How do Indonesian EFL Students’ Writing Strategies and Writing Process Differ from English L1 Students?

This research aimed to investigate the writing strategies applied by English as a Foreign Language (EFL) students in Indonesia. It also investigated how the writing process undergone by Indonesian EFL students differs from students using English as their first language (L1). The writing strategies questionnaire from Petrić and Czárl (2003) was used to find out the strategies used by the EFL students. In addition, the adult writing process model of Hayes (2012), which was constructed based on the writing process of English as first language students, was used as the framework to investigate the writing process applied by Indonesian students. The participants were 135 English Education department students from two Indonesian universities. Participants’ answers to the questions in the questionnaire over the three stages of writing (pre-writing, drafting, and revising) were analysed using a 5-point Likert scale. The result revealed that the participants applied writing strategies at a moderate level. They only frequently used 15 of the 38 strategies provided


INTRODUCTION
Writing is considered a difficult skill, be it written in a first language and even more, in another language, such as English. The difficulty is especially faced by undergraduates in Indonesia majoring in English Education. This is because as students in the department, they have to use English to answer written examinations, and write assignments. Moreover, as part of their graduation requirements, they have to produce an undergraduate thesis in English at the end of their studies. The challenge is made even more difficult by the fact that English is a foreign language in Indonesia, where it is only used in English lessons or at international events (Fajrina, 2022).
Research has indicated significant differences in the writing process between writing in English as a first language (L1) and as a second language (L2). This includes differences in applying writing strategies in three writing stages: planning, transcribing, and reviewing (Silva, 1993). EFL students have demonstrated a greater application of strategies during the drafting stage than during the prewriting or revision stages (Chen, 2011;Maarof & Murat, 2013;Xiao, 2016). However, another study of 50 university students in Malaysia revealed that the strategies were used more frequently during the planning stage than throughout the writing and revising stages (Aluemalai & Maniam, 2020).
Turning to the Indonesian context, a similar study was conducted on freshmen majoring in English Education (Dari et al., 2022). Using the same questionnaire (that of Petrić & Czárl, 2003) utilized in the research mentioned previously, Dari et al. (2022) found that the students employed more strategies during the writing stage than strategies in the prewriting and revising stages. Another study in Indonesia was conducted to investigate how graduate students used writing strategies while composing their work by using think-aloud protocols (Arifin, 2017). According to these data, skilled students used writing strategies more frequently and repeatedly than their less skilled peers.
Regarding the types of strategies used, high-level proficiency university students in Malaysia utilize more metacognitive, cognitive, affective, and effort regulation strategies compared to low-level proficiency participants (Raoofi, et al., 2017). While in Indonesia, Ardila (2020) revealed that compensation and affective strategies are the two most commonly used strategies by Indonesian students.
A search for studies of the writing strategies used by native speakers of English revealed one study conducted in England by Abdul-Rahman (2011). She compared the writing strategies of three groups of university students: British students who were English first language (L1) speakers, versus Libyan and Chinese students who used English as their foreign language (EFL). The findings showed that the three groups of participants utilized similar strategies. However, the L1 students approached the strategies differently to that their EFL peers. The different approaches in applying the writing strategies indicated the different processes conducted by the participants during the process.
The aforementioned studies looked at the stage of writing where the writing strategies were frequently used (Aluemalai & Maniam, 2020;Arifin, 2017;Chen, 2011;Maarof & Murat, 2013), and the most common types of writing strategies employed (Ardila, 2020;Raoofi, et al., 2017). However, no study has been conducted yet with participants in Banda Aceh, Indonesia. Moreover, there has not been a study that compared the strategies employed by English L1 and Indonesian EFL students. This study intended to fill in the gap by examining how the writing processes and strategies used by EFL students in Indonesia differ from those used by students who speak English as their first language. The research questions for this study are: 1. What are writing strategies frequently used by Indonesian EFL students? 2. What are the types of writing strategies frequently used? 3. How does the writing process applied by Indonesian EFL students differ from that of students of English native speakers?

