Çiftci, H., & Ozcan, M. (2021). A contrastive analysis of Traditional Grammar Translation Method and Communicative Language Teaching in teaching English grammar and vocabulary. International Online Journal of Education and Teaching (IOJET), 8(2). 709-729. Annotated link of the article. Received : 16.08.2020 Revised version received : 24.09.2020 Accepted : 25.09.2020 A CONTRASTIVE ANALYSIS OF TRADITIONAL GRAMMAR TEACHING AND COMMUNICATIVE LANGUAGE TEACHING IN TEACHING ENGLISH GRAMMAR AND VOCABULARY Research article Harun Çiftci Isparta University of Applied Sciences harunciftci@isparta.edu.tr Mehmet Özcan Mehmet Akif Ersoy University mozcan@mehmetakif.edu.tr Harun ÇİFTCİ is an instructor at Isparta University of Applied Sciences, School of Foreign languages. Currently, he is pursuing his PhD studies at Muğla Sıtkı Koçman University, Graduate School of Educational Sciences, Department of English Language Teaching. Mehmet Özcan received his doctorate degree at Middle East Technical University, Institute of Education Sciences, Department of English Language Teaching. Currently, he works as a professor at Mehmet Akif Ersoy University, Faculty of Education, Department of English Language Teaching. Copyright © 2014 by International Online Journal of Education and Teaching (IOJET). ISSN: 2148-225X. Material published and so copyrighted may not be published elsewhere without written permission of IOJET. mailto:harunciftci@isparta.edu.tr mailto:mozcan@mehmetakif.edu.tr https://orcid.org/0000-0003-1444-641X http://orcid.org/xxxx Ciftci & Ozcan 710 A CONTRASTIVE ANALYSIS OF TRADITIONAL GRAMMAR TEACHING AND COMMUNICATIVE LANGUAGE TEACHING IN TEACHING ENGLISH GRAMMAR AND VOCABULARY Harun Çiftci harunciftci@isparta.edu.tr Mehmet Özcan mozcan@mehmetakif.edu.tr Abstract This study aimed to compare whether Traditional Grammar Translation Method (henceforward TGTM) or Communicative Language Teaching (henceforward CLT) was more effective in teaching English language grammar and vocabulary. The participants of this study were sixty-eight 9th grade students attending to a high school. Mixed methods research design was grounded comprising a pre-test and post-test quasi-experimental design. The results of the pre and post-tests, quizzes and the oral productions of the participants were analyzed based on error analysis. It was observed that the students, who were taught English grammar via TGTM made fewer errors than those who were taught via CLT in both their written and oral productions. Additionally, vocabulary production was higher for the students who were taught based on TGTM than those in CLT. As a result, the quantitative and qualitative findings of the study suggested that TGTM was more practicable and productive than CLT in teaching and learning English grammar and vocabulary. Keywords: TGTM, CLT, Grammar, Vocabulary, Errors 1. Introduction The question how to teach grammar in language teaching has been under discussion for decades, especially, whether to teach it explicitly or implicitly. Debate over this issue ascended in 1970s, especially with the emergence of communicative based approaches (Ellis, 2006). During this continuum two core ideas came into prominence: deductive and inductive way of teaching grammar. Deductive teaching is a way which centers upon explicit grammar teaching; on the other hand, inductive teaching is a way of implicit grammar teaching without foregrounding the language structures. These two views have been by large associated with the methodologies Grammar Translation Method (henceforward GTM), in which deductive way of grammar teaching is implemented and CLT, in which inductive way of grammar teaching is implemented in teaching contexts. In a GTM classroom, the students are expected to translate the literary texts in which structures are included from simple to complex. Through these translation activities, the grammatical structures and vocabulary of the target language (henceforward TL) are expected to be conceptualized and learned by the students. A general criticism directed to GTM is that it puts a barrier which prevents language learners’ interaction in TL (Conti, 2016). However, it contributed a lot by maintaining its validity in language teaching and learning practices, especially in EFL classrooms (Xia, 2014). It puts fewer burdens on teachers as it is easy to mailto:harunciftci@isparta.edu.tr mailto:mozcan@mehmetakif.edu.tr International Online Journal of Education and Teaching (IOJET) 2021, 8(2), 709-729. 711 shape classroom objectives and evaluation processes when compared with other methods and approaches (Brown, 2000). When considered from the language learners’ perspective, since the interaction takes place in the mother tongue (henceforward MT) of the language learners, they feel less stressed during learning TL (Chang, 2011). Starting form 1960s, there has been a shift from a structural view to a communicative view in language teaching and learning contexts. With the emergence of natural approaches, CLT was recognized as the methodology which promotes communication and interactional skills to convey meaning in the classroom environment (Richards & Rodgers, 2002). The main goal of CLT is to teach TL via meaningful activities by avoiding explicit grammar teaching. The learners are expected to acquire grammatical structures in an implicit way and the vocabulary is learned through functional and situational exercises. The ostensible criticism directed to CLT has been the violation of grammar teaching which is regarded as an efficient way for maintaining intelligible communication in the classroom environment (Thompson, 1996). However, CLT still keeps its validity by promoting more authentic and interactive learning tasks with comprehensible input and learners’ language output (Koosha & Yakhabi, 2013) and helping teachers to motivate their students to focus on fluency rather than form (Belchamber, 2015). The pending question whether grammar should be taught or not is still a controversial issue for many language learning contexts and within this debate CLT and GTM have been eminently investigated to answer this question. Howbeit, as opposed to the current view of GTM in teaching English, it was traditionally taught by translating it into native languages or vice a versa before the introduction of English grammar in a systematic way (Widodo, 2006). Thence, the importance and originality of this study is that it holds the traditional method TGTM in teaching English grammar and vocabulary and compares it with a reputed approach CLT. Furthermore, no other studies have provided an in-depth analysis on the comparison of TGTM and CLT through a qualitative and quantitative perspective in teaching English grammar and vocabulary in a Turkish EFL context. 2. Literature review In language teaching history, there have been many theories and methodologies about how to teach and learn languages over centuries. There have also been lots of rapid shifts and changes in these theories and methodologies due to the various reasons. These reasons vary with respect to the influences and instances in different periods. Prior to emergence of the language studies scientifically, particularly before the 15th century, trade and religion were at the forefront of these factors (Howatt, 1984). Although English gained official importance in the last quarter of this century, the medium of education and aristocratic society was still Latin as it was the only language that had grammar (Howatt, 1984). The 16th and 17th centuries witnessed a tremendous refugee exodus from France, Italy, and Spain to England. After this immigration, teaching English became a profession with the contribution of these refugees who started to teach English to non-English speaking Europeans. Howbeit, English could not have superseded French or Latin, despite the efforts performed by the refugee teachers until 18th century (Howatt, 1984). After the introduction of GTM in Germany in the 1780s, learning a foreign language gained importance. This situation also paved the way for methodological approaches and methods which resulted as a reaction to classic methods in teaching and learning languages. In research methodology, GTM and CLT have been the major methodologies that are exoterically investigated by the researchers based on their