How do EFL university student-writers prepare their draft? An analysis of writing strategy use in EFL writing instruction

The use of writing strategy across writing levels in a virtual EFL writing course was still rarely found in the literature. To fill this gap, therefore, this current research aims at exploring writing strategy in EFL writing instruction. Informed by a research framework of descriptive case study, university student-writers who regularly attended a virtual academic writing course filled out the online writing strategy inventory (Yang & Plakans, 2012) and submitted the argumentative essays. The findings show that the writing strategy was diverse across levels. At the stage of before drafting the essay, all student-writers reread task requirement and understood the type of essay and organization, searched for valid references, and tried to avoid plagiarism. Moreover, student-writers with higher levels also summarized ideas, analyzed sentences and contents, and made writing plan. At drafting the essay, they double checked the requirement, revised the essay, reread the essay and the sample texts, and provided valid arguments. However, student-writer with lowest level was not used those strategies. At the stage of drawn a conclusion that the more writing strategies used during drafting the essay, the more score and the higher quality of the essay was possibly achieved.

write; hence, this may implicate faculties/instructors in creating writing activities in virtual classrooms. Hence, this present research was addressed to answer the following question: what are writing strategies used by the university student-writers across writing levels in drafting and preparing their argumentative essays during EFL virtual writing course? online mapping and online reading. Social media (i.e., Facebook, Twitter) has mediated writing course and activity particularly in teaching mechanics (Altakhaineh & Al-Jallad, 2018) and improved students" motivation. Besides, peer and teacher electronic feedback or e-feedback (Abdullah, Hussin, & Shakir, 2018) has decreased the students" level of writing anxiety, and implementation of automated tracked system of teacher feedback contributed to improve the quality revision and revised texts (Cheng, 2019). Additionally, automated writing evaluation (AWE) tool has enhanced the grammar accuracy on the students" essays on the third composition (Liao, 2016). Though the expansion of technology have provenly brought positive implication in writing performance, some issues related to technological competences of teachers and learners and internet access have become problems for many (Boudjadar, 2015).

Writing strategy in EFL writing instruction
Generally, writing strategy can be defined as "a conscious mental activity, employed in pursuit of a goal, often with an aim to solve a problem in writing" (De Silva, 2015, p. 2). The strategy can be also referred to some techniques used by the learners in creating a writing product and improving its quality (Apridayani, Yungkun, Thoch, & Ruththong, 2021). In a simple manner, Okasha and Hamdi (2014, p. 675) wrote writing strategies as "ways of controlling writing process to produce wellorganized production crystallized by high quality". The strategies can be grouped into direct strategy and indirect strategy (Oxford, 1990). To be more specific, O"Malley and Chamot (1990) and Apridayani, et al. (2021) classified writing strategies into four categories: cognitive, metacognitive, social, and affective strategies. These then have gained much attention on discussions and research in the field of writing instruction.  Apridayani, et al. , (2021), O'Malley andChamot (1990) Numerous studies of writing strategies have been extensively found, particularly looking at how its effect on writing achievement and other writing attributes. The study of Saudi EFL students" writing strategies found that students with low anxiety are more users of writing strategies than the students with high level (Asmari, 2013). The dimensions of writing strategies cover before starting the writing, during the writing, and during revision. Moreover, the investigation by Mahnam and Nejadansari (2012) to twenty three adult EFL learners in Iran showed that pre-writing strategies give a significant effect on the students" writing achievement. The writing strategies the students chose for their writing were using relevant texts, negotiating of topics, making concept map. To more specific on concept mapping as pre-writing strategy in a focused instructional strategy, it improves the quality of students" argumentative essays in the aspects of point of view, unity, coherence, development, organization, and thinking (Al-Shaer, 2014). Moreover, the instruction of strategy gives positive impact on writing performance (De Silva, 2015) and writing attitude (Okasha & Hamdi, 2014), particularly at pre-writing stage in which focuses on developing main concepts, logical connection among ideas and writing organization (Al-Shaer, 2014).
Apart from its benefits in writing achievement, the use of writing strategies towards other learner attributes such as gender and writing ability anchored the contrasting results. Brainstorming as pre-writing strategy gives no significant differences on its sub-categories: listing, outlining, and question and answer among males and females (Hashempour, Rostampour, & Behjat, 2015). In almost a similar fashion, study by Apridayani et al. (2021) concluded that, though most university Thai EFL students applied cognitive strategies during writing, there was no significant differences among the low and high writing achievers. Additionally, despite its positive relationships among writing strategies, it was no significant relations found on the use of writing strategies and their writing achievement.