Writing Strategies
Language learning strategies are defined as "specific actions taken by the learner to make learning easier, faster, more enjoyable, more self-directed, more effective, and more transferable to new situations" (Oxford, 1990, p. 8). Moreover, Abas and Aziz (2018) stated that learning strategies are actions employed by second language students during the process of producing a text. Writing strategy is also defined as "the sequence in which a writer engages in planning, composing, revising and other writing-related activities" (Torrance et al., 2000, p. 83).
Second language acquisition (SLA) academics like O'Malley and Chamot (1990) and Oxford (1990) have categorized strategies for language learning. The goals of the strategies are to increase the conversational skills of students whose English is their additional language (Oxford, 1990). According to Oxford (1990), the majority of learning strategies are appropriate for use in both domains, even if there are distinctions between learning L1 and additional language and it is indisputable that some learning strategies work better in one environment than another. Oxford's (1990) classifications of writing strategies are used in this study because it is well documented in the literature (Tzivinikou et al., 2021). The strategies were classified as direct or indirect. Memory strategies, cognitive strategies, and compensating strategies are all covered by direct strategies. Memory strategies are utilized "for remembering and retrieving new information" (p. 14), which can be accomplished by memorizing an abstract word using locations. Cognitive strategies can help with "understanding and producing language" (p. 14). Compensation strategies are helpful to assist students to fill in a hole in the information flow (Oxford, 1990). Fajrina,J. Everatt,J. Fletcher,C. Astall & A. Sadeghi,How do Indonesian EFL students' writing strategies and writing process differ from English L1 students? | 910 In contrast, indirect strategies are classified into metacognitive, affective, and social strategies. Metacognitive strategies are those employed "to coordinate the learning process" (Oxford, 1990, p. 15). Metacognitive strategies are "students' global skills and knowledge about cognition for helping them raise their self-awareness, direct their own learning, and monitor their own progress" (Larenas et al., 2017, p. 89). Affective strategies are employed "to regulate emotions" (Oxford, 1990, p. 15). Social strategies are learning strategies that include collaborating with peers (Oxford, 1990). This research used writing strategies proposed by Oxford (1990) to examine the various types of writing strategies utilized by the students.

EFL Students' Use of Writing Strategies
Second language (L2) and foreign language (FL) learners' application of writing strategies was indicated to differ in several ways from that of native English speakers. When writing in a first language, writers have readily available words in their vocabulary repertoire and are able to use grammatical structures in an automatized way. However, writing in a second language requires writers to be linguistically educated and competent in the target language (Schoonen et al., 2003). Other than linguistic knowledge, writing strategies also influence the quality of academic writing (Wong, 2012).
A study in Malaysia indicated that proficient ESL students employ the same strategies as their less proficient counterparts. The primary difference between the two groups was the frequency with which the proficient ESL students used the strategies as opposed to the less proficient ones (Abdullah et al., 2011). Another study of Malaysian university students discovered that proficient students used more metacognitive strategies than their less proficient peers (Raoofi et al., 2014).
Comparing the writing strategies applied by EFL and L1 students, Abdul-Rahman's (2011) study found two groups of students used three strategies differently: organization strategy during the prewriting stage, content strategy, and mechanical approach during the revising and editing stage. For example, in the planning stage, when EFL students require additional explanation on a topic, they will discuss it in their study group, whereas L1 students will seek clarification from their supervisor.
This study investigated the frequency and the types of writing strategies used by EFL students in Indonesia regardless of the differences in participants' proficiency in English. By referring to Hayes' (2012) writing model, this study also investigated how the writing process of Indonesian EFL students differs from those of students of English native speakers.

