ENGLISH REVIEW: Journal of English Education p-ISSN 2301-7554, e-ISSN 2541-3643 Volume 11, Issue 2, June 2023 https://journal.uniku.ac.id/index.php/ERJEE 479 UNVEILING COHERENCE: ANALYZING GRAMMATICAL DEVICES IN RECOUNT TEXT Dewi Nurpitriyani Department of English Education, Faculty of Educational Sciences, UIN Syarif Hidayatullah Jakarta Indonesia E-mail: dewi.nurpitriyani20@mhs.uinjkt.ac.id Didin Nuruddin Hidayat Department of English Education, Faculty of Educational Sciences, UIN Syarif Hidayatullah Jakarta, Indonesia E-mail: didin.nuruddin@uinjkt.ac.id Nida Husna Department of English Education, Faculty of Educational Sciences, UIN Syarif Hidayatullah Jakarta, Indonesia E-mail: nida.husna@uinjkt.ac.id Alek Department of English Education, Faculty of Educational Sciences, UIN Syarif Hidayatullah Jakarta, Indonesia E-mail: alek@uinjkt.ac.id APA Citation: Nurpitriyani, D., Hidayat, D. N., Husna, N., & Alek. (2023). Unveiling coherence: Analyzing grammatical devices in recount text. English Review: Journal of English Education, 11(2), 479-488. https://doi.org/10.25134/erjee.v11i2.5703 Received: 19-02-2023 Accepted: 27-04-2023 Published: 30-06-2023 INTRODUCTION Writing is an essential skill because it allows us to express our ideas and thoughts as well as convince and persuade others in written form (Novariana et al., 2018; Sianturi et al., 2020). However, writing is not an easy process and generating a text that the reader can easily understand is complicated since it involves a long process and multiple practices (Chung et al., 2020). In the case of English for Foreign Language (EFL) students, writing is considered much more difficult as students tend to transfer their first language into the target language (Aziz et al., 2020). Besides, students struggle to express their ideas into sentences, turning sentences into paragraphs, and turning paragraphs into a whole discourse for their lack of writing practice (Nasiba & Norboyevich, 2022). Furthermore, the Abstract: The purpose of this study is to examine the grammatical cohesive devices used in the recount text writing of bilingual students. The cohesive device is crucial because it consists of grammatical and lexical devices that contribute to the writing’s strength and accuracy, as well as its contextual understanding. Twenty bilingual ninth-grade students of a junior high school participated in this study. The students’ recount text writings were used as the main data in this research. The qualitative descriptive method employed in this study to thoroughly assess and describe students’ writing and its’ cohesive device used. Further, the data analysis used was content analysis consisting of three steps: categorization, abstraction, and coding. The result shows that the bilingual ninth-grade students were able to use grammatical cohesion in their recount text writing. The result revealed the reference is most dominated in 68.42% and the second device is in conjunction with 29.97%. Meanwhile, substitution showed the low rank with 1.61% and the last was ellipsis with 0%. Furthermore, in the reference device category, personal reference becomes the most highly used at 85%, meanwhile in the conjunction category, additive category becomes the most frequent one at 46% as well as nominal substitution at 50% in the substitution device. It was also found that the ninth-grade students of bilingual class can write well using numbers of grammatical cohesive devices even though there was still some monotonous repetition in their writing. he findings of this study have pedagogical implicature as the need to enhance students’ knowledge of grammatical cohesion device because understanding of cohesiveness is necessary for pertaining to the text’s cohesion and coherence particularly in writing ability. Keywords: grammatical cohesion devices; recount text; students’ writing. mailto:dewi.nurpitriyani20@mhs.uinjkt.ac.id mailto:alek@uinjkt.ac.id Dewi Nurpitriyani, Didin Nuruddin Hidayat, Nida Husna, & Alek Unveiling coherence: Analyzing grammatical devices in recount text 480 problems get more complicated in Curriculum 2013 as students are required to master a variety of genres simultaneously with their distinct purposes, language features, and structures specified for each genre (Ariwibowo et al., 2023). Three elements of constructing well-structured writing are unity, coherence, and adequate development (Asfah, 2019). Accordingly, they also stated that good writing should be cohesive and coherent. Cohesion is required to establish connections between what is being said in the writer’s mind and what the writer tends to say in a text. Cohesive is the element that binds the text together by redundant information at the semantic, syntactic, and discourse structure levels (Khalaf, 2020). Akbar et al. (2022) also used a similar definition, stating that cohesion connects phrases to retain their fundamental meaning. A well-chosen cohesive device will result