Mixed-Method Research On EFL Graduate Students’ Academic Writing Practices

This study aims to identify (i) how EFL graduate-level students at various Turkish universities regard the level of difficulty in terms of the different sections of a scholarly work in their academic writing practices, (ii) whether their perceptions concerning the difficulty of the various sections show a significant difference depending on their demographics, (iii) the solutions they employ when they are challenged with difficulties in academic writing and (iv) their views about the process of academic writing in general. Data from 34 graduate EFL students were reported. The study adopted a mixed-method research design, and the data were collected with Academic Literacies Questionnaire (ALQ) (Chang, 2006; Evans & Green, 2007). The participants also responded to open-ended questions about the challenges they face in academic writing and their solutions. The results revealed that EFL graduate students had problems with academic conventions, and found expressing themselves succinctly problematic. However, they were familiar with the mechanics of the target language.

However, this kind of 'ad hoc' (Jalongo, Boyer, & Ebbeck, 2014, p. 242) approach is far from ideal in that it is not systematic and inclusive. Institutional support is essential for graduate students to improve their academic writing skills through skilled support from knowledgeable authorities (Kamler, 2008). Kamler and Thomson (2006) also suggest that academic writing and writing for publication should be an integral part of the curriculum for postgraduate level students. Considering that only a few studies explored academic writing practices and the difficulties learners experience in Turkish contexts, the study will further investigate the underlying integral components of academic writing.
Writing in a specific discipline mandates an active, comprehensive, and demanding engagement with the principles and facts of that specific discipline (Rose, 1985). Supporting this view, Hyland (2007) points out that academic writing requires the kind of thinking, particularly in academic literacy, which requires students to have the ability to synthesize ideas and sustain arguments to be able to write in English for academic purposes and these skills are deemed to be crucial for academic success.
Non-native English speakers may face grammar, lexis, and syntax difficulties when writing at the university level (Rose, 1985). With academic norms, English academic rhetoric, defending or disputing arguments, addressing the audience, and making scholarly writing cohesive and coherent, academic writing may be difficult for students (Belcher, 1994). Thus, academic writing for graduate students at the tertiary level requires graduate-level academic literacy and a sophisticated way of thinking. In this sense, Bronson (2004) argues that non-native graduate level students may experience problems not only because of the mechanics of academic writing, such as lexis and grammar, but also due to an inadequate grasp of academic writing standards and the failure to understand the expectations of the related institutions or lecturers in any specific field.
Moreover, regarding language difficulty in second language writing, Hinkel (2003) has identified the characteristics of weak writing like overuse of the copula verb be or frequent uses of vague nouns in academic writing of non-native academics/students who write in English. Other kinds of organizational problems, including macro-level of the thesis and micro-level of the paragraph, have also been reported in addition to some other challenges such as surface-level problems including punctuation, mechanics of spelling, bibliographic referencing, and punctuation as well as substantiation of ideas and strength of claims made (Allison, Cooley, Lewkowicz, & Nunan, 1998). To resolve some problems related to academic writing challenges, Swales (1987) and Hirvela (1997) explored the field study approach, which aims to sensitize students in specific genres and aids the teaching of discourse-specific scholarly writing by introducing some strategies used by proficient writers in academic writing. Gosden (1995) also emphasizes that junior scholars can be trained by combining this genre sensitizing approach and self-regulatory strategies, and junior scholars could become independent research article writers.
Graduate students encounter various writing tasks to reach their chosen degrees during their education. Their experiences regarding their academic writing skills differ from one program to another. However, the nature of the tasks is similar in some respects. Firstly, writing tasks become increasingly sophisticated and demanding as graduate students study in their programs. Secondly, the assigned tasks in various programs must be written 'academically' (Swales & Feak, 1994, p. 7). Graduate students must use the type of English when they are writing on academic subjects for their scientific purposes, and the style mandated by this type of writing requires students to use a clear, simple, and precise language when they are supposed to express their opinions, findings, and arguments in this academic style. Research showed that students who participated in an academic writing course that included a variety of writing themes, exercises, and genres improved their writing skills (Rakedzon & Baram-Tsabari, 2017).