effectiveness or ineffectiveness. Uysal and Bardakci (2014) investigated Turkish primary-level English language teachers’ Ciftci & Ozcan 712 beliefs and practice patterns of teaching grammar, vocabulary, and the reasons behind these practice patterns. Their study displayed that teachers overall preferred GTM rather than CLT since time constraints, crowded classes and especially for low motivated students. Many of the ELT teachers have difficulty in coping with teaching English grammar as they pursue CLT in their teaching contexts and have difficulties when subject is grammar in their classrooms (Sato &Kleinsasser, 1999). Chang (2011) argued in his study that GTM was proved to be a suitable and successful teaching approach in English grammar teaching since it met the students’ needs. Gorsuch (1998) introduced a Japanese English language grammar teaching method named Yakudoku in which grammar teaching is performed by merely translating Japanese into English, or English into Japanese. The results of Gorsuch’s study revealed that Yakudoku had advantageous results in teaching English grammar and vocabulary since it provided better results in the university entrance exams. As opposed the studies which argue that GTM is more effective in teaching and learning English grammar and vocabulary, there are also studies which argue that CLT is more effective than GTM in classroom practices. In a comparative study on the principles of GTM and CLT, it was argued that CLT was far more effective than GTM in students’ handling with communicative problems (Abbas & Ali, 2014). Conti (2016) pointed out that CLT was more effective in preparing the students for real-life situations. GTM and CLT were additionally compared with respect to their feasibility in the classroom environment by administering questionnaires as was in Jewad and Verma’s (2014) study. The study revealed that CLT was preferred and regarded more useful than GTM in the classroom environment. 2.1. What is TGTM? Up to now, no previous study has distinguished TGTM and GTM based on teaching grammar and vocabulary. Concerning the differences between TGTM and GTM, it will be beneficial to approach this differentiation from two different perspectives: their view on grammar teaching and their argument about translation. Firstly, in TGTM, there is not an emphasis on explicit grammar teaching; hence, the grammatical topics are not arranged on a linear sequence i.e. from simple to complex. Additionally, the grammatical items in the teaching materials are not explicitly exemplified with the exercises for the purpose of reinforcing these items. However, explicit grammar teaching is the core premise of language teaching process and the grammar topics are arranged from simple to complex in an organized sequence and they are explicitly exemplified with the exercises for the purpose of reinforcing these items. (Larsen-Freeman, 1986). For vocabulary teaching and learning, short vocabulary lists and practices are provided to reinforce this vocabulary in a GTM classroom (Richards & Rodgers, 2002). Secondly, translation is the basic medium of instruction and the content is translated word- by-word from TL to MT of learners in TGTM. The whole teaching hour is attributed to this premise. The teacher provides only the correct translated form of the sentences without focusing on the structure of the sentence. Additionally, the learners are the active participants during these teaching hours. Thus, the learners are expected to learn the grammar implicitly. However, translation is regarded as an activity in a GTM classroom (Larsen-Freeman, 1986). It is used to consolidate the grammatical items and the vocabulary. Grammar and vocabulary teaching are one of the most widely investigated concern in the history of English language teaching and learning as a foreign language. Within this framework, two methodologies have come into prominence; GTM which calls for an explicit way of grammar teaching and CLT which calls for an implicit way of grammar teaching. However, English language was only a vernacular language and had to wait for three centuries to be taught as a language (Baugh & Cable, 2002). After the proclamation of English as the International Online Journal of Education and Teaching (IOJET) 2021, 8(2), 709-729. 713 official language by King Henry V and the introduction of printing press in Britain by William Caxton, the first texts were started to be prepared to teach English based on the basic premise of TGTM (Howatt, 1984). The use of TGTM was prevalent in teaching and learning English during this continuum. By considering this historical background of English grammar and vocabulary teaching, recent research lacks studies that investigate and compare these two methodologies: TGTM and CLT. Based on the findings of researcher’s MA study, this study attempted to answer the following research questions: 1. Is TGTM or CLT more effective in English grammar teaching? 2. Are there any differences between TGTM and CLT in the teaching and learning of vocabulary, especially modifiers? 3. Are there any structural differences between written and spoken language produced by the participants who were taught based on TGTM or CLT? 3. Method 3.1. Design Mixed methods and a quasi-experimental research design were designated to collect qualitative and quantitative data. Since the classes are pre-determined by other variables such as the national exam TEOG and school conditions, this study was designed as nonrandomized TG and CG pre-test/post-test design containing two intact groups (Leedy & Ormrod, 2015). 3.2. Participants The participants in this study were two intact 9th grade classes consisting of 34 students in each class attending to a high school in Turkey. The ages of the participants ranged from 14 to 15. Throughout the study, the classes were randomly assigned as the ‘Treatment’ and ‘Contrast’ groups. The gender balance in the groups (TG & CG) was nearly the same. Table 1. Gender of the students participating the study Gender Groups Male Female Total Number of participants Number of participants Number of participants % % % TG 10 29.5 24 70.5 34 50 CG 12 35.30 22 64.70 34 50 Table 1 demonstrates that the number of female students is 22 (64.70 %) in CG and 24 (70.5 %) in TG and the number of male students is 12 (35.30 %) in CG and 10 (29.5 %) in TG. As an important parameter used in identifying the proficiency levels of the participants was the mean values of English exams in Transition Examination from Primary to Secondary Education known as TEOG in Turkey. TEOG was implemented in each mid-term to place the 8th grade students into the schools of Secondary Education Schools in Turkey. In this placement, 30% of the mean of the Grade Point Average of the 6th, 7th, and 8th grades and 70% of TEOG scores obtained at the end of the 8th grade were considered by Ministry of National Education (MEB, 2013). Ciftci & Ozcan 714 Table 2. The results of English tests in TEOG The mean values of English exam in TEOG Groups First TEOG Second TEOG Total m % M % m % TG 18.02 90 19.44 97 18.7 93 .5 CG 18.41 92 18.79 96 18.6 94 There is not much difference in the mean values of each group as the entry scores range for this school context the same for every student (Table 2). The first TEOG results indicate that CG had a slight better success 18.41 (92%) than TG 18.02 (90%). However, the results of the second TEOG were exactly the opposite as TG had the score of 19.44 (97%) which is higher than those of CG 18.79 (96%). 