in a well- structured and coherent text. Meanwhile, inappropriate usage of a cohesive device will result in readers’ confusion. Cohesive devices also have a strong influence on writing’s cohesion as they provide learners with a variety of syntactic that can be used to make any piece of discourse more cohesive (Schiffrin et al., 2001; Ludji et al., 2022). An effective way to construct a good paragraph in writing is from its coherence and cohesion. According to Marashi (2021), coherence is the ability to join one word to another. The sentences that must hold together demonstrate the text’s coherence, which indicates that the transitions between words and phrases must be logical and easy to follow (Putri, 2021). Each word, phrase, and sentence should be simple to comprehend and organized rationally. Halliday and Hasan (1976) added that cohesion is symbolized and shared partially through grammar or the vocabulary in the text provided. Cohesive devices are significant for creating effective writing, both coherence and cohesion. Yule (2019) argues that cohesion could be drawn as a tie and connection within the text. It can also be pointed as a part of the system to correlate an item with either the first or the following text. Without cohesion and coherence, writing would be unclear and misunderstood by the readers. As Islami et al. (2021) address, cohesion contributes to text readability, comprehensibility, and clarity of arguments. Although cohesion and coherence seem different, they have some common features; both are evolving. Cohesion is symbolized and shared partially through grammar or the vocabulary in the text provided. Therefore, it includes two types of cohesion: grammatical and lexical. Grammatical cohesion represents the tie, the rope, and the link conveyed and indicated in the grammatical system of a language, such as reference, substitution, conjunction, and ellipsis. In contrast, lexical cohesion refers to the connection achieved by selecting vocabulary (Verdiansyah et al., 2019). The cohesion of grammar and lexicon is one of the essential factors for the process of receiving the organization for human thoughts. Thus, if the grammatical and lexical cohesion are appropriate enough, the discourse is structured well. Previous research from Rudiana (2021) evaluated the use of grammatical cohesiveness in argumentative essays written by students. He discovered that students commonly employed the sort of grammatical cohesion devices and that it enhanced students’ writing successfully. Moreover, Shahzad (2021) found that the students focus on some grammatical devices and neglect others. In the paragraphs, the misuse and overuse of devices showed their lack of grammatical knowledge and a bad influence on English. As an instructor, she noticed more influence on them of their mother tongue. From these two studies, we can conclude that a grammatical cohesion device is significant in building coherence and cohesive writing, both effective and meaningful. Furthermore, other related research from Nurhidayat et al. (2021) and Syam (2020) revealed how significant grammatical cohesive devices are to build a well and meaningful writing since using grammatical cohesive devices, enhances us to use references to show our personal thought or to address the existence of the writer by using reference device, it may also to add something, or to oppose something by using conjunction, or to make our writing more natural, we are also able to substitute to avoid the redundant or repetition in our writing by using substitution device, and or omit something in the term of ellipsis device. The findings above indicate that students continue to struggle with using cohesive devices. Thus, studying the coherent strategies used in students’ writing is vital to ascertain where gaps exist and what aspects of students’ writing should be reviewed and improved. Cohesion is one of the vital factors in determining the quality of writing; meanwhile, cohesive devices are significant because they include grammatical and lexical devices that help strengthen the writing and make it more accurate and contextually understandable. Subsequently, without discourse analysis, ENGLISH REVIEW: Journal of English Education p-ISSN 2301-7554, e-ISSN 2541-3643 Volume 11, Issue 2, June 2023 https://journal.uniku.ac.id/index.php/ERJEE 481 grammar, vocabulary, and other linguistic components, the literary result will be inadequate and inaccurate (Meylinda, 2023). Cohesive devices indicate relationships between sentences and elements of the text. This means that the proper use of coherent devices allows readers to grasp the connection between what comes before and what comes after. Cohesive devices play a crucial role in the context of English as a Foreign Language (EFL) learners. According to Cheng and Tsang (2022), cohesive writing is highly relevant for all EFL students, as it is a key factor in producing more coherent written