The ability to write academically necessitates time and dedication from language learners. Challenges emerge during numerous stages of the academic enculturation process according to research on the particulars of graduate writing experiences. Reading and covering previous research (Kwan, 2009); navigating the process of publication once the writing process has been completed (Badenhorst & Xu, 2016;Casanave & Li, 2015); composing and conducting research in various genres such as theses, proposals, and peer-reviewed articles (Aitchison & Lee, 2006;Feak & Swales, 2009;Hyland, 2015;Negretti, 2017); and writing anxiety and managing work-life balance when writing (Huerta, Goodson, Beigi, & Chlup, 2017) are among the most frequently reported difficulties in the writing process.
In addition, it is a skill that develops as part of graduate students' academic tasks and tertiary level studies as they think, argue, reason, and interact with what has been argued or discovered in their field. Accordingly, they must specialize in the vocabulary and expressions related to their field, the types of texts (e.g., essays, reports, summaries, or research articles) they are required to write and also the structure and organization of these academic texts (Мкртчян, 2020). Smith, Densmore and Lener (2016) also maintain that graduate students should develop their writing skills early in their careers. This development requires them to have a comprehensive understanding of good writing, writing skills methods, and some specific tips on writing specific writing such as entries in research notebooks, reports and research papers, and book reviews. Previous research has primarily focused on EFL students' difficulties in academic writing classes rather than their experiences and expectations and how they handle problems when writing. As a result, the current study used quantitative and qualitative data collection techniques to delve into the experiences of Turkish EFL graduate students in their academic writing classes.

Purpose of the Research
There has been a considerable body of research regarding the challenges posed by academic writing, either from the lecturers' perspectives (Singh, 2019) or undergraduate students' and lecturers' perspectives (Mudawy & Mousa, 2017). Moreover, much of the research concerning academic writing practices have focused on students studying in contexts where English is used as the first language (L1). The purpose of this study is twofold as it focuses on the challenges and suggested solutions to academic writing practices of EFL graduate students in Turkey. The perspectives of EFL graduate students in terms of the challenges they experience during their scholarly endeavors and their practices in overcoming these challenges at different universities within Turkey were investigated. Accordingly, the study addressed the following research questions: 1) How do EFL graduate-level students at various Turkish universities regard the difficulty in the different sections of a scholarly work in their academic writing practices?
2) Do EFL graduate students' perceptions regarding the difficulty of the various sections show a difference depending on their (a) gender, (b) age, (c) level of English, and (d) academic writing level?
3) What are EFL graduate students' general beliefs regarding the process of academic writing? 4) What are the solutions offered by EFL graduate students in terms of the constellation of abilities, understandings, habits, attitudes, and beliefs that help them overcome the challenges of academic writing?

Literature Review
Although numerous studies have been conducted to scrutinize various aspects of academic writing, studies on the substance of postgraduate students' experiences and expectations for academic writing are scarce. It indicates that the relationship between the problem, academic writing, and the experiences and expectations of postgraduate students should be addressed for academics to understand better the entire phenomenon (Esfandiari, Meihami, & Jahani, 2022). Besides, Rose, and McClafferty (2001) suggested more than a decade ago that little effort had been made in the past to handle graduate-level writing teaching in a 'systematic' (p. 27) fashion. This section reviewed and reported the findings of studies that looked into students' academic writing experiences and expectations.