Models of the Writing Process
Numerous writing models have been created to learn in depth the actions and mental processes of writers. One of the first attempts was the model created by Flower and Hayes (1981), which has primarily been used to identify the writing strategies used by first-language students (Sevgi, 2016). The most recent model for explaining adult writing was developed by Hayes (2012) (see Figure 1) which he claimed to be the most complete model. In contrast to Flower and Hayes's (1981) model, this one incorporates additional elements that affect the writing process at the control, process, and resource phases.  (Hayes, 2012) The control phase covers motivation, goal-setting, the current plan, and writing schema. A comprehensive explanation of the model is given by Hayes and Olinghouse (2015, pp. 481-487). Motivation is necessary to ensure the writer completes the writing piece. Setting a goal for the text helps the writer stay motivated. The goal can be to persuade the readers about anything or to debate, describe, narrate, or depict. The goal of the writer will affect how the content is shaped. The main goal, known as the current plan, will be broken down into achievable subgoals. The writing schema includes an understanding of the text's qualities as well as the writing strategies employed. The instructions received, life experiences, and cultural upbringing all have an effect on the way a piece is organized. These elements: motivation, goal-setting, the present plan, and writing schema, which is found in the control phase of the model, influence and direct the composing activity.
The process phase covers both the task environment and the writing process. The writing process -or the writers' thought process -begins when topics are submitted. An outline is then written on a sheet of paper to express this concept. Following that, the concepts are written down as a writing draft during the transcription phase. Any time during the writing process, a piece of writing can be assessed. The evaluator is free to change the suggested idea, the translator's outline, and the transcription. Activities linked to review or planning are excluded at this level. The reason for this is that the writing process is recursive and non-linear, requiring planning and editing.
The task environment of a writer includes social and physical factors that affect the writer's surroundings and the writing process. Social components might take the form of peer and instructor comments and ideas, as well as criticism and collaboration. These suggestions and criticisms offer the proposer information during the writing phase that aids in the concept's development. Physical elements include things like written plans, transcription equipment, and previously written content. Any references a writer makes while drafting a suggestion are considered task materials. Brief notes from the written plans are used to generate writing subject ideas or writing outlines, which the translator then puts into a draft. During the writing process, the transcriber employs transcribing technology, such as typing the written content on a computer. Later, the transcriber went back and revised the earlier text if necessary or to come up with ideas for writing.
Four elements at the resource phase: reading, working memory, attention, and long-term memory, are utilized for communicating, making decisions, and solving problems in addition to writing. The writer must pay attention to maintain concentration during the composing activity. Long-term memory refers to the writer's ability how to compose writing. These could include linguistic abilities as well as relevant historical details and current events. The fluency and quality of the writing will therefore depend on the writer's general familiarity with the subject and experience with the language used in writing. Working memory was used during the writing process to organize the information stored in long-term memory. Reading as a resource can involve reading original writing or works by other authors. Reading other writers' works is done with the intention of using them as references, whilst reading one's own previously written work is crucial for editing or continuation procedures.

Participants
The data was collected from 135 participants (123 females and 12 males) enrolled in two public universities in Banda Aceh, Indonesia. The participants were all English Education majors.

Instruments
The writing strategies questionnaire of Petrić and Czárl (2003) was adapted for this study. The questionnaire was created to examine the "self-reported writing strategies of a large number of non-native English speakers who write in English for academic purposes" (p. 188). The cognitive model of first language processing by Flower and Hayes (1981) served as the foundation for the questionnaire. It has been extensively used to look into the strategies second language learners employ to carry out their writing assignments (Al Kubaidi, 2014;Bai et al., 2014;Dari et al., 2022;Maarof & Murat, 2013;Xiao, 2016;Zhang, 2015). In this study, the questionnaire captured students' opinions about the strategies they employed when completing their writing in English, so was not limited to the writing tasks participants did for this study.
Two sections are included in the questionnaire: the first section contains information about the characteristics of the participants. Six questions made up the first part of the questionnaire: gender, native language, number of years spent learning English, participation in an English writing course before enrolling in the university, types of texts they typically write in English, and level of enjoyment of writing in English. Students' strategies for writing texts were included in the second part of the questionnaire. This section was split into three parts: prewriting (8 items), drafting (14 items), and revising (16 items). Types of writing strategies are divided into direct and indirect strategies (Oxford, 1990).