3.2.1. Instructional materials Grammar and vocabulary teaching via five literary texts which were chosen from different literary genres to ensure the reliability and validity of the data was implemented to the groups by the researcher. Table 3. The list of the literary texts Weeks Literary Texts Literary Genres 1-2 The Haughty Princess Folk Tale 3-4 The Zipper Descriptive Narration 5-6 King Arthur Historical Narration 7-8 First Day at School Poem 9-10 All My Sons (Act 1) Play These literary texts illustrated in Table 3 were taught TG based on TGTM; on the other hand, they were also taught CG based on CLT. 3.3. Data collection procedures and analysis Data were collected by means of written and spoken productions of the participants. Written data were collected from pre-test and post-test, and the quizzes which were applied after the implementation of each text. The questions directed to the participants in these tests were mainly focused on revealing their knowledge about the structures and vocabulary identified in these texts. These questions were also checked by two other ELT teachers working at the same school. Spoken data were collected by recording the oral productions of the participants by using a high frequency response sound recorder during the implementation of the identified texts. These sound files were then transcribed into word format (for transcriber reliability, 20% of the data was also transcribed by two different ELT teachers at the same school) and the identified errors in both these transcriptions and the quiz papers were codified into different International Online Journal of Education and Teaching (IOJET) 2021, 8(2), 709-729. 715 colors for the purpose of analyzing TG and CG on the production level. To capture their real language competence, participants were not allowed to use dictionaries or any other supplementary resources while building written or spoken corpora in this corpus-based research. In the qualitative and quantitative data analysis, the errors made by the participants were analyzed which provided insights about how much they learned and progressed in English grammar and vocabulary. With these aims, the errors produced by the participants were classified into their source category to have a better understanding for their progress (Corder, 1967). The ungrammatical parts of the sentences were typed in italics and in cases of the existence of more than one error in the same sentence, the error type which affects the intelligibility of that sentence was dealt as the error type. The erroneous productions of the participants were also checked by two native speakers of English from USA and UK. According to researcher’s approach to ungrammaticality, the sentences which were identified as ungrammatical were determined in comparison with Standard English as well. Table 4. Categorization of the errors A-INTERFERENCE A. 1. L1 interference with the TL 1.1. Syntactic errors 1.1.1. Word order 1.2. Insertion and deletion 1.3. Case marking 1.4. Third-person singular 1.5. Semantic level 1.6. The (Non) use of articles A.2. TL interference with TL B- DEVELOPMENTAL ERRORS B.1. The misapplication of rules or features B.2. Overgeneralization (Ozcan, 2012, p. 2311-2322) Errors made by the learners were basically divided into two main groups as interference errors (e.g. We hide and seek played.) and developmental errors (e.g. If I was a king, buy a big and beautiful castle.) in the analysis of the data (Table 3). In qualitative analysis, data collected from the participants were analyzed with an explanatory perspective. The structural features and erroneous parts of the sentences produced by the participants were the primary focus for the interpretation. In this identification and interpretation process, errors were taken into Ciftci & Ozcan 716 consideration as performance errors and competence errors (Ozcan, 2012). Hence, the classification of errors with respect to their source category relies on their underlying reasons rather than their grammatical category. The identified errors were grouped, as previously stated, and interpreted depending on the research questions. In some cases, clarification questions were also asked to the participants for consolidating the category of the error. In quantitative analysis, the number of the erroneous patterns and sentences, pre-test, post-test and quizzes were analyzed whether there was a statistically significant difference in the errors produced by the groups so as to compare the productivity levels based on sentence grammaticality. The retrieval process of the vocabulary produced by the groups was also compared accordingly. 4. Results The error types that displayed the highest frequencies in the pre/post-tests and quizzes of the participants are dealt according to their categories and possible sources causing these errors were discussed by illustrating samples from each group. The numbers of the errors and their percentages were also analyzed according to each of these identified categories. In the comparison of the error types of TG and CG, it was observed that interference errors comprised one-third of the total errors. Both groups made totally 739 (20.8%) interference errors and these errors were mainly clustered around the (non) use of articles with a high frequency of 328 (9.2%) errors in TG and 411 (11.6 %) in CG. Table 5. Interference errors in the written data Interference Errors TG CG (1) I hate the school but I love the Mr. Çiftci because he is a wonderful person. (Okuldan nefret ederim ancak Bay Çiftci-y-i seviyorum çünkü o harika bir insan.) (3) He clean the room. (Odayı temiz-ler) (5) Zipper is wonderful a invention. (Fermuar harika bir icattır. (7) I’m excitied (excited) and scaried. (9) Cemil is hung up Deliha. (Cemil, Deliha’ya asılır.) (2) (The) Bagger hasn’t got money. (4) I and my mum goes to school. (Ben ve annem okula gider-iz.) (6) Oğuz is rude a character. (Oğuz kötü bir karakterdir.) (8) He is huntring (hunting). (10) I saw my main school friends. (Anaokulu arkadaşlarımı gördüm.) Interference errors significantly emerged in the insertion of ‘the’ in previous nouns which are used as the object of the sentence and the non-use of definite article the in the written data. When the researcher asked clarification questions to the participants in each group, the usage of definite article ‘the’ before the nouns that are the objects of a transitive verb as in (1), they stated that they used definite article the as the accusative case marker (Table 5). This kind of an error, prevalent in the quiz papers of TG, may be an attempt to find an equivalent of the accusative case marker (the Mr. Çiftci- Bay Çiftci-y -i) which exists in their MT. It was also observed that the omission of definite article ‘the’, which should modify the nominative case of the word, was frequently observed as was exemplified in sample (2). The main reason for this omission may be the fact that Turkish language system does not contain a linguistic element which accounts for definite article the explicitly. Turkish language system does not International Online Journal of Education and Teaching (IOJET) 2021, 8(2), 709-729. 717 call for definite article the when the word to be modified by the definitive encoder in ‘’subject position’’ and the nominative case of a noun refers to a known entity. This type of errors was mostly observed in the quiz papers of CG, especially with the nouns that were discussed before in the texts produced by the participants. The errors which stemmed from the failure of adding or omitting the third-person singular agreement suffix -s to the verb in The Simple Present Tense as was in sample (3) were more frequent in CG than in TG. Turkish language system does not necessitate third-person singular inflectional morpheme -s attached to the verb used with third-person singular subject pronouns while other subjects impose an extra morpheme to the stem (O temiz-ler clean-aorist, Ben temiz-ler-im, Sen temiz- ler-sin, Onlar temiz-ler-ler). As opposed to the previous example, the agglutination of third-person singular inflectional morpheme -s to the plural nominative cases as in (4) ‘Kızlar sever-ler’ and ‘Erkekler sevmez’-ler may be activated due to the participants’ compulsion to attach third-person agreement marker -s (Ozcan, 2012). The errors stemming from syntactic error category were mainly clustered around phrase structures in TL. Especially, the use of modifiers and articles according to word order in their MT influenced their TL syntactic order. As for the samples (5) and (6), the participants preferred to insert the indefinite article a/an between the modifiers and the nouns that were modified. The positioning of articles and adjectives in a noun phrase in Turkish language system (kötü bir karekter and harika bir icat), urged the participants to impose the same word order (rude a character and wonderful a invention) in TL. These syntactic errors in the productions of TG are the result of word by word translations of L1 sentences into L2 or vice versa. Insertion and deletion errors mainly stemmed from adding ‘i’ to an adjective ending