Higher education institutions around the world attach a great deal of importance to reinforcing graduate students' academic writing skills, even in countries where English is used as the first language, and academic writing skills are also deemed to be the critical indicators of superior experience in many postgraduate programs (AlMarwani, 2020). Scholarly works exposed to anonymous peer evaluation serve a crucial role in determining whether or not faculty members will be promoted or tenured. (Luey, 2007;Rocco & Hatcher, 2011). Accordingly, mastery of academic writing is a strong indication of scholarship to the extent that graduate students and faculty members are supposed to publish their works written in English to progress professionally (Kwan, 2010). Researchers studying graduate students' writing challenges have discovered barriers to producing scholarly works in practically every element of their academic performance. Writing for their chosen fields requires individuals to make fundamental alterations in how they approach knowledge, learning, written expression, and themselves before attaining a level of comfort in scholarly writing, as reported in research into what makes writing challenging for graduate students (Ondrusek, 2012).
Graduate students must have high-level academic writing skills since they are expected to integrate convergent ideas and synthesize various perspectives by paying attention to voice, accuracy, and audience. Moreover, graduate students face greater demands in terms of having to cover a broader scope in writing (in terms of breadth and depth) and are able to "integrate disparate ideas, synthesize perspectives, and extend theory",-which requires higher-level construction skills (Lavelle & Bushrow, 2007, p. 809). Moreover, according to Fergie, Beeke, Mc Kenna, and Créme (2011), much study concentrated on writing in higher education contexts in undergraduate work. Nevertheless, there was a minimal focus on graduate-level students' writing experiences, despite PhD writers' recent increased attention.
Casaneve and Hubbard (1992) requested graduate instructors at one university to provide precise information regarding the writing requirements for first-year doctoral students, the criteria they use to grade students' writing, and the writing challenges of native-and normative-English-speaking (NS and NNS, respectively) students. The 85 sets of data represented 28 departments, with the humanities/social science and scientific/technology disciplines approximately evenly split. The survey's findings raised pedagogical concerns about global versus local writing problems, the significance of vocabulary education, the necessity for discipline-specific writing instruction, and the timing of ESL support service writing sessions for graduate students.
Ntereke and Ramoroka (2013) studied the effectiveness of academic writing education from students' perspectives in a mixed-methods study. They studied 46 first-year students at the University of Botswana. The results of their study showed that most studies concentrated on enhancing teaching rather than considering students' perspectives on writing education. As a result, they wanted to find out how students regarded the effectiveness of academic writing in their courses. It was demonstrated through qualitative interviews and questionnaires that students found activities helped them write an essay. The majority of them felt that more time should be spent preparing them for essay writing and that they required more practice exercises, such as gathering and synthesizing material from various sources.
Bair and Mader (2013) undertook a collaborative self-study to determine the source of graduate students' academic writing challenges and find solutions. A group of ten professors from a college of education got together to describe the issue and examine data from faculty and student questionnaires, course papers, course assignments, and course assessments. They discovered differences in teacher and student perceptions of graduate writing preparation and between the espoused and implemented curriculum. Both academics and students recognized problems with synthesizing theory and research. They examined the necessity for teacher-scholars in today's educational climate, curriculum improvement problems, and many program-specific actions to address deficiencies in academic writing and critical thinking.
On the other hand, in a qualitative study, Morton, Storch and Thompson (2015) conducted a case study on the perceptions of three multilingual students on their academic writing. Thus, their research concentrated on students' impressions of what it meant to write academically at university. The study lasted two 12-week semesters and was based on an academic literacies model developed at one of Australia's top universities. The findings of the interviews revealed that students' main disciplines had an impact on their beliefs and attitudes regarding academic writing. Furthermore, the findings pointed to socio-academic relationships, implying that interpersonal writing relationships influence students' academic writing development.
McCarthy and Dempsey (2017) addressed chemistry graduate students' uneven preparation in technical writing and developed a graduate-level course focusing on innovative research proposals in their research. The focus of this course was on developing novel research projects. Extensive group discussions, small-group exercises, and regular in-class small-group peer evaluations were all part of the overall course format. Student surveys, staff feedback, and student success in getting graduate fellowships all showed that this course was a helpful graduate education component since its inception. The researchers also detailed the course format, teaching approach, and course evaluation.