Data Analysis Procedures
To answer the first and the second research questions, the students were instructed to rate each statement in the questionnaire on a 5-point Likert scale, with 1 being 'never use', 2 being 'usually not use', 3 being 'sometimes use', 4 being 'usually use', and 5 being 'always use'. Oxford (1990, p. 300) divided the frequency of use of the strategies into three groups: (i) low, which includes statements that are rarely or never used (between 1.0 and 1.4) and statements that are usually not used (between 1.5 and 2.4), (ii) moderate, which includes statements that are sometimes used (between 2.5 and 3.4), and (iii) high, which includes statements that are usually used (between 3.5 and 4.4) and statements that are always or nearly always used (between 4.5 and 5.0). Analyses of the students' answers to the questionnaires were performed using SPSS version 25.0.
Participants' answers to the high frequency used strategies were analysed to investigate the stage of writing where the writing strategies were frequently used, and the most common types of writing strategies used as classified by Oxford (1990). Participants' responses to the questionnaire should also indicate the writing process they went through. The findings were then compared to Hayes' writing model which provides a basis for understanding first-language students' writing process (see Hayes, 2012;Hayes & Olinghouse, 2015), and to Abdul-Rahman's (2011) study on the differences in the writing strategies used between EFL and L1 students.

Participants' Frequency of Writing Strategies Used
To find the answer to the first research question, 'what are writing strategies frequently used by Indonesian EFL students?' the mean and standard deviations (SD) for reported strategy use at each stage of writing are displayed in Table 1. The mean for all responses to the items in the questionnaire was 3.34 and SD was .438. Based on Oxford (1990) (see sub-section 3.3 on classifications of frequency levels), this demonstrates that overall, the strategies were utilized at a medium frequency (2.5 to 3.4). The strategies in the drafting stage were the most utilized with M=3.67, but at a moderate level in the prewriting stage (M=3.32) and revising stage (M=3.03).

Prewriting
The participants' answers to the prewriting strategies are displayed in Table 2. Participants' choice of strategies, based on the number and percentage answering each item option, as well as the mean and standard deviation of option choices, are included. High response levels were related to 48 participants (35.6%) indicating that they always made an effort to be knowledgeable of the requirements. This finding implies that participants appreciated the knowledge of the essay's requirements, which may include being aware of the topic and word limit. Some 49 participants (36.3%) used a written sample produced by a native speaker or a more experienced writer as a guide. This implies that participants believed using a writing model will improve their writing.
Furthermore, 36 participants (26.7%) indicated that they always employed strategy number six, which involves taking brief notes and noting important terms connected to the topic. This high-use strategy should assist participants in maintaining concentration and limiting their writing to the current topic. To conclude, the participants reported applying three strategies frequently and the remaining five strategies only occasionally. The data from the prewriting stage revealed a relatively low level of organizational strategies. Participants' response to the first strategy, making a schedule for the writing process, showed that they tended to lean toward the never use side of the scale (approximately 24% of students chose never or usually not use for each).
The Libyan students in Abdul-Rahman's (2011) study also indicated a lack of organizational strategies in the planning stage. In contrast, British students prefer to have a timetable to assist them in organizing their work. Based on her interview with the students, Abdul-Rahman (2011) concluded that students' educational background, L2 writing instruction, and feedback they received in the classroom may influence the students in employing such strategies. Language and Education, 10(2), 907-925, 2023

