std. Deviation | Std. Error | No. of Errors |
Lower Intermediate | 24 | 1.5417 | 1.74404 | .35600 | 37
|
Upper Intermediate | 29 | 1.6552 | 1.75816 | .32684 | 48 |
Advanced | 22 | .8182 | 1.22032 | .26017 | 18 |
Total | 75 | 1.3733 | 1.63388 | .18866 | 103 |
Table 9
ANOVA results of the Mean Scores for Tenses
Source of variation | Sum of squares | df | Mean Square | F | Sig. |
Between Groups | 9.764 | 2 | 4.882 | 1.872 | .161 |
Within Groups | 187.783 | 72 | 2.608 |
|
|
Total | 197.547 | 74 |
|
|
|
As can be seen in Table 8, the obtained means are nearly similar within the three groups. According to the ANOVA results in Table 5.9, the significant value of .161 at F = 1.872 was much greater than .05, which shows that the mean differences across the three groups were not significant. To put it differently, the participants in the three levels made the same rate of this error, which indicates no development, meaning the error was fossilized and the null hypothesis is accepted.
This type of error was made when the learners used an unnecessary preposition, omitted a required preposition, or used a wrong preposition after another preposition. L1 negative transfer, intralingual factors, and the students’ lack of attention may be among the sources of these errors, as in these examples:
· Stepping *outside of the stressful situation can help you to overcome it (instead of out of).
· We should try to make better our class *of these difficulties (instead of out of).
HO5: the mean differences on each error type in prepositions in different levels of LI, UI, and AD are the same.
Table 10
Descriptive Statistics for Prepositions
| N | Mean | Std. Deviation | Std. Error
| No. of Errors
|
Lower-Intermediate | 24 | 1.9167 | 1.74248 | .35568 | 46 |
Upper-Intermediate | 29 | 1.0000 | 1.43925 | .26726 | 29 |
Advanced | 22 | 1.4545 | 1.73829 | .37060 | 32 |
Total | 75 | 1.4267 | 1.56361 | .19094 | 107 |
Table 11
ANOVA results of the Mean Scores for Prepositions
Source of variation | Sum of squares | df | Mean square | F | Sig. |
Between Groups | 11.059 | 2 | 5.529 | 2.081 | .132 |
Within Groups | 191.288 | 72 | 2.657 |
|
|
Total | 202.347 | 74 |
|
|
|
As can be seen in Table 10, the obtained means are nearly similar within the three groups. According to the ANOVA results in Table 5.11, the significant value of .132 at F = 2.081 was much greater than .05, which shows that the mean differences across the three groups were not significant. To put it differently, the participants in the three levels made the same rate of this error, which indicates no development, meaning the error was fossilized and the null hypothesis is accepted.
Another problematic area for the Persian learners learning English as a foreign language is the use of articles.
English has definite and indefinite articles. The use of articles depends on the noun premodified by the article. A definite article is used with specific references. The indefinite article with generic reference is used with a singular countable noun when the reference represents the whole class.
Articles are not used similarly in Persian as in English. For example, “I am a student”, the English indefinite article “a” is used before the student. However, this is not the case in Persian.
Errors of Articles contain Omission of “the”, Addition of “the”, Omission of a/an, Addition of a/an, Misuse of articles.
Thus, the Persian learners may produce so many errors in articles. Some examples are as follows:
· My major is the best in---- world. (omission of “the” definite article)
· I like to study at the home. (addition of the definite article “the”)
· Within a half an hour, my friend studied more than 3 chapters (addition of indefinite article “a”).
HO6: the mean differences on each error type in articles in different levels of LI, UI, and AD are the same.