Tremblay-Wrag, Mathieu Chartier, Labonté-Lemoyne, Déri and Gadbois (2021) also conducted a study on graduate writing students in the Canadian context since dropout rates and program lengthening have been alarmingly high in Canada's higher education system. Their findings revealed a unique concept created by Thèsez-vous, a non-profit organization specializing in creating physical and human environments to aid academic writing. Over the last four years, the organization has developed and standardized a writing retreat program for graduate students from various areas and universities around Quebec. The structure and operation of the writing retreats were described, as well as an analysis of the goals: 1) set realistic individual goals for academic writing; 2) find optimal writing conditions, and 3) eliminate isolation. According to decisive findings, the adopted approach delivered positive outcomes in the development of academic writing talents by forming a community of practice during writing retreats and subsequent interaction. This growing network of graduate students represented a significant opportunity, they concluded. Wijaya (2021) studied self-regulated learning strategies in academic writing by English education master students. Fifteen master's students and three interviewees were invited to complete the questionnaire and engage in the interview activities for data collection. The results of the study demonstrated that the majority of English education master students have turned into more lifelong and proficient academic L2 writers, as evidenced by their continued resilience, efforts, and commitment to complete numerous academic writing projects.

Research Design
The current study espoused a mixed-methods research design. In a mixed-methods research design, the researcher adopted a procedure to collect, analyze, and mix quantitative and qualitative methods (Creswell & Plano Clark, 2011). Likewise, the study employed two strands of data, quantitative and qualitative, to further understand the research problem concerning the academic writing difficulties of EFL graduate students and their solutions. As Miles and Huberman (1994, p. 42) state, combining quantitative and qualitative data yield 'a very powerful mix.' Thus, both means of data provided the researcher with the means to follow up the quantitative study with a qualitative one (Creswell, 2002, p. 535).
The study also utilized the convergent parallel design, suggesting that the researcher simultaneously collected both quantitative and qualitative data, merged the findings derived from the data and used the results to have a clear picture of the research problem. Both data sets were analyzed separately, and the results were compared to see if they supported each other or contradicted each other. The results were described separately in the discussion section. The participants' responses to the open-ended questions were included in the design. They offered several advantages in letting the researcher be flexible in the issues to be discussed in the following sections of the paper (Dawson, 2002). Following the quantitative statistical results, qualitative quotes were provided to either 'confirm or disconfirm the statistical results' by employing a thematic analysis strategy (Creswell, 2002, p. 542).

Sampling
34 EFL graduate students participated in the study by responding to items on a scale and open-ended questions delivered through the same questionnaire. Participants were asked to contribute to the research based on the following criteria: (a) respondents must be at the thesis level in master's or PhD programs, and (b) they must be enrolled in the EFL departments at the research site or university.
The demographic characteristics of the participants are indicated in Table 1. As shown below, female participants (23) outnumbered males (11) in the study: they almost doubled males in number. When the age range of the participants was concerned, most of them (10) were distributed between 31-35 range and following this, 9 participants were between 20-25 years old. Almost all participants stated that their level of English was high (advanced 32 participants), and only 2 of them stated that their level of English was upper-intermediate during the time of the study. Among these, almost 59% of them claimed that they had an advanced level of writing competency. In comparison, 29% of them stated their level of writing skill was very good, which suggested that almost all participants were competent in the English language in terms of academic writing, which suggested that they were familiar with the academic writing process.
The study employed a snowball sampling strategy to collect data for the quantitative section. This allowed the researcher to recruit many participants doing their master's or PhD degrees around Turkey. Accordingly, the participants were located in different cities in Turkey throughout the research since they became a member of the research as their colleagues let them know about the research at their universities. To reach a sufficient number of participants, the researcher reached 34 participants by assuming that the sample size was large enough for a correlational study that related variables (Creswell, 2011). The study employed a purposeful sampling snowballing strategy to identify the participants with their defining characteristics for the qualitative section, suggesting that participants were intentionally selected to understand the research phenomenon.