Drafting
The strategies employed during the drafting stage are provided in Table 3. The mean, standard deviation, the number of participants, and the percentage of the responses are displayed in the table. At this writing stage, participants utilized nine high-level strategies. For strategy number one, 51 participants (37.8%) said they always started with an introduction. This shows that participants were mindful that the introduction should give readers information about the essay's topic. According to strategies 2, 3, and 4, participants experienced difficulty with fluency where they usually or always pause whenever they finish writing a sentence or a paragraph, they go back and read it to acquire ideas for how to continue. It can be assumed that the participants performed poorly during the planning stage, in addition to showing that they were attempting to create their writing in a coherent manner. This notion is reinforced by the fact that during the planning stage, the strategy of taking brief notes was utilized frequently, whereas the strategy of writing an outline was only D. Fajrina, J. Everatt, J. Fletcher, C. Astall & A. Sadeghi, How do Indonesian EFL students' writing strategies and writing process differ from English L1 students? | 916 occasionally employed. Hu and Chen (2007) assert that writing a whole phrase rather than just noting terms and brief summaries of the subject are more effective.
The eighth strategy, simplifying a sentence to express their ideas, was always used by 28 of the participants (20.7%) and usually used by another 65 (48.1%) participants. This strategy was the second-most frequently employed strategy when participants found it difficult to continue writing. Most participants (34.8%) claimed they would write in their first language before translating it to English and tried to identify a synonym or related word for the word they need in English (40.0%). The majority of participants also regularly used the dictionary (strategy no. 11). Only 9.6% of participants (97) always used a monolingual dictionary, whereas 57 (42.2%) always used a bilingual dictionary. This type of dictionary is argued to provide students with the definitions they require (Richards & Renandya, 2002). The participants in this study stated they applied both printed and electronic dictionaries, but preferred the latter since it was more practical. The other five strategies were less frequently employed. In this stage of writing, participants indicated more frequent use of strategies compared to the other stages. Abdul-Rahman (2011) found no substantial difference between L1 and EFL students in terms of how they approach this stage of writing.

Revising
The strategies used during the revising phase are displayed in Table 4. As previously, the mean and standard deviation (SD) are shown together with the participants' frequency of response (and percentages) for each option. Only three strategies were found to be applied at a high level of use. The strategy of examining the essay's requirements was usually used by 60 participants (44.4%), while 46 (34.1%) claimed they always use this strategy. This is consistent with the fact that they frequently used strategy no. 2 during the prewriting stage; i.e., being aware of the writing prerequisites before beginning to write. This demonstrated that despite the fact that this strategy only addresses the writing's surface rather than its content, these participants still value paying attention to the requirements for writing. Only 28 participants (20.7%) indicated that they always showed their text to someone and solicited their feedback, but another 42 (31.1%) indicated that they did so frequently.
The participants' final frequently used strategy was to review errors once they had received feedback from their teacher. The majority of participants -55 (40.7%)claimed they always employed this strategy. The last two strategies suggested that participants valued criticism, especially from their lecturers, as a way to enhance their work. However, in most cases, students are required to produce a draft and submit it to the teacher as the final draft. Therefore, students would not have the chance to change their work based on the feedback if the teacher just provided the feedback after returning their paper and not during the writing process.
Ten strategies reached a medium degree of frequency of use during the revising phase. The other three strategies were used at low frequency. To sum up, three strategies at a high level, another three at a low level, and the remaining ten at a moderate level -were used by the participants in this revising stage.
Participants' responses to the strategies they used during the revising stage suggest that these students may put less effort into revising their work. This may also be evident in their responses to questions related to making changes in vocabulary, sentence structure, essay structure, and content. This finding supports Abdul-Rahman's (2011) study where she found that EFL students used strategies related to revising the content of their work less than the L1 students did. The EFL students focus more on the language structure while the L1 students focus more on clarity. Based on the percentages of the writing strategies used, it showed that the participants employed more strategies during the drafting stage (9 strategies) than during the prewriting (3 strategies) and revising stages (3 strategies). Of the 38 strategies, only 15 were frequently used by the participants. Hu and Chen (2007) assert that depending on the need for writing, students' use of strategies will either increase or decrease. This implies that assignment difficulty may have an impact on students' decisions regarding their writing strategy.
The findings of this study are consistent with previous studies (Chen, 2011;Dari et al., 2022;Maarof & Murat, 2013;Xiao, 2016) where the drafting stage was observed to be where the EFL students used writing strategies more frequently. However, caution may still be needed before assuming that all strategies used will be found to be at a moderate level. For example, the writing strategies questionnaire was created in 2003, at a time when students did not have access to resources such as an electronic dictionary, nor to online artificial intelligence tools that can offer lots of assistance in writing in English. Strategy use may increase with such availability.