in simple past tense marker ‘-ed’ for regular verbs and omission and insertion of a letter from the word. Adding -i to an adjective ending in ‘-ed’ as in the sample (7) type of error produced by the participants. The underlying reason for the emergence of this erroneous inflection of ‘-ied’ to an adjective ending in ‘-ed’ (excited-excitied) may stem from an analogy due to the suffixation process in the regular forms of the verbs in Simple Past Tenses which end in /y/ preceded by another consonant. Since the Past Simple Tense and past participles of irregular verbs have many different forms that do not adhere to a distinct or predictable pattern contrary to the case in Turkish language, the participants might have difficulty in fulfilling the conjugation rules. Insertion and deletion errors might also stem from a derivational process processed in the minds of participants as was in the sample (8). The participant derived the verb huntre by attaching derivational suffix -re to the verb hunt. Additionally, he also obeyed the rules of affixation process of the inflectional morpheme -ing by omitting -e at the end of the verb he derived. Semantic errors made by the participants may stem from various reasons; however, the analysis of the data in this study showed that these errors mainly stemmed from the participants’ transferring the semantic content of the words and phrases from their MT into TL (Ozcan, 2012). The words ‘hang up and main’ were used correctly in the syntactic order of TL; however, they were used in the same notion as was in their MT. In the first example (9), the participant tried to emphasize that ‘Cemil philanders with Deliha.’ The example in (10) reflects the same procedure on the semantic level ‘I saw my friends who were at the kindergarten.’ According to the analysis of the data in interference error category, CG had a higher frequency of errors than TG. Especially, the number of the errors in the (non) use of articles third-person singular agreement and semantic level error categories were higher in CG than those of TG. Howbeit, in the word order, insertion and deletion error categories TG made more errors than Ciftci & Ozcan 718 CG. As for the interference errors stemming from target langue interference with target language, the groups did not make any errors in this category. Developmental errors constituted two third of the total errors analyzed in the written data. In this category CG had a high frequency with 1568 errors which comprised (80%) of their total errors, and TG made 1230 errors which comprised (78.5%) of their total errors. The misapplication of rules or features was the first type of developmental errors in which the groups had higher frequencies than other developmental error types. Table 6. Developmental errors in written data Developmental Errors TG CG (11) I am exciting. (13) Where went to the princess? (15) I did call my mother. (17) King is rules his country. (12) I am scaring. (14) I happied first day at school. (16) I readed the book. (18) Man isn’t know swimming. The errors stemming from the misapplication of rules or features were divided into two sub- categories as the word-based and sentence-based errors (Table 6). In the first sub-category, the participants misapplied the rules or features on the word- basis. Word-basis errors especially stemmed from the usages of adjectives ending in -ed and -ing. The participants were mostly confused the use of adjectives ending in ‘- ed’ and ‘-ing’. In samples (11) and (12), the process in the usage of the adjectives ‘scaring’ and ‘exciting’, the participants might perceive these adjectives as verbs by agglutinating them with Present Continuous Tense marker -ing. In the second category, the participants misapplied the rules or features on the sentence-basis. The errors stemming from the sentence-based ones mainly stemmed from the misapplication of rules or features in interrogative sentences and conditionals. The underlying reason why TG made more errors than CG in interrogative sentences as exemplified in the sample (13) might stem from the nonexistence of auxiliary verbs as a separate morpheme in interrogative sentences structure of Turkish language. Furthermore, a word-to-word translation of MT (13) ‘Prenses nereye gitti?’ reflected L1 interference of the interrogative form in the emergence of these structures. Overgeneralization errors constituted the second error category in developmental errors. Overgeneralization errors stem from the application of a form into similar constructions where it should not be applied (Corder, 1967). The application of Simple Past Tense conjugation marker ‘ed’ in irregular verbs was prevalent in the productions of CG. As exemplified in sample (16) the lack of irregular verb structure in Turkish language system might be the reason why the participants preferred to apply ‘-ed’ form as a Past Tense marker for irregular verb forms. In addition to this usage, some adjectives were also inflected as a verb by implementing the same process as was in sample (14). This situation also revealed that the participants were not able to comprehend the use of stative verbs with predicative adjectives which are used after copular verbs. The use of is as a Simple Present Tense marker shows similar process in the productions of the participants. Although the third-person singular agreement suffix -s was attached to the verbs in (17), the participants’ usage of copular verb is as the Simple Present Tense marker revealed that it was conceptualized as an extra morpheme which should be used with the main International Online Journal of Education and Teaching (IOJET) 2021, 8(2), 709-729. 719 verb. However, in the sample (18), the participant did not attach -s to the verb know since the sentence was in negative form. Thus, the participant obeyed the rule of negativity in Simple Present Tense. The samples illustrated in overgeneralization error category may also be a way of reduction of the learning load and simplification of the structures which are formidable for the participants. Apart from the analysis of the interference and developmental errors made by the participants in the written data, the vocabulary retrieved and produced by the groups was also analyzed in a descriptive way. The analysis revealed that EG retrieved and produced more vocabulary than CG. This retrieval and production process of the targeted vocabulary were presented quantitatively in the quantitative data analysis. The error types that displayed highest frequencies in the oral productions of the participants were also dealt according to the error categories as identified previously and possible sources causing these errors were discussed, respectively. The total number of the errors in the spoken data was 120 and 49 (4.8%) of these errors were interference errors and 71 (59.2%) of them were developmental errors. Table 7. Interference errors in spoken data TG CG (20) He is a from Irish. (22) She might make a good husband. (24) One day, King Arthur rides his horse in the woods and he sees a beautiful castle. (26) Character feeling is sad, scare and exciting. (28) He is likes himself. (21) Person feel like outsider (23) Merlin and The Lady of the lake has powers! (25) This person is scaring and exciting. (27) Merlin is helps Arthur. (29) It means when bell is ringing school is finish. In the sample (20) in Table 7, the use of indefinite article a preceding the preposition from may stem from the direct translation of the sentence from their MT (O bir İrlandalıdır). As illustrated in the sample (22), the participant used a totally direct translation of a phrase in the slang language of his MT. The intended meaning for the phrase ‘make a good husband’ was ‘to get married to a good and wealthy person’. The phrases used in the notion of their MT revealed that some idiomatic and vernacular phrases especially those related to the customs of marriage in Turkish culture were handled as the direct translation into TL. The semantic process in (21) reflected not only a direct translation of the word in his MT, but also, a derivation for the word within its semantic scope was provided by the participant. The preposition outside was translated as foreign (yaban) and by deriving it with the derivational suffix -er, he produced the foreigner (yaban-cı) in