Data Collection Procedure
The instrument was conducted via social media, namely, Google forms, since it is one of the most popular tools to collect data from several participants practically. The web page composed by the researcher was sent to the participants through a link via their WhatsApp or e-mail addresses, and they were informed about the aim of the research prior to answering the questions. Since the study was conducted during the Covid-19 pandemic, collecting data online also let both parties be safe in data collection.
To collect data, the researcher asked participants to identify others who were also doing their graduate studies at the time of the study to become participants of the sample. When the participants had questions, they quickly reached the researcher to clarify their understanding. Consent forms were achieved; the participants voluntarily answered the items on the scale and the open-ended questions and were informed that they had the right to quit the research at any time they wished during the research. The identities of the participants were also kept anonymous. Concerning collecting qualitative data, the researcher avoided the temptation to express her views and stay neutral to prevent bias during the interviews and let the participants be open and express their real feelings.

Data Collection Instrument
To collect data on the academic writing practices of graduate students, the researcher espoused selected items from the Academic Literacies Questionnaire (Chang, 2006;Evans & Green, 2007). In addition to the section which embodies the demographics of the participants, the scale utilized for this research consists of two sections: "Challenges Faced in Academic Writing Practices" (20 items) and "Overcoming the Challenges in Academic Writing Practices" (6 items). The participants were asked to assess the difficulty level of the challenges in their academic writing practices on a scale from 1 (very difficult) to 4 (very easy) for both sections. The reliability analysis revealed relatively high consistency compared to the minimum accepted value of 0.70 (Pallant, 2010). The Cronbach's Alpha reliability of the original scale indicated .903. The reliability analysis was conducted for this study again, and the reliability score was .908, suggesting high internal consistency.
To provide the triangulation of different data sets, the researcher also conducted semi-structured interviews online for the participants to respond (self-prepared open-ended questions) to generate helpful information about their experiences.

Data Analysis
The data derived from the questionnaire were analyzed with Statistical Package for Social Sciences (SPSS, version 26). Kolmogorov-Smirnov and Shapiro Wilk tests were used to test whether the data were distributed normally. Both tests demonstrated that the data showed normal distribution, p>.05 (see Table 2). Frequency counts mean the percentages as shown in the tables, and a series of parametric tests were reported based on the output derived from the statistical software SPSS (version 26). For the qualitative section, however, thematic analysis was used to analyze the responses to the open-ended questions, which aided the researcher in identifying, analyzing, and reporting the findings in the form of codes and themes in full description (Braun & Clarke, 2006). The themes derived from the thematic coding were reported in the results section.

Results
The current study aimed to explore academic writing difficulties that Turkish graduate EFL students face in their academic writing practices and their solutions to those difficulties when writing. The descriptive statistics and inferential statistics were reported under four headings concerning (i) frequency counts concerning the perceived difficulties in EFL graduate students' academic writing practices, (ii) frequency counts showing how the participants overcome the difficulties, and (iii) if the perceived difficulty levels show a change for their demographics and (iv) the responses to the open-ended questions. Scale: 1 = Very Difficult, 2 = Difficult, 3 = Easy, 4 = Very Easy As given in Table 3, writing the methodology section (68%), expressing ideas clearly/logically (68%), summarizing/paraphrasing (68%), proofreading written assignments (76%), planning writing assignments (71%), planning writing assignments (71%) and writing recommendation section (71%) are ranked as the six top difficulty areas in EFL graduate students' academic writing practices (mean range between 3.00 and 3.15). However, the participants rated using appropriate academic style, linking sentences smoothly, revising written work, writing the conclusion, referring to sources and writing references as very easy and easy.