Participants' Types of Writing Strategies Used
To address the second research question, the types of strategies employed were examined in accordance with Oxford's (1990), 'what are the types of writing strategies frequently used?' The findings indicated that in the prewriting stage, participants employed one cognitive strategy to comprehend and generate the language, as well as two metacognitive strategies that assist to coordinate the learning process. The metacognitive strategies were used in understanding the requirements for the essay and studying the writing sample to learn how to write the essay. The application of these two metacognitive strategies was somewhat consistent with Abdul-Rahman's (2011) findings. She found that only Chinese students outline their work at the planning stage by determining the writing requirements to determine what the tutors actually wish for. Other kinds of outlining were favoured by Libyan and British students. In this stage, students from Libya and China examine a writing sample. The students from China and Libya believed that a writing model assisted them in acquiring the English language's grammatical structure and writing organization. However, it was not mentioned that British students used a writing model. The participants in this study prewrote their work by making brief topic-related notes as a cognitive approach.
The participants employed metacognitive strategies frequently during the writing stage. These included having an introduction and pausing to reread their work after finishing a sentence or a paragraph. These strategies seemed to be employed to generate ideas on how to carry on with their task. Additionally, the students used compensating strategies. According to Oxford (1990), who researched language acquisition processes, intelligent guessing and overcoming obstacles are two examples of compensating strategies. In the current study, participants who had trouble articulating their ideas utilized compensatory strategies by simplifying what they wanted to write. In order to compensate for their inability to recall a term in English, either they would look for an English synonym for the word or write the word down in Indonesian before translating it to English.
The participants also used cognitive methods by consulting a dictionary in addition to metacognitive and compensatory strategies. They claimed that utilizing an electronic bilingual dictionary was preferable to a monolingual version. The lack of a monolingual dictionary may have been due to the fact that the translation may have included a new word, making it even more challenging for the students to comprehend, or it may have been because the students needed to understand the text they were reading right away (Laufer & Aviad, 2006). In contrast, employing a bilingual dictionary spares the learner from learning new words by defining a foreign term using a word or phrase from the native tongue, in this case Indonesian, to illustrate its meaning. Applications for their cell phones, such as U-Dictionary, were cited as the type of dictionary that the participants said they used regularly. They argued that having the dictionary installed on their phone is convenient. In addition to receiving word-byword translations in "a click" utilizing this type of dictionary, users may also access example sentences followed by a direct translation into their native tongue. The sample phrases help the participants comprehend how to use the word properly. This result confirms the findings of Ardila's (2020) study, which showed that students preferred compensatory strategies.
The participants used cognitive strategies during the revision stage by making sure their essays agreed with the specifications and by using the lecturers' input to assess their work. The writing prompt sheet contains the requirements, including the topic for the writing assignment, the word limit, etc. The input from teachers typically focused on the language's structure rather than immediately commenting on the text's content and rhetorical devices. According to a study by Budiharso (2014) on the application of writing strategies by EFL students in Indonesia, when giving feedback, EFL teachers tend to focus more on grammar, mechanics, and mechanical conventions of writing, such as the presence of the main idea, introduction, supporting details, and conclusion. There was no in-depth discussion of the subject or rhetorical strategies. This demonstrated that the teacher gave greater consideration to the language's structure and the presence of the aspects needed to write a text. Contrarily, the text's subject matter and the way the students presented it (rhetorical strategies), which might be impacted by cultural elements or educational experiences, were not discussed.
The students also used a social strategy during the revision stage by presenting their work to their peers and requesting feedback or assistance. In a related study, Abdullah et al. (2011) discovered that students communicated with one another by asking for an English word's translation. The research found that students not only consult dictionaries for the word's translation but also ask their friends for assistance.
To sum up, metacognitive strategies were employed more frequently in this study than other strategies: two were used during the prewriting stage, four during the drafting stage, and another two during the revising stage. This result supports the previous finding by Raoofi et al. (2014) who found that skilled students prefer to utilize metacognitive strategies, followed by cognitive, compensation, and social strategy. In a different study, Raoofi et al. (2017) found that students employed more metacognitive, cognitive, affective, and effort regulation strategies than other writing strategies.