compositions. Furthermore, it guides students in improving their writing skills. In Indonesia, mastering English writing is considered an essential skill for students. One of the necessary skills for effective writing is the ability to complete coherence and cohesion. Notably, bilingual schools in Indonesia offer a unique educational environment that further highlights more exposure to English. This notion raises interesting questions regarding students' use of grammatical cohesive devices in bilingual schools. Recount text, a genre characterized by providing information or retelling experiences, is often examined as an expression that reflects personal opinions and emotions (Mingsakoon & Srinon, 2018). Notably, when students write about themselves, their engagement with the writing task tends to be higher, which motivates them to write more (Barratt-Pugh et al., 2021). In recount text, students must follow language features such as tenses, action verbs, conjunction, and so on (Iriana, 2018). Given this view, it becomes evident that recount text is a suitable genre for encouraging student writing and examining the use of grammatical cohesive devices. Regarding the importance of students’ writing ability development, exploring and investigating the use of grammatical cohesive devices in the students’ recount text writings will give more valuable insight into how students engage cohesive devices. Furthermore, this is also expected to give a more profound understanding of how these cohesive devices into the writing to make it more coherence and clearer. Then, we will be able to identify the weaknesses and strengths in every aspect of each grammatical cohesive device and help uncover the specific challenges and difficulties students encounter when attempting to create cohesive and coherent texts. All in all, research on grammatical cohesive devices is to inform teachers or other language and education practitioners to increase students’ writing skills and contribute to developing more effective language teaching methods. Considering these notions, it is noteworthy to explore the specific aspects of grammatical cohesive devices employed by students in bilingual classes, which may demonstrate better variation and proficiency than in other educational settings due to their English exposure. Thus, this research aims to investigate the use of grammatical cohesive devices in the bilingual ninth-grade students recounting text writings. The research questions were proposed as follows: (1) What common grammatical cohesive device types are frequently used in the ninth-grade bilingual students’ recount text writing? (2) What kinds of grammatical cohesive device specifications are found in the ninth-grade bilingual students’ recount text writing related to its grammatical cohesive device specification? METHOD The method used in this study is qualitative descriptive analysis to characterize a phenomenon and its characteristics (Doyle et al., 2020). In this type of method, the data is presented in a description to give detailed results. The grammatical cohesion devices that students employ in their recount text writing are discussed in detail through analysis and description. The students’ recount text writings were analyzed and categorized to identify the types or subtypes of grammatical cohesion devices; reference, substitution, conjunction, and ellipsis used by the students and their dominance in the writings. 20 of the ninth-grade bilingual students of MTs Pembangunan UIN Jakarta were chosen as the subjects of this research. They were asked to share their experience of having online classes during the Covid-19 pandemic into recount text writings. In this study, the instrument was a writing test. It measures students’ writing ability to tell their experiences in online learning during the Covid- 19 pandemic into recount text writing and to identify types and subtypes of grammatical cohesion devices used in their writings. Twenty bilingual ninth-grade students of MTs Pembangunan UIN Jakarta were required to write recount texts on their experiences attending online classes during the Covid-19 pandemic. The writing should follow the recount text generic structure consisting of orientation, event, re- orientation, and language feature characteristics, which include using simple past tense to convey Dewi Nurpitriyani, Didin Nuruddin Hidayat, Nida Husna, & Alek Unveiling coherence: Analyzing grammatical devices in recount text 482 information that has already occurred. The text was typed on A4 paper with a 3 cm margin on each side, at least three paragraphs with a minimum of seven sentences in each, saved as PDF, and uploaded to the Google Drive provided. The steps were collecting students’ recount text writing, reading and calculating every grammatical cohesion device in students’ writing, and classifying them based on their subtypes with the formula: P = percentage N = types or sub-types of grammatical