Almost 68% of the participants reported having a very high degree of difficulty in writing the methodology section. They rated writing the discussion section, summarizing/paraphrasing and proofreading the written assignments as very difficult. Besides, the identical items were again reported to be complicated by 30%, 44% and 53%, respectively. Students also reported that proofreading and planning writing assignments and writing implications were considerably tricky (more than 70%), suggesting that most EFL graduate students have difficulties regarding organization and coherence in academic writing. On the other hand, only 2.9% of them rated using appropriate academic style, linking sentences smoothly, revising written work, referring to sources and writing conclusion items as very difficult, which might reveal that they are familiar with the academic writing conventions, and they find these sections of academic writing easy.
On the other hand, more than half of the students reported the following items as easy and very easy: using appropriate academic style, writing findings/analysis section, writing coherent paragraphs, synthesizing information/ideas, linking sentences smoothly, revising written work, writing introductions, referring to sources, writing a conclusion and writing references/bibliography. Table 4 demonstrates the frequency counts of the six different solutions used by the participants to overcome difficulties in their academic writing practices. Depending on the frequency count, the participants revealed that the practical solution they employed is "Be persistent and try to express themselves in different ways" (56.8%), which suggests they look for ways to relate the text to the audience in several ways. Following this, the participants reported to "seek help from other classmates, for example, checking the writing in English" (73.5%) and "Discuss with the lecturer to get information on how to approach assignments" (41.7).
The least popular solution employed by them was "write in their first language (Turkish) and then translate it into English" (2.9%). Moreover, 20.6 of them also reported "take additional writing course(s), and 35.3 of them stated that they "use editors to edit their work".
On the other hand, as an answer to the third research question, to see if the perceived difficulty levels show a difference concerning the participants' (a) gender, (b) age, (c) level of English, (d) academic writing level, the result of the T-test revealed that the perceived difficulty level by the participants does not change by gender (p>0.005, p= .084). ANOVA test also indicated that the perceived difficulty level did not change by age (p>0.05, p=.407). Similarly, the perceived difficulty levels did not show a significant change in their level of English as the One-Way Anova test showed (P=.095) and by writing level as the significance level of the Kruskal-Wallis test revealed p as .188.
Concerning the main points that the participants pay attention to when writing, 'use of academic vocabulary and commonly used phrasal elements' (37.5%), comprehensibility (34.37%), 'use of scientific language' (34.37%) and paraphrasing (28.5%) were believed to be the prominent factors. They also attached importance to reading different resources (18.75%), reliability (18.75%), and the organization of the text (18.75%). Five participants stated replicating previous studies, and five stated the ability to synthesize as essential steps when writing. However, only 2 participants stated that they found making citations important.
Most participants benefited from participating in academic writing courses or receiving online training. P4, for example, noted: 'I took an undergraduate academic writing skills course. I did not take private courses at the master's and doctorate levels, but I can say that my academic writing skills have improved since we had research proposals or article writing assignments in all the courses I took.' P15 also stated: 'I took multiple academic writing classes in both MA and Ph.D. I also attended online courses and MOOCs, which highly contributed to my writing skills.' Moreover, supervision and supervisor feedback were highly valued by the participants. For instance, P31 commented: 'My second article, which I wrote with my advisor through cooperation, was a good writing experience for me. I think that I have improved myself and my academic writing with the valuable opinions and suggestions of my advisor and different lecturers in my jury.' However, P2 interestingly mentioned the damaging effect of a supervisor when writing. She reported: 'I used to make a lot of grammatical mistakes in my master's thesis, even though I had another friend proofread my manuscripts. And I can't say that I was very successful in using academic vocabulary in graduate school, but I'm better now. My supervisor would discourage me a lot as if I was doing it on purpose. That's why I myself will never be such a teacher and a supervisor in the future, no matter how bad the other person is in terms of producing written materials. I think the worst thing is to devastate (original emphasis on the word during the interview) the courage, interest, and self-confidence of the student in the academy'. This comment is highly revealing regarding the possible adverse effect of a supervisor, which must be productive for the graduate students to solve writing-related problems during the writing process.