TL. Spoken data revealed that the groups made errors on the semantic level in which the words and phrases transferred from the notion of their MT into TL. The errors stemming from the failure of third-person singular -s usage was common in the oral productions of CG when compared with TG. Especially, the use of a singular verb in the sentences containing two nouns as the nominative cases called for the participants a singular verb form in Simple Present Tense as in the sample (24) in Table 7. In sample (23), the syntactic order in the predicative ‘rides his horse in the woods’ was translated to MT as ‘giderken atıyla ormanda’ instead of ‘atıyla ormanda giderken’ in MT. This revealed that the syntactic order of TL also affects their syntactic order of MT in translation activities of TG. The analysis of Ciftci & Ozcan 720 the data related to the interference errors of the participants revealed that CG made more errors than those of TG in their oral productions. Developmental errors in spoken data were also grouped as the misapplication of rules or features and overgeneralization. The groups made more misapplication of rules or features errors when compared with overgeneralization errors. In developmental error category, CG made 50 errors which comprised (56.1%) of their total errors while TG made 21 errors which comprised (67.8%) of their total errors. In the misapplication of rules or features errors in the spoken data, the errors stemming from the confusion of the adjectives ending in -ed and -ing were frequent in the oral productions of TG. These adjectives were mostly predicative adjectives that are used after linking verbs. Strong evidence suggests the usages of the adjectives scare, scaring and exciting (25 and 26) in English language do not follow the same process as in the Turkish language as in sample. As was in the analysis of the written data, the groups made fewer errors in the overgeneralization error category than in the misapplication of rules or features error category. The use of -is as the marker of Simple Present Tense were the most common error type in the oral productions of CG in the overgeneralization error category. The samples (27) and (28) in Table 7 revealed that the copular -is was used as a means of Simple Present Tense marker in the productions of the participants. In the sample (28), although the third-person singular agreement suffix -s were attached to the verbs, the participants’ pertinacity in the usage of -is as the Simple Present Tense marker reflects a strong evidence that it was learned as an auxiliary used with all the main verbs in Simple Present Tense. The vocabulary retrieval process of the participants was analyzed based on oral productions. In the recognition process of the vocabulary which was learned in the literary texts were also expected to be used in the participants’ oral productions. Totally 21595 words were analyzed in the spoken data. The analysis revealed that TG did not make any errors, whereas CG made only one error in this process. (30) Because she don’t like anybody and beggar is her lost change (chance). In sample (30), the activation process of change (/tʃeɪndʒ/) instead of chance (/tʃɑːns/) the initial sound /tʃ/ may have played a significant role on the selection and production of this word in the mental lexicon of the participant. The participants’ mean scores and standard deviations of pre-test were presented in Table 8 Table 8. The results of pre-test Groups The mean values of pre-test 95% Confidence Interval of the Difference n m SD T df Sig. (2-tailed) Lower Upper TG 34 21,4412 13,4111 8.288 66 ,472 16.182 26.870 CG 34 21,5882 15,4875 With respect to the results in the pre-test, there is not a statistically significant difference (t (66) =8.288; p=, 472, d= 0.01) between the mean proficiency score of EG (M=21.44; SD=13.41) and CG (M=21.58; SD=15.48). According to the sections of vocabulary, sentence completion and paragraph writing in general, the two groups were considered to be equal with respect to the results. International Online Journal of Education and Teaching (IOJET) 2021, 8(2), 709-729. 721 The number of the vocabulary produced in the written data of the groups was analyzed for the purpose of comparing the progress on vocabulary learning in this study. Table 9. The distribution of the vocabulary produced in the written data Quiz Number Number of the Vocabulary Percentages TG CG EG % CG % Pre-test 3760 3950 6.2 6.4 1 3991 4279 6.5 6.9 2 3585 3328 5.9 5.4 3 5028 4334 8.2 7 4 5248 4641 8.6 7.5 5 5361 4006 8.8 6.5 Post-test 5388 4285 8.9 7 TOTAL 32361 28823 53.1 46.9 The results of the pre and post-test and quiz papers were compared based on vocabulary production of the groups (Table 9). As for the results, TG produced higher number of vocabulary usage with 32361 words (53.1%) when compared to CG with 28823 words (46.9%). The purpose of implementing quizzes after each literary text is to collect data which would be used in comparing the grammatical progress of the groups. The results of these quizzes of the groups were compared as to whether TG or CG had made any significant progress in learning English grammar. Table 10. The results of the quizzes Groups The mean values of pre-test 95% Confidence Interval of the Difference n m SD T df Sig. (2-tailed) Lower Upper TG 34 59,1529 12,38807 ,828 66 ,411 -4,64345 11,18463 CG 34 55,8824 19,42171 As presented in Table 10, the results of the quizzes revealed that although TG had a higher mean value, there was not a significant difference (t (66) = ,828; p<.05) between TG (M=59,1529; SD=12,38807) and CG (M=55,8824;SD=19,42171). Ciftci & Ozcan 722 Table 11. The results of post-test Groups The mean values of pre-test 95% Confidence Interval of the Difference n m SD T df Sig. (2 tailed) Lower Upper TG 34 74.7059 11.4111 19.33 66 ,000 70.724 78.687 CG 34 51.3235 15.4875 When the results of post-tests were compared, there was a statistically significant difference (t (66) =19.323; p<.05) between the mean proficiency score of TG (M=74.70; SD=11.41) and CG (M=51.32; SD=15.4875). The total number of the vocabulary produced by the groups was 32134 (100%) as shown in Table 12. TG produced 18860 (58.7%) words, whereas CG produced 13274 (48.3%) words. Table 12. The distribution of the vocabulary produced in spoken data Groups Number of the Vocabulary Percentages (%) TG 18860 CG 13274 TOTAL 32134 58.7 48.3 100 In the oral productions of the groups, only the words produced by the participants were taken into consideration (Table 12). TG produced 18860 words (% 58.7) while CG produced 13274 (% 48.3) in their oral productions. 5. Discussion Grammar and vocabulary teaching have always been one of the major controversial concerns in different approaches and methodologies. The historic form of GTM in the history of English language teaching and learning, which is identified as TGTM in our study, has never been studied with this respect. Hence, we not only focused on the written productions of the learners but also their oral productions and addressed the research questions to find answers; whether TGTM or CLT is more effective in English grammar teaching, whether there are any differences between TGTM and CLT in the learning of vocabulary specifically modifiers, and whether there are any structural differences between written and spoken language produced by the participants who were taught on the basis of TGTM and CLT? Error analysis is regarded as a very important premise in foreign language learning contexts since it is the first step to identify a successful language learning process. The errors made by the participants in their written and oral productions were used as the main pivots in responding the research questions. A language corpus with 93.318 words was built by the researcher to have a better insight into participants’ language use involving their correct, incorrect, and unused structures. All errors made by the groups were regarded as competency errors although there may be other factors as the grounds of errors which cannot be differentiated. These data International Online Journal of Education and Teaching (IOJET) 2021, 8(2), 709-729. 723 then tabulated with respect to the error categories. In the following section, the results of the study are discussed by taking research questions into consideration, respectively. 