Indonesian EFL Students' Writing Process through The Lens of Hayes'
Writing Model (2012) The third research question, 'how does the writing process applied by Indonesian EFL students differ from that of students of English native speakers?' was answered by comparing the participants' responses to the demographic questions in the first part of the questionnaire and their responses to the writing strategies questionnaire and Hayes's (2012) writing model of the writing process. Participants' responses should reflect their writing process as EFL students. On the other hand, Hayes's (2012) writing model, which was created based on the writing process of L1 students, should reflect the writing process conducted by L1 students.
The background questionnaire responses revealed that the sample had a different proportion of male and female students (123 females and 12 males). In the English Education Departments of the universities that took part in the study, female students outnumbered male students by a ratio of around 75%: 25%. All participants acknowledged having Bahasa Indonesia or Indonesian as their national language. However, 64.4% of students said Indonesian was their second language because their regional language was their first language. The participants were in their fifth, seventh, and ninth semesters of study. Prior to enrolling in college, each participant had completed six years of English coursework in high school, consistent with the high school requirements in Indonesia. Participants admitted that they frequently wrote quick notes for assignments, and 80% of students said they enjoyed writing in English because it assisted them to get better in their language skills. Others, on the other hand, had trouble understanding English, lacked English vocabulary, or only produced essays in English since it was required as homework for writing classes. Through the lens of Hayes' writing model, this study examined how EFL students utilized writing strategies as presented in Figure 2. The demographic responses from the participants revealed that external motivation influenced them more than internal motivation. This was demonstrated by the participants' writings, which were primarily academic in nature (writing essays for school, for example), and less creative (writing stories for enjoyment).
Participants were assisted in formulating task goals by the frequent use of the strategy during the prewriting stage of being aware of the work requirements. The task's conditions, which included the fact that it had to be finished in a certain amount of time and met the limited word count, were also made clear to the participants. This strategy is implemented at the control level and meets the organizational requirements for writing tasks.
Participants reported that they were less prepared during the prewriting stage since they did not set a schedule to complete their work, and only had a brief plan in mind or some basic topic-related notes before starting to work on their current plan. This result validates Abas and Aziz's (2018) study on postgraduate English Language Studies majors. Participants in their study admitted to having some ideas about how their work should be organized but did not write them down. In contrast, the L1 students in Abdul-Rahman's (2011) study was well-prepared by having a timetable to help them organize their writing schema.
The EFL students in this present study reported utilizing a writing model because it offers a sample of the required work. The sample was used as one of the work materials. The EFL students in Abdul-Rahman's (2011) study argued that a sample assignment assisted them in developing similar writing structure plans, and to develop writing schemas as a result. In contrast, L1 students did not report using a writing sample in producing their work.
The students in this study acted as proposers by drafting their work based on the task materials during the writing phase. The EFL participants in this study indicated that during this writing process, they would probably think and sometimes write in Indonesian before translating it into English. Abdul-Rahman's (2011) EFL students employed an identical strategy by thinking and creating sentences in their minds using their native tongues before translating them using written English. Conversely, L1 students used spoken English to formulate their ideas and later describe those ideas in written English.
In this writing phase, the EFL students reported frequently stopping to revise the content that had already been written. A similar strategy was also applied by the students in Abas and Aziz's (2018) study, where the EFL students frequently stopped to gain more ideas before continue writing. The participants took on the role of transcribers by applying writing conventions and creating a written text. The transcribers managed the information needed to complete the task in working memory at this stage. Only sometimes they acted as proposers and evaluators to modify the original concepts.
The task environment is made up of both the social and physical task contexts. Asking for comments or criticism from peers on a task that had already been completed was the extent of the social task environment. However, in Indonesian classrooms, asking a friend for help with the assignment would be considered cheating (Fajrina, 2022) as indicated by the moderate use of the strategy request assistance from peers (see Table 3). Because of this, the figure does not depict a collaborative role as Hayes's (2012) writing model does.
Writing models and prompts were employed as task materials in this study. A writing model was also used by EFL students in Abdul-Rahman's (2011) study to compose their work. Participants in Abas and Aziz's (2018) study claimed, however, instead of using a writing model, they would use their devices, such as cell phones and laptops, to look for assistance, for example, to gather ideas before continuing to write or look up suitable words using an online dictionary. For transcribing, several participants used laptop computers. Additionally, they used a bilingual dictionary installed in their smartphone because it is useful and provides a clear translation rapidly. They also acted as evaluators by proofreading their work for errors before submitting it to their teacher.
Participants paid attention to the essay's prerequisites in the planning and revising process as indicated by high frequencies of the strategies aware of the essay's prerequisites in both phases. They also paid attention to their work during the writing process by pausing several times either after writing a sentence or a paragraph. Students were indicated to take benefit from their long-term memory and reading at the resource level. This was indicated by the high level of applying the strategies to pause several times and reread their work during the writing process to gain more ideas on how to continue writing their work. Working memory contained the knowledge needed to finish the writing assignment. Those resources are used in the activities of proposing, translating, evaluating, and transcribing (Hayes & Olinghouse, 2015).
The writing process model in Figure 2 is slightly different from Hayes's (2012) writing model (see Figure 1). This is due to the fact that although ESL/EFL scholars have used Hayes's (2012) writing model throughout time to analyse the writing processes of ESL/EFL students, its original intent was to investigate the writing processes of students who were native English speakers. Therefore, it is natural that participants' writing processes diverge slightly from those of students whose first language is English. As Abdul-Rahman (2011) claimed that the inconsistencies in approach when putting the ideas into practice could happen due to variations in classroom instruction, sociocultural difficulties, or other culturally specific factors.