Furthermore, writing practice was also considered one of the most effective solutions in the academic writing process. For example, P9 pointed out: 'I always keep my writings and essays. When I lose motivation, I just open it and look. When it is necessary to produce, it is best to proceed in small steps.' P13 also reported: 'I can see the difference between the articles I wrote during my undergraduate education and my current thesis. Also, I think that the biggest share of this difference comes from reading a lot first and then writing. I want to go back and completely change the studies that I have read so far. This is an example that shows that development continues all the time.' It is clear from the quotations that the importance of reflection and doing writing practice is emphasized.

Discussion
The purpose of this mixed-methods study was to examine how graduate-level EFL students regard the level of difficulty in terms of the different sections of a scholarly work in their academic writing practices and if their elt.ccsenet.org English Language Teaching Vol. 15, No. 7;2022 perceptions concerning the difficulty of the various sections reveal a significant change depending on their demographics as well as the solutions they employ when they are challenged with difficulties in academic writing. In general, EFL graduate students' views regarding the process of academic writing were also scrutinized.
The present study's findings revealed that EFL graduate students experienced significant difficulty in academic writing with regard to organization and development of their ideas. Writing discussions, summarizing, and paraphrasing particularly require the skill of decent organization and control over a constellation of skills to succeed. This view was also supported by the study of Cargill and O'Connor (2009). They noted that students in their study also had difficulty organizing and developing their ideas systematically. Thus, the responses of the EFL graduate students in this study revealed a similar crucial point: they had significant difficulty in expressing their ideas clearly and logically, and they had trouble summarizing and paraphrasing, which all compose the significant steps of a successful academic writing process.
The results are also similar to the findings of Bian and Wang's (2016) study as they postulate cohesion and coherence are two problematic areas in academic writing as students utilize 'unclearly signposting connections between sentences, paragraphs, and chapters through inappropriately using linking words or introductory and concluding remarks' (p. 27). In the same vein, the responses of EFL graduate students in this study demonstrated that they found planning writing assignments and writing recommendation sections very difficult and difficult. In this sense, the study's findings are also in line with the findings of Azizah and Budiman's (2018) study in that 65% of the respondents in their study reported that they had significant difficulty in organizing and developing their ideas in writing. Jomaa and Bidin (2017) also noted that students might fail to deliver information accurately when paraphrasing as they struggle to keep the original meaning. Thus, EFL graduate students might be supported to practise writing more often to form a pattern when they are doing academic writing finally. Likewise, Karimnia (2013) also maintained that students had difficulty accurately reporting their results and effectively corroborating a claim, indicating that students need to be backed up with the necessary skills to describe results and support claims when writing.
On the other hand, EFL graduate students' responses revealed that linking sentences smoothly is easy. However, the finding of this study diverges from Al Fadda's (2012) study in that learners experienced difficulties, mainly when they tried to combine sentences in their writing. Referring to sources and writing references were also reported to be easy within the body of this research, probably because the participants were familiar with the styles of citation and the requirements of different journals. Supporting this view, Azizah and Budiman (2018) suggested in their study that students found it easy to cite and make references to the published literature. However, the results of this study conflict with the research of Peat (2002) in that reference and citation in writing are the primary areas where learners make mistakes. In line with the same finding, Jomaa and Bidin (2017) also pointed out that although PhD students received formal training on how to cite, they reported that they still had difficulties when making citations since they needed to focus on other aspects such as the organization, essay writing, and the content for academic writing. Thus, citations could be paid more time and attention as they form an essential part of academic writing. It should be emphasized in academic writing since students could make some mistakes when making citations.
On the other hand, the statistical analyses run to test if the perceived difficulty levels show a difference concerning the demographics of the participants revealed no significant differences concerning the gender, age, or level of English. Academic writing levels revealed no significant differences, suggesting that EFL graduate students' difficulty perceptions regarding different sections of scholarly work derive from other reasons.