5.1. Is TGTM or CLT more effective in English grammar teaching? Based on the findings in the written data, CG had a higher frequency of errors than TG in interference error category. In this category, the number of the errors in the (non) use of articles, third-person singular agreement and semantic level error categories were higher in CG than those of TG. In the (non) use of articles, TG better processed the (non) use of articles than CG since they transferred the knowledge of their MT article system into the article system of TL via translation activities (Ionin et al., 2004). The case is also in line with the third person agreement errors made by the groups in their productions. Howbeit, in the word order and insertion and deletion error categories, TG made more errors than CG. The possible reason why TG made more errors in these categories stem from the fact that the word order of TL influenced their productions. Contrary to this study, Chang (2011) argues that translation impedes the negative transfers such as word of choice, word order and sentence structures from MT into TL or vice versa. This study revealed contrary results with this respect. As for the interference errors stemming from target language interference with target language, the groups did not make any errors in this category. This may stem from the fact that the participants are not proficient enough to discriminate and process the rules of TL which would help them to interpret the structural similarities and differences between their MT and TL. The groups made more developmental errors than interference errors stemming from the fact that the participants did not have enough practice and knowledge in TL. This case contradicts with the findings of Taylor’s (1975) study which states that elementary language learners made more interference errors than developmental errors. In the misapplication of rules or features error category CG made more errors than TG while in overgeneralization error category, TG made more errors than CG in developmental errors of the written data. Based on the findings in the oral productions of the participants, the interference errors revealed that CG had more errors than those of TG in their oral productions. In the (non) use of articles error category, the groups made more errors than the other error categories in interference errors. In word order, case marking and semantic error categories, the groups made fewer errors. Among these error categories, TG made more errors in word order and semantic error categories than those of CG. However, in the (non) use of articles and third person singular error categories CG made more errors than those of TG. These findings are line with the findings of Master (1997) who states in his study that since CLT is a meaning focused approach which puts primary focus on meaning and fluency of the communication, the (non) use of articles might be regarded as the components which do not hinder communication. With respect to insertion and deletion and TL language interference with TL error categories, the groups did not make any errors. As for the analysis of the data in developmental error category in the oral productions of the participants, the results were similar with those of the results of written data; in other words, the groups made more errors in developmental error category than interference error category. To be more specific, in the misapplication of rules or features error category, the groups made more errors when compared with overgeneralization error category. In these sub-categories of developmental error category, CG made more errors than those of TG. Translation, which is a prerequisite of TGTM, was an advantage for the participants in TG in producing more grammatically correct utterances since they relied on the comparison of their MT and TL (Guerra, 2014). This case also helped them to participate more actively during the classes (Pan & Pan 2012). According to the results of pre-test, the groups did not show significant difference at the beginning of the implementation process of the study. However, CG was more successful Ciftci & Ozcan 724 than TG by achieving a slight difference in the mean value. When considered the quiz results of the study, TG was more successful than CG in all the quizzes except for the first quiz. Lastly, the post-tests revealed that TG made a significant progress when compared to CG. In conclusion, the error types were classified, and the number of the errors made by the groups was identified with their underlying reasons to answer the first research question. Although there were some exceptional cases, TG made fewer errors and obtained better results in the quizzes and post-test in the implementation process of the study. Returning to the first research question posed at the beginning of the study, it is now possible to state that TGTM was more effective than CLT in the learning English grammar. 5.2. Are there any differences between TGTM and CLT in the teaching and learning of vocabulary, especially modifiers? The second question was addressed to answer whether there were any differences between TGTM and CLT in the teaching and learning of vocabulary, specifically modifiers. When compared vocabulary production between the written data and spoken data of the participants, it was observed that there was a significant difference. The participants in both groups produced more vocabulary in their written productions than in their spoken productions. The participants had the opportunity to control and review their written productions as opposed to their oral productions (Krashen & Terrell, 1983). Therefore, this opportunity provided them to use the intended vocabulary and sentences more carefully and fruitfully in their written productions. Within this framework, TG outnumbered CG in the vocabulary production in the written and spoken data. As previously stated in the section 4.2.2, the oral productions of the groups revealed a significant difference in the number of the vocabulary. TG produced more words when compared with CG since the participants in TG had a chance to monitor their productions via translation. Conversely, the reason why CG produced fewer words might have resulted from the inhibition which the participants’ anxiety levels triggered during the speaking activities. This case contradicts with the findings of Akhtar’s (2014) study in which it is argued that the CLT helps the students to learn the vocabulary and understand the text when compared to traditional methodologies. Additionally, in his study Al-wossabi (2016) mentions that in a CLT classroom, only a few competent students participate in speaking activities while the others are not volunteer to join these activities. In the recognition of the vocabulary in the literary texts, TG was more successful than CG. Especially in the sections which participants were expected to complete the sentences, it was observed that TG was more productive in completing the sentences with the intended vocabulary and expressions. Furthermore, in paragraph writing section TG produced more syntactically and semantically correct sentences via meaningful usages of these targeted vocabulary. This might result from the fact that translations activities during the implementation of the study, particularly in these sections, had a positive impact on the recognition of the intended vocabulary (Dagilienė, 2012). Another significant point in vocabulary learning process in this study was that the groups confused some words stemming from the similarity of their phonological and orthographical forms. In this comparison TG made fewer recognition errors than CG. Strong evidence in this process suggests that translation activities, in other words verbatim translations of the words, had a positive impact on the discrimination of phonologically and orthographically similar words on behalf of TG. As a result, the number of the vocabulary retained by the participants in TG was higher than those of CG. International Online Journal of Education and Teaching (IOJET) 2021, 8(2), 709-729. 