CONCLUSION
The findings indicated that the participants did use a variety of writing strategies, including metacognitive, compensatory, social, and cognitive ones. However, the students did not yet apply the strategies regularly enough to get the most benefit. This was shown by the general writing strategies that were employed with moderate frequency. They only used strategies at high frequency at the drafting level which supports the findings from previous studies. The findings from this study urged writing teachers to direct their students to apply writing strategies to improve their writing abilities, especially the strategies that the students can get lots of benefit from. According to earlier studies on teaching writing to EFL students, strategies that are explicitly taught to students enhance the quality of their writing.
This study compared and contrasted the application of the writing strategies used between EFL and L1 students. It was revealed that the Indonesian EFL students used writing abilities somewhat differently from the students who were native English speakers. Participants were observed going through the writing process as described in the model, albeit with a few small changes. These differences between the participants in this study and those in Hayes' study and also Abdul-Rahman's (2011) study, who were English native speakers, suggest different approaches to putting writing strategies into practice during the writing process. The education gained in the classroom most likely had an impact on the choice to employ a specific strategy. More research on the elements that influence the choice of writing strategies should be conducted in order to fully comprehend the rationale for employing various strategies.
One of the study's limitations is that it was difficult to determine if the participants' responses to questions about writing strategies were based on what they actually did or what they thought was the best answer. Interviews with the students would have helped researchers to comprehend in-depth their reasons for choosing particular strategies. Think-aloud protocols, in which students are directed to describe the processes done throughout the writing process, can be used as another instrument. The instrument may be able to provide a basis for determining the strategies used in practice though they are not easy to utilize with big numbers of individuals involved as in the present research.

ACKNOWLEDGMENTS
The Indonesia Endowment Fund for Education (LPDP) funded the study presented in this paper.