cohesive devices T = total grammatical cohesive devices produced by students In accordance with Lindgren et al. (2020), content analysis consists of three steps: categorization, abstraction, and coding. Categorization is a procedure in which the researchers grouped the data based on similarities to construct a generalization. The data is classified according to the material that pertains to the research’s theoretical framework on cohesive devices. The second is an abstraction, which refers to the process through which the researchers developed a broad explanation by elaborating on a specific explanation. Then the third is code. When coding the data, the researchers read it thoroughly from beginning to end, looking for every line with information relevant to the research’s primary category. The researchers should match the elements discovered in the text to the categories established previously during the categorization process. It makes it easier for the researcher to locate the data to be evaluated. After completing all procedures, the researchers drew a conclusion regarding the types of cohesive devices that bilingual students typically employ in their writing and the accuracy of the cohesive devices used by bilingual students. RESULTS AND DISCUSSION This research's result is divided into two parts to reveal the research questions stated before. The first part focuses on the problem of what common grammatical cohesive device types are frequently used in ninth-grade bilingual students’ recount text writing. Each device was categorized into its own part in the percentage. Meanwhile, the second part was the answer to the kinds of grammatical cohesive device specifications found in the ninth-grade bilingual students' recount text writing, and it explained in detail its grammatical cohesive device specification. The use of grammatical cohesion devices This section explained the most frequent grammatical cohesion devices used by the ninth- grade of bilingual students of MTs Pembangunan UIN Jakarta in their recount text writings are shown in the table below: Table 1. Frequent grammatical cohesive devices Type of Grammatical Cohesion Students’ Grammatical Device Use Students’ Cohesion in Percentage (%) References 1,018 68.42% Substitution 24 1.61% Conjunction 446 29.97% Ellipsis 0 0 Total 1,488 100% Table 1 shows a reference as the most frequent grammatical cohesiveness device used by bilingual ninth-grade students with 68.42 % of all grammatical cohesion devices. The second type of grammatical cohesive device is 29.97%. Meanwhile, substitution was used at just 1.61 %, while ellipsis received 0%. This result draws into the assumption that ninth-grade students are unfamiliar with substitution and ellipsis. In line with the findings of Trisnaningrum et al. (2019), who discovered that participants were more familiar with both reference and conjunction than substitution or ellipsis, the study found that participants were more familiar with the reference and conjunction. Furthermore, the findings revealed that students' writing was unable to make use of grammatical cohesiveness techniques because of their incapacity to do so. To obtain more particular information, the researchers divided the grammatical cohesive devices used by the bilingual ninth-grade students into specific types and subtypes of grammatical cohesive devices, which show the highest among other grammatical cohesion devices in reference with 861 in personal reference, followed by conjunction with 204 in additive conjunction, and the last is substitution with 12 nominal substitution as the most frequent subtype of grammatical cohesion devices used. The detail data is shown in the table below: ENGLISH REVIEW: Journal of English Education p-ISSN 2301-7554, e-ISSN 2541-3643 Volume 11, Issue 2, June 2023 https://journal.uniku.ac.id/index.php/ERJEE 483 Table 2. Types of grammatical cohesive devices use List of Student References Substitution Conjunction Ellipsis Total Perso nal Dem onstr ative Com parat ive Nom inal Verb al Clau sal Addi tive Adve rsati ve Tem poral Cau sal Nomi nal Ver bal Cau sal Student 1 48 3 1 2 1 0 2 6 2 5 0 0 0 70 Student 2 37 10 0 1 1 0 6 3 4 2 0 0 0 64 Student 3 57 6 1 0 0 0 6 4 3 3 0 0 0 80 Student 4 57 2 2 0 1 1 5 1 4 4 0 0 0 77 Student 5 30 4 2 0 0 0 10 3 4 4 0 0 0 57 Student 6 67 11 0 1 1 0 10 2 10 6 0 0 0 108 Student 7 37 3 2 0 1 0 10 1 16 2 0 0 0 72 Student 8 35 3 0 1 0 0 6 1 9 2 0 0 0 57 Student 9 37 5 3 0 1 0 18 5 4 1 0 0 0 74 Student 10 40 13 5 0 0 1 13 4 9 1 0 0 0 86 Student 11 50 2 8 1 0 0 13 4 4 1 0 0 0 83 Student 12 48 3 5 1 0 0 15 9 5 8 0 0 0 94 Student 13 38 4 4 2 1 0 15 2 2 2 0 0 0 70 Student 14 57 3 2 0 0 1 3 1 4 4 0 0 0 75 Student 15 35 7 6 1 1 0 13 5 5 4 0 0 0 77 Student 16 33 4 0 1 0 0 6 1 9 2 0 0 0 56 Student 17 45 3 1 1 1 0 14 3 6 6 0 0 0 80 Student 18 30 3 1 0 0 0 9 2 3 3 0 0 0 51 Student 19 38 7 3 0 0 0 17 5 4 1 0 0 0 75 Student 20 42 11 4 0 0 0 13 4 7 1 0 0 0 82 TOTAL 861 