725 5.3. Are there any structural differences between written and spoken language produced by the participants who were taught on the basis of TGTM or CLT? The most striking difference between the written and spoken languages of the groups was that although there were many erroneous sentences, the participants produced more sentences in their written language than in their spoken language. This case might have stemmed from the fact that they were not under pressure in answering the questions as the case was in the classroom- talking hours. Firstly, what increases this pressure on them in the classroom-talking hours might be the anxiety level resulting from the notion or fear of making errors in these classroom-talking hours (Öztürk and Gürbüz, 2014). Secondly, the peer-mocking might hinder them to be productive in their spoken language. And lastly, their previous language learning in English did not put much emphasis on oral activities since they were prepared for their central exam TEOG, which was based on only testing the learners’ text-based performances. These factors were similar to those reported by Liu (2007) who stated that the anxiety level, fear of making mistakes and being laughed at and lack of practice and vocabulary are the main reasons which impede productivity in the spoken languages of learners. Considering these factors, they were more productive in their written language than spoken language. When the written and spoken languages produced by the groups are compared, CG produced more structurally erroneous sentences in both their written and spoken languages than those of TG. CG produced text-independent sentences, while TG produced text-dependent sentences based on their translation activities. Due to these text-dependent sentences, TG had an advantage to analyse and compare the structure of their MT with that of TL. This analysis might provide a positive attitude not only in their written productions but also in their oral productions as was in Şevik’s (2007) study in which it was stated that the use of MT provides an opportunity for EFL learners at different levels in comprehending the subject and using the positive effects of MT in this way. Although Duff (1994) argues that translation is a text- dependent activity and it does not involve oral interaction as a communicative activity, Ross (2000) argues that translation used as the basic skill of TGTM in this study is the most important social skill which enriches communication and transferring the meaning. It was also found that the groups produced structurally correct sentences in their oral productions than those of in their written productions. This may stem from the fact that the participants felt free from anxiety in their written productions when compared to their oral productions. Conversely, in the speaking hours, the participants were aware of their wrong utterances would cause more burdens on them; therefore, they participated in the speaking activities when they felt ready for their correct answers. In some phrases, such as containing indefinite articles and modifiers, TG produced more erroneous structures when compared with those of CG. In these types of sentences produced by TG, a direct impact of the interference of MT might have played a determinant impact in their erroneous productions. This case was not analyzed in the oral productions of TG compared to CG. Another important difference between the written and spoken language of the groups was that they produced more syntactically correct sentences in their spoken language. This case also supports the hypothesis that the groups were more careful and meticulous before the production stage of their utterances than before the production stage of their written productions. In the comparison of the groups in terms of productivity, both in the written language and in the oral language, TG has produced more structurally correct sentences than CG. The Ciftci & Ozcan 726 underlying reason for this result stems from the fact that TG could make in-depth analysis in the translation process of the TL and this resulted in making fewer errors than CG. In conclusion, since the fundamental aim is to teach TL in accordance with the Standard English in the education system in Turkey, the results of this study revealed that the participants who were taught on the basis TGTM were more successful than the participants taught in CLT both in their written and oral productions. 6. Conclusion The use of MT has been regarded as the main obstacle which prevents the flow of interaction and communication in the teaching and learning of English as a second or foreign language. Therefore, the proponents who argued that fluency should be the basic premises also stressed that English grammar should be taught an implicit way (Wong & Barrea-Marlys, 2012). Since 1970s, Turkey has undergone a shift from structuralist view to a communicative view in English language teaching and learning (Kırkgoz, 2007), and thence curriculum developers and syllabus designers have focused on designing curriculum, syllabus and course books on the basis of CLT. However, from the very beginning of its application in Turkish EFL classrooms, its effectiveness has always aroused questions in its practitioners’ minds (Ozsevık, 2010). Within the framework of this study, it was aimed to reveal whether TGTM or CLT was more effective in teaching English grammar and vocabulary. As a result, TGTM was more effective than CLT in teaching English grammar and vocabulary. These results were not only valid in the written productions of the participants but also in their oral productions. Additionally, the participants who were taught based on TGTM were more productive in learning and using the vocabulary in their written and oral productions. This study gives some practical and theoretical implications for stakeholders, curricula designers as well as learners as regards to the implementation of TGTM in EFL secondary and high school settings. With the aim of eliminating the blur whether to teach grammar and vocabulary explicitly or implicitly, this study proposes that TGTM should be inserted not only in EFL curriculum and syllabus design but also in the exam system of EFL contexts. The quantitative and qualitative data which were provided here was limited to 68 Turkish EFL students attending to 9th grade students attending to a public high School. Thus, the qualitative and quantitative data discussed represent the whole body of Turkish learners to the extent the number of the participants this study covers. Hence, this contrastive analysis and their results can be generalized accordingly. A longitudinal study can be conducted at different levels of EFL learners to generalize the findings. Additionally, this longitudinal study might give a better understanding in the progression of the EFL learners’ oral productions when it is implemented for a longer period. 7. Conflict of Interest The authors declare that there is no conflict of interest. 8. Ethics Committee Approval The authors confirm that the study does not need ethics committee approval since this study is derived from an MA thesis which was also presented and published in National Thesis Centre. International Online Journal of Education and Teaching (IOJET) 2021, 8(2), 709-729. 