Table 12
Descriptive Statistics for Articles
| N | Mean | Std. Deviation | Std. Error | No. of Errors |
Lower-Intermediate | 24 | .9167 | 1.13890 | .23248 | 22 |
Upper-Intermediate | 29 | .7931 | .97758 | .18153 | 23 |
Advanced | 22 | .5455 | .91168 | .19437 | 11 |
Total | 75 | .7600 | 1.01129 | .11677 | 56 |
Table 13
ANOVA results of the Mean Scores for Articles
Source of variation | Sum of squares | df | Mean square | F | Sig. |
Between Groups | 1.634 | 2 | .817 | .794 | .456 |
Within Groups | 74.046 | 72 | 1.028 |
|
|
Total | 75.680 | 74 |
|
|
|
As is shown in Table 12, there was a decrease in the mean of the errors from LI toward the UI and AD groups, but the ANOVA results in Table 13 indicate no significant mean differences across the three levels (p=.456, F = .794). Therefore, it seems that the error had remained fossilized. To put it differently, the participants in the three levels made the same rate of this error, which indicates no development, meaning the error was fossilized and the null hypothesis is accepted.
This category is the lowest number of errors among the linguistic categories employed in this study. The students avoided using active and passive voices in their compositions. Both English and Persian have active and passive with different constructions. English passive construction uses auxiliaries and word order change. Persian passive construction is a matter of vowel change without changing the order of the words in a sentence. Such differences may cause errors that are due to interference.
These errors are divided into sub-categories as stated below:
Passive auxiliary Be omission, Passive with intransitive verb Be addition, Preposition confusion.
Below are some samples of such errors:
· I ---- shocked by the results (passive auxiliary Be omission).
· They were decided to give up University (passive with intransitive verb /Be addition).
· The progress is caused from studying (preposition confusion).
HO7: the mean differences on each error type in active and passive voices in different levels of LI, UI, and AD are the same.
Table 14
Descriptive Statistics for Active and Passive voices
| N | Mean | Std. Deviation | Std. Error | No. of Errors |
Lower-Intermediate | 24 | .2500 | .44233 | .09029 | 6 |
Upper-Intermediate | 29 | .1034 | .30993 | .05755 | 3 |
Advanced | 22 | .3636 | .78954 | .16833 | 7 |
Total | 75 | .2267 | .53457 | .06173 | 16 |
Table 15
ANOVA results of the Mean Scores for Active and Passive voices
Source of variation | Sum of squares | df | Mean square | F | Sig. |
Between Groups | .866 | 2 | .433 | 1.537 | .222 |
Within Groups | 20.281 | 72 | .282 |
|
|
Total | 21.147 | 74 |
|
|
|
As can be seen in Table 14, the obtained means are nearly similar within the three groups. According to the ANOVA results in Table 15, the significant value of .222 at F = 1.537 was much greater than .05, which shows that the mean differences across the three groups were not significant. To put it differently, the participants in the three levels made the same rate of this error, which indicates no development, meaning the error was fossilized and the null hypothesis is accepted.
These errors are divided into sub-categories as stated below:
The omission of the verb “Be”, Addition of the verb “Be”, Misuse of the verb “Be”, Omission of other verbs, Misuse of other verbs.
The most erroneous area in this category is the use of the verb “Be”. Both omission and addition of verb “Be” are found as the following example:
· When a student---- studying heavily he hurts himself. (Be omission).
· The teacher says it is essential that students are given special attention. (Misuse of be).
Some other types of errors are bellowed:
· He know English very well (instead of knows).
· She have good behavior (instead of has).
· She should ask a lot of questions (instead of ask).
HO8: the mean differences on each error type in verb forms in different levels of LI, UI, and AD are the same.
Table 16
Descriptive Statistics for Verb forms
| N | Mean | Std. Deviation | Std. Error | No. of Errors |
Lower-Intermediate | 24 | 3.6667 | 2.29682 | .46884 | 88 |
Upper-Intermediate | 29 | 1.4483 | 1.66017 | .30829 | 42 |
Advanced | 22 | 1.4091 | 1.70878 | .36431 | 31 |
Total | 75 | 2.1467 | 2.14787 | .24801 | 161 |
Table 17
ANOVA results of the Mean Scores for Verb forms
Source of variation | Sum of squares | df | Mean square | F | Sig. |
Between Groups | 81.563 | 2 | 40.781 | 11.301 | .000 |
Within Groups | 259.824 | 72 | 3.609 |
|
|
Total | 341.387 | 74 |
|
|
|
According to Table 16, the number of this error type decreased from 88 errors in the LI to 42 errors in the UI, and 31 errors in the AD group. Since there was a decrease in making the error by the UI and then the AD group, it seems that the error was developmentally disappearing. Moreover, based on the ANOVA results in Table 17, the mean difference among the three groups was significant (F = 11.301, p = .000). This means that this error type was developmental and the null hypothesis of the study was rejected.