107 50 12 9 3 204 66 114 62 0 0 0 1,488 To get more detail on the percentage of each type, the following table provides a description of each kind of sub-type grammatical cohesion found in the bilingual students' recount text writings. Three different pie charts consist of reference, conjunction, and substitution due to ellipsis not found in the students’ writing. Each section indicates a sub-type of grammatical cohesion device and the device's frequency. The pie chart figures are shown in the following ways: Figure 1. Reference According to the pie chart above, reference is the form of grammatical cohesive device most frequently utilized compared to other types. A personal reference is the most prevalent subtype with 85%, followed by demonstrative reference at 10% and comparison at 5%. Furthermore, the researchers discovered a large number of reference devices that were employed in this study. The number of reference devices used by bilingual ninth-grade students illustrates their preference for personal references to make their recount text writing more accessible and to link their own experiences to the readers, which is the major purpose of the recount text. Personal pronouns such as I, my, our, his, her, and it are frequently used in formal writing. Additionally, the text makes use of demonstrative references such as this, that, and there, as well as comparative references such as less and more. Next, the most frequent grammatical cohesive device is followed by conjunction in 29.97%. In conjunction, additive conjunction has the most frequent, which is 48%, temporal 25%, adversative 15%, and causal 14%. The pie chart is as follows: Dewi Nurpitriyani, Didin Nuruddin Hidayat, Nida Husna, & Alek Unveiling coherence: Analyzing grammatical devices in recount text 484 Figure 2. Conjunction Conjunction, as a means of connecting one clause to another, is frequently used among bilingual ninth-grade students. There are four types of conjunction devices: additive, adversative, temporal, and causal. The additive conjunction joins two language parts that are in the same place. While adversative conjunctions arise when the first clause’s statement contradicts or is not equivalent to the second clause’s statement. Then, a temporal conjunction is used to connect events in time, but a causal conjunction is used to connect causes and effects that occur in the same phrase. Researchers discovered that additive conjunctions were more frequently used than adversative conjunctions. This is seen by the numerous instances in which conjunctions like “and” and “also” are used in repetition. The second is temporal conjunction, in which students relate their experiences to time using numerous examples such as then, next, after that, first, and at that time. The usage of adversative conjunctions, on the other hand, is restricted to forms such as but in a few. The last kind of conjunction is causal, which students frequently use because and thus. Additionally, nominal substitution uses 50% of substitution devices, followed by verbal substitution at 37% and clausal substitution at 13%. The chart is as follows: Figure 3. Substitution Substitution is a cohesive device that substitutes later phrases that are not repeated in the next clause or sentence. The replacement device generates more dynamic sentences by substituting for the term. The research discovered that substitute sentences were relatively evenly distributed and varied. These terms contain substitutes such as one, one’s, do, did, and thus. This demonstrates that there is a connection between the two sentences. The kinds of grammatical cohesive devices specifications The description of grammatical cohesion devices used by the bilingual ninth-grade students recount text writing presented in Table 3 below: Table 3. The kinds of grammatical cohesive devices specifications Type of grammatical cohesion devices A subtype of grammatical cohesion devices The kinds of grammatical cohesion devices in students’ The number of grammatical ENGLISH REVIEW: Journal of English Education p-ISSN 2301-7554, e-ISSN 2541-3643 Volume 11, Issue 2, June 2023 https://journal.uniku.ac.id/index.php/ERJEE 485 Recount Text Writing cohesion devices uses References Personal I, me, my we, our, us, she, her, he, him, his, it, they, them, their 861 Demonstrative this, that, there, now 107 Comparative more, less, better 50 Substitution Nominal one, one’s 12 Verbal do, did, done 9 Causal so 3 Conjunction Additive and, and also, in addition, moreover 204 Adversative but, however, even though, despite all, instead 66 Temporal then, the last, after that, before, beforehand, next day, until, in the end, at the same time, previously, later, another day 114 Causal because, for, so 62 Ellipsis Nominal Verbal Causal TOTAL 1488 Reference The study discovered 861 personal references the ninth-grade bilingual students used in recounting text writings. The students use personal pronouns such as