Method
This current study was guided by a descriptive case study research paradigm (Nunan, 2013), which demonstrates an in-depth portrayal of a specific bounded phenomenon using different data sources in specific circumstances (Casanave, 2015;Duff, 2020). This study also looked into the writing strategies utilized by EFL university student-writers in a virtual writing course. To be more specific, this current investigation described the writers" strategy in drafting, composing, and completing argumentative essay in three dimensions: before, while, and after drafting.
Research site was at the academic writing course in English Education Department Postgraduate Program at a private university in Jakarta, Indonesia. The writing course lasted 1.5 hour-synchronous sessions and 1-hour asynchronous session once a week for 14 weeks and aimed at enabling the studentwriters to gain competences in developing and creating several essay genres. The participant pool was 70 university student-writers from the first semester who registered and regularly attended the course. Until the end of the course, however, only 38 participants filled out the online writing strategy inventory as well as submitted the essays and completed the task requirements on Google Classroom. They were 23 females and 15 males with the age ranged between 22-49 years old. They speak Bahasa Indonesia as their first/second language.

Figure 1. Writing task on a virtual class
For data collection, multiple data sources were carried out to get a deeper understanding on what writing strategies the participants used in preparing and drafting the argumentative essays, , and improving its quality. A writing test was assigned to develop one argumentative essay with a given topic as one of writing tasks during the course (see figure 1). For scoring and grading, scoring rubric for essays from Oshima and Hogue (2006) was used to assess the students" essays in five criteria: format, mechanics, content, organization, grammar and sentence structure. The scores were then consulted to CEFR level default scores (Jiménez-Muñoz, 2014). Moreover, the scale of reports and essays from CEFR (Council of Europe, 2020) was adopted to confirm the leveling and scoring of the students" essays and the scores.
To obtain the data of the writing strategy use, online inventory was distributed to the participants after essay submission. The inventory was modified from Strategy Inventory for Integrated Writing (Yang & Plakans, 2012) that covers three dimensions of drafting: before drafting (16 statements), while drafting (14 statements), and after drafting (6 statements). Furthermore, to confirm data trustworthiness and consistency, an online questionnaire was to gather the participants" perceptions and strategy in developing argumentative essay In a phase of data analysis, the gathered data from writing test and inventory were divided into several levels based on the CEFR levels (A1-C2) and the writers" strategy use (before, while, and after). They were then analyzed descriptively in the form of numbers and text referred to the levels and scores of the students" essays.

Findings and discussion
This paper reported writing strategy used by the university student-writers in drafting and preparing their argumentative essays during EFL virtual writing course. Students" essay and responses on the writers" strategy inventory were collected during a virtual academic writing course.

B2.2
Can produce an essay or report which develops an argument systematically with appropriate highlighting of significant points and relevant supporting detail.

B2.1
Can produce an essay or report which develops an argument, giving reasons in support of or against a particular point of view and explaining the advantages and disadvantages of various options. Can synthesize information and arguments from a number of sources.