727 References Abbas, S., & Ali, W. (2014). Comparison between Grammar & Translation Method & Communicative Language Teaching. International Journal of Advanced Research, 2(6), 124–128. Retrieved from https://www.researchgate.net/publication/323346253_COMPARISON_BETWEEN_G RAMMAR_TRANSLATION_METHOD_COMMUNICATIVE_LANGUAGE_TEAC HING Akhtar, R., N. (2014). An experimental study using Communicative Language Teaching Method (CLTM) at higher secondary level to enhance students’ reading comprehension skills. NICE Research Journal, 7, 1–24. Retrieved form https://www.researchgate.net/publication/275042781_AN_EXPERIMENTAL_STUDY _USING_COMMUNICATIVE_LANGUAGE_TEACHING_METHOD_CLTM_AT_ HIGHER_SECONDARY_LEVEL_TO_ENHANCE_STUDENTS'_READING_COM PREHENSION_SKILLS. Al-wossabi, S. (2016). Speaking in the target language: Issues and considerations. Journal of Language Teaching and Research, 7(5), 886–893. http://dx.doi.org/10.17507/jltr.0705.08. Baugh, A., C., & Cable, T. (2002). A history of the English language (4th ed.). London, England: Routledge. Belchamber, R. (2015). The advantages of Communicative Language Teaching. The Internet TESL Journal, 13(2), 1–7. Retrieved from http://iteslj.org/Articles/BelchamberCLT. Brown, D. (2000). Principles of language learning and teaching (4th ed.). New York, USA: Pearson Education. Chang, S., C. (2011). A contrastive study of Grammar Translation Method and Communicative Approach in teaching English grammar. English Language Teaching, 4(2), 13–24. https://doi:10.5539/elt.v4n2p13. Conti, G. (2016). Grammar Translation and Communicative Language Teaching compared. Retrieved from https://gianfrancoconti.wordpress.com/2016/01/12/grammar-translation- and-communicative-language-teaching-compared/. Corder, S., P. (1967). The significance of learners’ errors. Int. Rev. Appl. Linguist., 5(1), 161– 169. https://doi: 10.1515/iral.1967.5.1-4.161. Dagilienė, I. (2012). Translation as a learning method in English language teaching. Studies about Languages, 21, 124–129. http://dx.doi.org/10.5755/j01.sal.0.21.1469. Duff, A. (1994). Translation: Resource books for teachers. Oxford: Oxford University Press. Ellis, R. (2006). Current issues in the teaching of grammar: An SLA perspective. TESOL Quarterly, (40)1, 83–107. https://doi: 10.2307/40264512. Gorsuch, G., J. (1998). Yakudoku EFL instruction in two Japanese high school classrooms: An exploratory study. JALT Journal, 20(1), 6–31. Retrieved from http://jalt- publications.org/jj/articles/2777-yakudoku-efl-instruction-two-japanese-high-school- classrooms-exploratory-study. Guerra, F., A. (2014). The usefulness of translation in foreign language learning: Students’ attitudes. International Journal of English Language and Translation Studies, 2(1), 153– 170. Retrieved from https://www.researchgate.net/publication/261641087_The_Usefulness_of_Translation_ in_Foreign_Language_Learning_Students%27_Attitudes. Howatt, A., P., R. (1984). A history of English language teaching. New York, USA: Oxford University Press. Ionin, T., Heejeong, K., & Wexler, K. (2004). Article semantics in L2 acquisition: The role of specificity. Language Acquisition, 12(1), 3–69. https://doi: 10.1207/s15327817la1201_2. https://doi:10.5539/elt.v4n2p13 https://gianfrancoconti.wordpress.com/2016/01/12/grammar-translation-and-communicative-language-teaching-compared/ https://gianfrancoconti.wordpress.com/2016/01/12/grammar-translation-and-communicative-language-teaching-compared/ https://doi.org/10.1515/iral.1967.5.1-4.161 https://doi.org/10.1515/iral.1967.5.1-4.161 https://www.researchgate.net/profile/Ana_Fernandez_Guerra?_sg=Y5Hwb2OtLS9Bd71I3gYruVxwHHG4-ISyG065spJPGFRf57O-45fCydc2sqIvXX_3kjn4Jsw.nnYKlqpdlaJz-4yNoNHPh2FPaDsD9FkFsh6-zBgXDHpQXbR_rVoZQxAHJBt4orL1UXLHI0yhw8hziMpa7Zb94Q https://doi.org/10.1515/iral.1967.5.1-4.161 Ciftci & Ozcan 728 Jewad, H., G., & Verma, S., S. (2014). A comparative study of the difficulties faced by the teachers in Communicative Language Teaching & Grammatical Translation Method in the primary schools of the rural areas of Allahabad. European Academic Research, 2(2), 2174–2205. Retrieved from http://euacademic.org/ArticleDetail.aspx?id=553. Kırkgoz, Y. (2007). English language teaching in Turkey: Policy changes and their implementations. RELC Journal, 2(38), 216–228. https://doi: 10.1177/0033688207079696. Krashen, S., & Terrell, T. (1983). The natural approach: Language acquisition in the classroom. London, UK: Phoenix ELT. Koosha, M., & Yakhabi., M. (2013). Problems associated with the use of Communicative Language Teaching in EFL contexts and possible solutions. International Journal of Foreign Language Teaching and Research, 1(2), 63–76. Retrieved from http://jfl.iaun.ac.ir/article_4185.html. Larsen-Freeman, D. (1986). Techniques and principles in language teaching. New York, USA: Oxford University Press. Leedy, P., & Ormrod, J., E. (2015). Practical research: Planning and design (11th ed.). Essex, England: Pearson Education Limited. Liu, M. (2007). Anxiety in oral English classrooms: A case study in China. Indonesian Journal of English Language Teaching, 3(1), 119–137. http://dx.doi.org/10.25170%2Fijelt.v3i1.132. Master, P. (1997). The English article system: Acquisition, function, and pedagogy. Pergamon, 25(2), 215–232. http://dx.doi.org/10.25170%2Fijelt.v3i1.132. Ministry of National Education (2013). A guide for the transition system from basic to secondary education. Ankara: MoNE. Retrieved from http://www.meb.gov.tr/sinavlar/dokumanlar/2013/kilavuz/2013_OGES_Klvz.pdf. Ozcan, M. (2012). Novelist calls out poemist: A psycholinguistic and contrastive analysis of the errors in Turkish EFL learners’ English. International Scholarly and Scientific Research & Innovation, 8(6), 2311–2322. Retrieved from https://waset.org/publications/2632/novelist-calls-out-poemist-a-psycholinguistic-and- contrastive-analysis-of-the-errors-in-turkish-efl-learners-interlanguage. Ozsevik, Z. (2010). The use of Communicative Language Teaching (CLT): Turkish EFL teachers’ perceived difficulties in implementing CLT in Turkey (Master’s thesis). University of Illinois, USA. Öztürk, G., & Gürbüz, N. (2014). Speaking anxiety among Turkish EFL learners: The case at a state university. Journal of Language and Linguistic Studies, 10(1), 1-17. Retrieved from http://jlls.org/index.php/jlls/article/viewFile/178/165. Pan, Y., & Pan, Y. (2012). The use of translation in the EFL classroom. Philippines ESL Journal, (9), 4–23. Retrieved from https://www.researchgate.net/publication/308633917_Use_of_Translation_in_the_Clas sroom_by_EFL_Teachers_in_Libya_A_descriptive_study. Richards, J., C., & Rodgers, T., S. (2002). Approaches and methods in language teaching. Cambridge: Cambridge University Press. Ross, N., J., (2000). Interference and intervention: Using translation in the EFL classroom. Modern English Teacher, 9(3), 61–66. Retrieved from http://web.tiscali.it/njross/interfereart.htm. Sato, K., & Kleinsasser, R. (1999). Communicative language teaching (CLT): Practical understandings. Modern Language Journal, 83(4), 494-517. https://doi: 10.1111/0026- 7902.00037. https://doi.org/10.1515/iral.1967.5.1-4.161 http://dx.doi.org/10.25170%2Fijelt.v3i1.132 http://jlls.org/index.php/jlls/article/viewFile/178/165 https://www.researchgate.net/publication/308633917_Use_of_Translation_in_the_Classroom_by_EFL_Teachers_in_Libya_A_descriptive_study https://www.researchgate.net/publication/308633917_Use_of_Translation_in_the_Classroom_by_EFL_Teachers_in_Libya_A_descriptive_study https://doi.org/10.1515/iral.1967.5.1-4.161 International Online Journal of Education and Teaching (IOJET) 2021, 8(2), 709-729. 729 Şevik, M. (2007). The place of mother tongue in foreign language classes. Ankara University, Journal of Faculty of Educational Sciences, 1(40), 99–119. https://doi: 10.1501/Egifak_0000000163. Taylor, B., P. (1975). The use of overgeneralization and transfer learning strategies by elementary and intermediate students of ESL. Language Learning, 25(1), 73–107. https://doi: 10.1111/j.1467-1770.1975.tb00110.x. Thompson, G. (1996). Some misconceptions about Communicative Language Teaching. ELT Journal, 1(50), 9–15. https://doi: 10.1093/elt/50.1.9. Widodo, H. (2006). Approaches and procedures for teaching grammar. Practice and Critique, 5(1), 122–141. https://doi: 10.1093/elt/50.1.9. Wong, C. C. Y., & Barrea-Marlys, M. (2012). The role of grammar in Communicative Language Teaching: An exploration of second language teachers’ perceptions and classroom practices. Electronic Journal of Foreign Language Teaching, 9, 61–75. Retrieved from https://www.scirp.org/(S(czeh2tfqyw2orz553k1w0r45))/reference/ReferencesPapers.as px?ReferenceID=1849149. Uysal, H., H., & Bardakci, M. (2014). Teacher beliefs and practices of grammar teaching: focusing on meaning, form, or forms? South African Journal of Education, 34(1), 1–16. https://doi:10.15700/201412120943. Xia, Y. (2014). Language theories and language teaching from Traditional Grammar to Functionalism. Journal of Language Teaching and Research, 5(3), 559–565. https://doi: 10.4304/jltr.5.3.559-565. https://doi.org/10.1515/iral.1967.5.1-4.161 https://doi.org/10.1515/iral.1967.5.1-4.161 https://doi.org/10.1515/iral.1967.5.1-4.161 https://doi.org/10.1515/iral.1967.5.1-4.161 https://www.scirp.org/(S(czeh2tfqyw2orz553k1w0r45))/reference/ReferencesPapers.aspx?ReferenceID=1849149 https://www.scirp.org/(S(czeh2tfqyw2orz553k1w0r45))/reference/ReferencesPapers.aspx?ReferenceID=1849149 https://doi.org/10.1515/iral.1967.5.1-4.161 https://doi.org/10.1515/iral.1967.5.1-4.161