I, me, my, mine, we, our, she, his, her, them, and it. In terms of writing about prior experiences or occurrences, they undoubtedly tell their own and incorporate the other person (male/female) and item as the references. Additionally, the findings indicate that students use this, that, there, and now as demonstrative references when referring to the total 107 and other, less, more when referring to the total 50. Thus, the results indicate that most students are familiar with the kinds of references in their recount text writing. Conjunction The study found 204 additive conjunction and the most frequent such as and, additionally, and then followed by adversative in total 66 such as but even though, despite all, and instead. After that, it is also found 144 temporal conjunctions such as then, the last, after that, before, beforehand, next day, until, in the end, at the same time, previously, later, another day and the last one is a causal conjunction such as because, for, and so. Substitution The study located 12 nominal replacement devices, such as one or one’s, followed by 9 verbal substitutions, such as do, done, or, did, and three clausal so. The explained result above reveals that the use of grammatical cohesion devices in the bilingual ninth-grade students’ recount text writing is dominated by reference devices as the most frequent device, which is 68.42%, followed by conjunction at 29.97%, and substitution at 1.61%. This present study's results are in line with several similar research such as those conducted by Trisnaningrum et al. (2019), which showed a result that 1048 grammatical cohesion devices were found in the students’ academic writing essays with reference devices as the most frequent ones at 53.3%. Moreover, another similar research from Satria and Handayani (2018) found that students’ descriptive text writing consisted of reference with 63.70%, conjunction with 30.66%, ellipsis with 6.35%, and substitution with 0.27%. Both research studies revealed that the reference device is the most frequent one compared to the other, and this is also in line with the writer’s research, which found reference 68.42% with the most frequent use sub-type in reference is the personal reference at 85%. This finding is possibly caused by the particular genre that the students wrote. Since they wrote about recount text which has the social function of telling the writer’s personal experience (Mediska & Adnan, 2019). They require the use of more personal references such as I, my, our, his, her than any Dewi Nurpitriyani, Didin Nuruddin Hidayat, Nida Husna, & Alek Unveiling coherence: Analyzing grammatical devices in recount text 486 other text types. Thus, the students’ massive use of reference in their recount writing in this study is as expected and as it is supposed to be based on the genre characteristics. Further, this finding implies that students were aware of the genre they were writing and aware of its function as well as its linguistic features. In addition, additive conjunctions are found still limited and monotonous. Students frequently use “either-or” and “also” to add something, but only a few uses in addition or moreover. This is also in line with adversative conjunction, which is frequently used but to contrast something only a few students use however and even though. The findings also identify a gap between additive conjunction and adversative conjunction which is three times higher, from 48% or 204 to 15% or 64. This finding is in line with Faizah et al. (2020), Harliani et al. (2021), and Jayanti and Hidayat (2021) results, in which the additive conjunction is repeated many times of all the forms of conjunctions from the writing they analyzed. The number of additive conjunctions revealed in this study compared to adversative conjunctions shows that were more concerned with adding information rather than opposing the information, which may reflect their understanding of the concept of sequencing events in the recount text. Meanwhile, the limited variety of conjunction used by the students might not only imply their glossary, their richness of vocabulary, as well as their frequency of writing practices but may also imply their thinking patterns and how their brains work which need to be investigated further. There is also a gap in the result of analysis data, specifically in ellipsis. Ellipsis was not found in the students’ writing. Ellipsis is defined as the deletion of unnecessary things in a sentence (Halliday & Hasan, 1976). It may also be called “zero substitution”. It is like omitting the word, e.g., I lay on my back and let them move as they would, so the move was not repeated. In the present study, where students tell their own experiences in writing in the form of recount text, they always use complete sentence. Meanwhile, the study also found that students used substitution the least in their recount writing. This shows that most students still lack competence