1
From the student-writers" essays, it was found that their CEFR levels were ranged from B2.1 to C1.2 and the scores were between 60 and 90 (see Table 2). One student was on level B2.1 and gained the lowest scores of all. However, the essays were mostly on B2 and C1 levels. It may indicate that the most of student-writers can produce argumentative essays with appropriate organization and solid supportive arguments.
Concerning to the use of writing strategy, it was divided into three dimensions: before drafting, while drafting, and after drafting. At the stage of before drafting the essay, student-writers in all levels of writing applied rereading the task requirement, trying to understand the type of essay and its organization, and searching for valid references and evidence with 100% percentage. Meanwhile, student-writers with level C1.2 learned much from sample texts and lecture by summarizing ideas, analyzing sentences and contents as well as planning their writing with 100% on percentage. Moreover, student-writers in all levels were found to avoid copy good sentences from the sample texts and lecture (see Table 3). It may indicate that they try not to fall into plagiarism due to maintaining the quality of their essays. At the stage of drafting the essay, the majority student-writers with level C1.1 and C1.2 admitted that they double checked their writing in the terms of task requirements, the use of sentences and paraphrases, rereading sample texts and what they have written, trying to give solid, valid arguments, and revising it over time. With the lowest percentage, only 65%-67% of them showed that they write the first draft and some ideas from what they read on writing template. On the other hand, student-writer with level B2.1 stated that some strategies were not applied when drafting the essay, such as writing some phrases, providing references and solid argument, and rereading what have written. At the stage of after drafting the essay, it should be a moment before task submission and the last chance to use writing strategies to improve the quality of their essay. From the inventory, it showed that student-writers in three levels (B2.2, C1.1, and C1.2) reported that they use some strategies in this stage with percentage between 93%-100%. They stated that they checked the evidence on the essay, reread the content and grammar, made some changes to ensure of no plagiarism. Moreover, not all of them admitted adding new points based on the sample texts, references, and the lecture on their essays, with percentage 93%, 86%, and 78% respectively. In contrast, the studentwriters with the lowest level showed to tick "No" to several statement. It indicated that some strategies were not conducted, for example checking evidence, rereading the essay, and making changes on the content. As a writing learner, student-writer applies some techniques when composing a text to improve writing skill and to produce high quality texts. The findings in the present study demonstrated that at the stage of before drafting the essay, all studentwriters reread task requirement and searched for valid references. Moreover, level C1.2 student-writers also summarized ideas, analyzed sentences and contents, and made writing plan. These are consistent to the result of the study by Apridayani et al., (2021). They found that Thai EFL writing learners apply cognitive strategies (i.e., use of linking words to ensure logical relationships between sentences/paragraphs) and metacognitive strategies (i.e., reviewing task requirement, looking at writing samples, writing down ideas before writing, and revising two or more times) in high levels of frequency use. furthermore, that a writing plan become one of writing strategies used by studentwriters in this research also supports the previous study by (Al-Shaer, 2014) in which demonstrated that mapping the topic as pre-writing strategy has helped the studentwriters focus on concepts, logical ideas and organization of the essay.
Moreover, on students" essay of level C1.1 and C1.2, the student-writer in the current study mostly use writing strategy, such as double checking the requirement, revising the essay, rereading the essay and the sample texts, and providing valid arguments when drafting the essay. In contrast, student-writer with lowest level was found not used those strategies. At the of after drafting, student-writers in all levels applied several writing strategies, for instance checking the evidence, rereading the content and grammar, making some changes to ensure of no plagiarism. However, not all of them added new points from sample texts, references, and lecture. To the studentwriter with the lowest level, several writing strategies were not used, such as checking evidence, rereading the essay, and making changes on the content. It is in the line with what is stated by De Silva (2015). The instruction of writing strategy gives positive effect on writing performance, and the students could be trained to use writing strategy effectively.
From the present study, it was found that the use of writing strategy was various across different levels in which tend to the more the use of writing strategy, the higher level of the quality of the students" essays. However, we can see that the findings from previous different research were poles apart. A study shows no significant differences between writing strategy by the low and the high writing achievers; meanwhile some another study proved that students using heavily on metacognitive strategy (i.e., planning, monitoring the writing progress) performed better than those who relied on cognitive strategy (Apridayani et al., 2021).
Moreover, on students" essay of level C1.1 and C1.2, the student-writer in the current study mostly use writing strategy, such as double checking the requirement, revising the essay, rereading the essay and the sample texts, and providing valid arguments when drafting the essay. In contrast, student-writer with lowest level was found not used those strategies. At the of after drafting, student-writers in all levels applied several writing strategies, for instance checking the evidence, rereading the content and grammar, making some changes to ensure of no plagiarism. However, not all of them added new points from sample texts, references, and lecture. To the studentwriter with the lowest level, several writing strategies were not used, such as checking evidence, rereading the essay, and making changes on the content.

Conclusion
This current research explored the writing strategy used by EFL university student-writers in composing argumentative essay in a virtual academic writing course. The use of strategy was divided into three stages: before, while, and after drafting. Meanwhile, the student-writers" product essays were categorized into four levels. The findings show that before drafting the essay, all student-writers reread task requirement and understood the type of essay and organization, searched for valid references, and tried to avoid plagiarism. At this stage, level C1.2 student-writers also summarized ideas, analyzed sentences and contents, and made writing plan.
The writing strategy at the stage of drafting the essay was various across levels. At level C1.1 and C1.2, double checking the requirement, revising the essay, rereading the essay and the sample texts, and providing valid arguments were the most used strategies when drafting the essay. In contrast, student-writer with lowest level was found not used those strategies. At the final stage or after drafting, student-writers in all levels applied several writing strategies, for instance checking the evidence, rereading the content and grammar, making some changes to ensure of no plagiarism. However, not all of them added new points from sample texts, references, and lecture. To the student-writer with the lowest level, several writing strategies were not used, such as checking evidence, rereading the essay, and making changes on the content.
The findings may show that higher level student-writers have more writing awareness to apply writing strategy during drafting their essays. It is believed that the more writing strategies used during drafting the essay, the more score and the higher quality of the essay was possibly achieved. Therefore, it is highly advisable for writing faculties/instructors to design an instruction of writing strategy across levels. Eventually, for further research, the investigations on specific writing strategy for specific text type and its effectiveness may lead to clearer understanding in the field of writing strategy and its implementation.