The study had two research questions:
(1) What are the major types of English grammatical errors made in compositions written by Persian EFL learners majoring in English (translation and teaching) at different levels of sophomore, junior and senior?
(2) Based on the rate of error occurrence across different proficiency levels, which of these errors are fossilized and which are developmental?
The first research question of this study addressed the overall rates and types of English grammatical errors across three levels of English proficiency. A total number of 445 grammatical errors were found in the corpus, dispersed among 5 error categories. The results indicate that the largest proportion of errors in each category was made in the incorrect verb forms (36.18%). Prepositional errors ranked next to verb forms errors (24.045%), tenses errors came after that (23.146%). Then articles errors (12.809%) and active and passive voices errors were found to be of less frequency (3.82%).
Michaelides (1990: p. 29) understood that “students created eight domains of errors according to seriousness and frequency of occurrence: incorrect word order, incorrect use of tenses, wrong use of prepositions, wrong use of articles, the omission of indirect object pronouns, etc.”
Many similar studies were conducted to investigate certain types of errors. Willcott (1978: p. 80) “investigated Arab students’ domain of errors make in the area of definiteness.”
Abbot (1980: p. 127) “investigated the errors generated by Arab students in the area of restrictive relative clauses. The study revealed that 57% of the attempted relative clauses were erroneous. The types of errors committed were: repetition of relative pronoun subject, repetition of relative pronoun object, wrong selection of relative pronouns and using redundant prepositions.” Likewise, Kharma (1981: p. 339) “examined Arab students’ errors in the use of the English definite articles.” Kharma’s results revealed that the majority of errors occurred in the use of “the” instead of “no articles” and vice versa. Some of these errors were attributed to Arabic language interference, others to wrong learning strategies or overgeneralization, etc.
The second question of the study addressed the grammatical errors which gradually disappeared from students’ writings (developmental errors) as well as those that persisted across the three levels of proficiency (fossilized errors). Of 5 error types, only one of them was put in the developmental category (verb forms).
The results indicated that 4 out of 5 error types were placed in the fossilized category; these errors persisted in the students’ writings, even though they were at the AD level. The most frequent fossilized errors in the corpus (across the three levels) were (1) prepositional errors, (2) tenses errors, (3) article errors, (4) active and passive voices.
5. CONCLUSION
Encouraged by previous studies that emphasize the difficulties that L2 learners find in the use of English grammar, this study focused on the use of 5 grammatical errors. The three compositions written by the students provide good evidence that students make a lot of errors in this area. Because of this, it was felt necessary to diagnose these grammatical errors and identify the most common ones so that English teachers would be familiar with different types of errors that their students make and be better able to tackle them.
The findings of this study call for more research in this area. First, future research can address grammatical errors not just in production, but also in recognition tasks. This can be done through other test methods, such as fill-in-the-blank exercises, multiple-choice questions, or translation exercises from L1 to L2. Second, due to the multifunctional aspects of English grammars, the conclusions made in this research need to be validated further by conducting more research on Iranian EFL students’ problems with other functions of grammars, It may also be helpful to replicate the study with a greater number of participants and a larger corpus of student writings. Finally, the present study calls for longitudinal case studies to carefully scrutinize and follow the grammatical development of Persian learners and identify the developmental and fossilized errors that emerge.
REFERENCES
Abbot, G. (1980). Towards a more Rigorous Error Analysis. International review of applied linguistics, 18, 121-134.
Allan, D. (1992). Oxford placement test 2. Oxford, UK: Oxford University Press.
Ancker, W. (2000). Errors and corrective feedback: Updated theory and classroom practice. English teaching forum, 38(4), 20-24.
Aslam, F. A. F., Shafique, N. S. N., & Nasrullah, S. N. S. (2023). A Study to Analyze the Fossilized Spelling Errors in Compositions of Second Language Learners at Intermediate Level. Journal of Policy Research, 9(3), 178-183.