in using grammatical cohesive well, as is also shown in Albana et al. (2020), Sari (2021), and Trisnaningrum et al. (2019), where ellipsis was not found in their results of the research. However, in this case, junior high students have not used the ellipsis and substitution in their writing due to several factors, such as their lack of awareness of the items, their level of understanding, and their level of grammatical. In contrast with the prior discussions, it is also found that some students could write many grammatical cohesive devices in their recount writing (shown in Table 2), with the highest being 108 and the least being 56. This result is even higher or the same as the subject in Albana et al. (2020), reaching 100 grammatical cohesive devices. This also revealed that junior high schools could use grammatical cohesive devices. The findings of this study also show pedagogical implicature, including the need to enhance students’ knowledge of grammatical cohesion devices. Increasing this knowledge and experience will increase students’ ability to use a variety of cohesive devices. Therefore, students should receive material about the text’s cohesion and coherence. In junior high school, they will be exposed to beneficial device cohesion, particularly in terms of writing ability. CONCLUSION In conclusion, the bilingual ninth-grade students of MTs Pembangunan UIN Jakarta employ a variety of grammatical cohesive devices. The students utilized all of the cohesive items to create cohesion and coherence in their writing. Cohesion enables students to comprehend certain missing components, resulting in a more precise understandable writing. Only ellipsis does not appear in the bilingual ninth-grade students’ recount text writing. No student makes use of ellipses in their work. Furthermore, it can be argued that students have a strong ability to write a coherent paragraph as a result of their ability to use grammatical cohesion devices: reference, substitution, and conjunction in their writing. It was also found that the ninth-grade students of bilingual class are able to write well using numbers of grammatical cohesive devices even though there was still some monotonous repetition in their writing. The findings of this study also show pedagogical implicature, including the need to enhance students’ knowledge of grammatical cohesion devices. Increasing this knowledge and experience will increase students’ ability to use a variety of cohesive devices. Therefore, students should be provided with material pertaining to the text’s cohesion and coherence. In junior high school, they will be exposed to device cohesion that is beneficial particularly in terms of writing ability. ENGLISH REVIEW: Journal of English Education p-ISSN 2301-7554, e-ISSN 2541-3643 Volume 11, Issue 2, June 2023 https://journal.uniku.ac.id/index.php/ERJEE 487 REFERENCES Albana, H. H., Marzuki, A. G., & Hidayat, D. N. (2020). Cohesive devices in student’s writing: A discourse analysis on argumentative text). Jurnal Pendidikan Humaniora, 8(1), 6–11. Akbar, P., Mujiyanto, K. J., & Sutopo, D. (2022). Assessing grammatical cohesive devices in the findings and discussion of research articles by graduate students. English Education Journal, 12(3), 427–436. Ariwibowo, T., Hidayat, D.N., Husna, N., & Alek. (2023). A discourse analysis of cohesion devices on students’ writing of recount text. Pendidikan Jurnal, Sosial, dan Agama, 9(1). 33-42. https://doi.org/10.53565/pssa.v9i1.687 Asfah, I. (2019). An analysis of paragraph writing abilities of English Department students UNM. ELT Worldwide, 6(1), 1–9. https://doi.org/10.26858/eltww.v6i1.9339 Aziz, Z. A., Fitriani, S. S., & Amalina, Z. (2020). Linguistic errors made by Islamic university EFL students. Indonesian Journal of Applied Linguistics, 9(3), 735–748. https://doi.org/10.17509/ijal.v9i3.23224 Barratt-Pugh, C., Ruscoe, A., & Fellowes, J. (2020). Motivation to write: Conversations with emergent writers. Early Childhood Education Journal, 49(2), 223–234. https://doi.org/10.1007/s10643-020-01061-5 Cheng, A., & Tsang, A. (2022). Use and understanding of connectives: an embedded case study of ESL learners of different proficiency levels. Language Awareness, 31(2), 155–174. https://doi.org/10.1080/09658416.2021.1871912 Chung, P., Patel, D. R., & Nizami, I. (2020). Disorder of written expression and dysgraphia: Definition, diagnosis, and management. Translational Pediatrics, 9(S1), 46–54. https://doi.org/10.21037/tp.2019.11.01 Doyle, L., McCabe, C., Keogh, B., Brady, A., & McCann, M. (2020). An overview of the qualitative descriptive design within nursing research. Journal of Research in Nursing, 25(5), 443–455. https://doi.org/10.1177/1744987119880234 Faizah, Sutopo, D., & Faridi, A. (2020). The use of cohesive devices in English Education journal articles written by graduate students of Universitas Negeri Semarang. English Education Journal, 10(2), 