Brown, H. Douglas (2000). Principles of language learning and teaching, San Francisco University, Pearson Longman.
Buzdar, H. Q. (2024). Identifying the grammatical errors in ESL students at graduate level: A Comparative analysis of multimedia and traditional modes of pedagogy. Pakistan Journal of Social Sciences, 44(1), 1-8.
Candling, R. B. (2001). Vocabulary and language teaching. New York: Longman Inc.
Chan, A.Y.W (2004). Syntactic transfer: Evidence from the Interlanguage of Hong Kong Chinese ESL learners, Modern Language Journal, 88, 56-74.
Do, P. T. V. (2023). Common written error analysis committed by EFL graders. A case study at a secondary school in Vietnam. Pegem Journal of Education and Instruction, 13(1), 1-12.
Dulay, H., Burt, M., & Krashen, S. (1982). Language Two. Oxford: Oxford University Press.
Elkilic, G. and Akca, C. (2008). Attitudes of the Students Studying at Kafkas University Private Primary EFL Classroom towards Storytelling and Motivation. Journal of Language and Linguistic Studies, 4(1), 1-22.
Ellis, R. (1985). Understanding second language acquisition. Oxford University Press.
Ellis, R. (1996). Second language acquisition research and language teaching. Oxford: Oxford University Press.
Gass, S. and Selinker, L. (2001). Second language acquisition: An Introductory Course Mahwah, NJ: LEA, chapter 3.2.
James, C. (2001). Errors in language learning and use: Exploring error analysis. Harlow, Essex: Addison Wesley Longman Limited.
Kharma, N. (1981). Analysis of errors committed by Arab university students in the use of the English definite and indefinite articles. Oral, 4, 333-345.
Larsen-Freeman, D. (2002). The grammar of choice. In E. Hinkel and S. Fotos (Eds.). New perspectives on grammar teaching in second language classrooms. Mahwah, New Jersey: Lawrence Erlbaum Associates.
Michaelides, N. N. (1990). Error Analysis: A teaching aid. English Teaching Forum Vol. XXVIII / 4: 28-30.
Mitchell, R. and F. Myles. (1998). Second language learning theories. Arnold.
Mitchell, R. and Myles, M. (2004). Second language learning theories. New York: Hodder Arnold.
Morelli, J. A. (2003). Ninth Graders’ Attitudes toward different approaches to Grammar Instruction. Unpublished Dissertation. The Graduate School of Education, Fordham University, New York.
Nasim, S. M., & Mujeeba, S. (2024). Arab EFL Students’ and Instructors’ Perceptions of Errors in Mechanics in Second Language Paragraph Writing. FWU Journal of Social Sciences, 18(1).
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., Nigeria.
Richards, J. C. Ed. (1974). Error analysis. Longman.
Selinker, L. (1972). Interlanguage. In Richards, J. C.
Selinker, L., & Lakshmanan, U. (1992). Language transfer and fossilization: The multiple effects principle. In S. Gass & L. Selinker (Eds.), Language transfer in language learning (pp. 197-216). Amsterdam: John Benjamins.
Sercombe, P. G. (2000). Learner language and the consideration of idiosyncracies by students of English as a second or foreign language in the context of Brunei Darussalam. In A.M. Noor et al. (eds.) Strategizing teaching and learning in the 21st century. Proceedings of the International Conference on Teaching and Learning. Faculty of Education: Universiti Kebangsaan Malaysia, Bangi.
Skehan, P. (1998). A Cognitive approach to language learning. Oxford University Press.
Stark, L. (2001). Analyzing the interlanguage of ASL natives. Newark: University of Delaware.
Vahdatinejad, S. (2008). Students’ error analysis and attitude towards teacher feedback using a selected software: a case study. Unpublished Master’s thesis. Universiti Kebangsaan Malaysia, Bangi.
Widdowson, H. G. (1990). Grammar and nonsense and learning. In H. G.
Willcot, P. (1978) Problems of definiteness in the Written English of Arabic speakers. E.L. T. Journal, 33(1), 67-83.