208–213. Halliday, M. A. K., & Hasan, R. (1976). Cohesion in English. Routledge. Harliani, N., Zamzam, A., Hanafi, N., & Mataram, U. (2021). A study on grammatical cohesion in the descriptive essays by the third semester students of English Education Program in the University of Mataram in the academic year 2019/2020. LISDAYA: Jurnal Linguistik (Terapan), Sastra, Dan Budaya, 17(1), 49–58. Iriana, A. L. (2018). Analyzing language features of recount text written by MA Hidayatu Umam students. RETAIN, 6(1), 26–37. Islami, H. H., Saleh, M., & Bharati, A. L. D. (2022). The use of cohesive devices in descriptive text by English Training participants at PST. English Education Journal, 12(1), 96–103. Jayanti, D. N., & Hidayat, D. N. (2021). Grammatical cohesive devices in reading text: A discourse analysis of English test for junior high school. JET (Journal of English Teaching) Adi Buana, 6(01), 1–6. https://doi.org/10.36456/jet.v6.n01.2021.2963 Khalaf, Y. S. (2020). Cohesive ties in English language. Journal of Archaeology of Egypt/Egyptology, 47(2). 521–532. Lindgren, B., Lundman, B., & Graneheim, U. H. (2020). Abstraction and interpretation during the qualitative content analysis process. International Journal of Nursing Studies, 108, 1–6. https://doi.org/10.1016/j.ijnurstu.2020.103632 Ludji, I. D. R., Hambandima, E. S. N., & Christiani, Y. N. (2022). Cohesive devices used in students’ argumentative essay: A discourse analysis. VELES (Voices of English Language Education Society), 6(1), 14–24. https://doi.org/10.29408/veles.v6i1.4969 Marashi, S. J. (2021). Impact of teaching cohesive devices to Iranian EFL learners: Case of IELTS writing task 2 examination. JL3T (Journal of Linguistics, Literature and Language Teaching), 6(2), 115–126. https://doi.org/10.32505/jl3t.v6i2.2050 Mediska, G., & Adnan, A. (2019). Improving student’s speaking by using Cocktail Party technique in teaching recount text for senior high school students. Journal of English Language Teaching, 8(2), 296–304. Meylinda, L. (2023). Exploring students’ perception on online EFL learning through WhatsApp application. English Learning Innovation, 4(1), 12–21. https://doi.org/10.22219/englie.v4i1.23622 Mingsakoon, P., & Srinon, U. (2018). Development of secondary school students’ generic structure execution in personal experience recount writing texts through SFL genre-based approach. Advances in Language and Literary Studies, 9(6), 112–119. https://doi.org/10.7575/aiac.alls.v.9n.6p.112 Nurhidayat, E. F., Apriani, E., & Edy, S. (2021). The analysis of cohesive devices used by tertiary English students in writing English paragraphs. International Journal of Multicultural and Multireligious Understanding, 8(4), 70–81. Novariana, H., & Tarjana, S. S. (2018). Senior high school students’ problems in writing. English Language and Literature International Conference (ELLiC), 2(1), 216–219. Putri, D. N. (2021). Cohesiveness in informal written text, song lyrics. ASELS, 1(1), 18–22. Dewi Nurpitriyani, Didin Nuruddin Hidayat, Nida Husna, & Alek Unveiling coherence: Analyzing grammatical devices in recount text 488 https://doi.org/10.51773/asels2021.v1i1.6 Rudiana. (2021). The realization of grammatical cohesion devices recount text. JALL (Journal of Applied Linguistics and Literacy), 5(1), 101–108. Satria, R., & Handayani, N. D. (2018). Grammatical cohesive devices analysis in descriptive writing by English department students in Putera Batam University. Prosiding Seminar Nasional Ilmu Sosial Dan Teknologi (SNISTEK), 1(1), 145–150. Schiffrin, D., Tannen, D., Hamilton, H. E., & Martin, J. R. (2001). The Handbook of discourse analysis. Blackwell Publishing. Shahzad, S. (2021). Investigation of grammatical cohesion errors in written paragraphs made by ESL students of Khwaja Fareed University of Engineering and Information Technology (KFUEIT). Linguistica Antverpiensia, 1(1), 4996-5005. Sianturi, L. P. D., Silalahi, D. E., & Purba, C. N. (2020). Improving students’ writing ability based on higher order thinking skills (HOTS) questions at 8th grade in SMP Swasta Kartika 1-4 Pematangsiantar. JETAFL (Journal of English Teaching as a Foreign Language), 6(2), 1–17. https://ejournal.uhn.ac.id/index.php/jetafl/article/ view/103 Syam, S. (2020). Reference: A grammatical cohesion in Seloka Journal introduction published by Semarang State University. Journal Polingua: Scientific Journal of Linguistic Literature and Education, 9(1), 5–9. https://doi.org/10.30630/polingua.v9i1.132 Trisnaningrum, Y., Alek, A., & Hidayat, D. N. (2019). Discourse analysis of grammatical cohesion devices in college students’ academic writing essay. IJEE (Indonesian Journal of English Education), 6(1), 79–90. https://doi.org/10.15408/ijee.v6i1.12502 Yule, G. (2019). The study of language. Cambridge University Press. Verdiansyah, Z. M., Sahrudin, & Degeng, P. D. D. (2019). Lexical cohesions in a grade X English textbook by Indonesian Education and Culture Ministry. Diglossia, 11(1), 14–23.