Regarding the results of the qualitative analysis, the participants reported that their previous writing experiences included doing academic writing courses and attending seminars and online courses. They particularly emphasized that they developed their writing through modelling, reading scholarly papers and trying to write research papers, which revealed that they improved as they were involved in the writing process. Besides, the major points they considered were comprehensibility and use of academic vocabulary and literary phrasal elements. Their responses revealed that they mostly preferred to attend academic writing courses face-to-face or online, summarizing, paraphrasing, and practising writing to solve the writing-related problems. Peer supervision and the support of a supervisor were also considered essential, which is also similar to the findings of the quantitative section of the current study. The participants reported that they benefitted from their supervisors and sought help from their peers. The findings of the quantitative section of the study demonstrated that the participants would not take additional writing courses. However, the qualitative section revealed that they still regarded attending courses and seminars as crucial to solving their writing-related problems and valued these courses and seminars. In line with this information, Hanjani and Li's (2014) study demonstrated that 111 students attending a genre-based academic writing course showed a favourable attitude towards the course, although they had some concerns.
The participants also stated that one of the practical solutions they utilized was paraphrasing their intended messages and persisted in expressing themselves in various ways. Like summarizing, paraphrasing is one of the vital means by which students grasp the original words and ideas of other scholars in a field; it is a commonly taught strategy in academic writing courses for both native and non-native speakers of English. However, it attracts little attention in both research and pedagogical literature. However, students are required to express the significant points of a text they have read 'succinctly' in their own words to be able to negotiate their meanings (Hedgcock & Ferris, 2009, p. 185). Similarly, the participants in the present study regarded summarizing and paraphrasing as essential components to consider when writing.

Conclusion
It is important to note that the data presented in the current study concern the views of EFL graduate students. However, many graduate students have similar tendencies and problems even in their L1. Holmes, Waterbury, Baltrinic and Davis (2018) and Buckingham (2008) have also noted these commonalities. Thus, perhaps especially noteworthy in this study is that although EFL graduate students may seem familiar with the mechanics of L2, they still have some problems since they fail to have an adequate grasp of academic language and conventions. It has been conclusively shown that writing the methodology section or using appropriate words and phrases to express their ideas succinctly during the writing process still poses problems for the participants.
The evidence reviewed here also suggests a pertinent role in the writing strategies that students must explicitly teach. It has been shown that students also have difficulties planning their writing assignments, summarizing, synthesizing, and summarizing information, which all pertain to the importance of making provisions during writing. Besides, writing an efficient abstract, proofreading and writing recommendations all pertain to the necessity of guidance for students. Since writing is a complex skill, students feel an urgent need for (peer or supervisor) supervision even though they are influential users of L2. Thus, rather than merely offering academic writing courses, students should be provided with practical guidance and ample opportunities to practice their research and academic writing skills, as these are congruent components of academic literacy.

Pedagogical Implications
There seems to be evidence indicating that students need to be equipped with the necessary writing knowledge and practice to gain an optimum academic writing experience. Besides, academic writing courses or curricula should be organized around distinctive characteristics of text genres as required by the different departments to help students write research papers. Rather than providing students with generalizations about the writing texts and how to write, students need to be reminded that they need to be aware of the importance of the writing process as a learning tool to improve themselves in the long term.
Students' previous experiences with writing also revealed that most students try to develop their academic writing skills through trial and error, examining articles, and reading scholarly work, which all point to the lack of adequate guidance. Thus, faculty members could clearly articulate the requirements for students rather than take their readiness for granted. Evaluation should be constant, constructive, and collaborative for the written assignments. Most importantly, faculty members should share the responsibility of tutoring with the staff who offer academic writing courses to have students more involved in the process and help them as exemplary models. Finally, academic courses offered at EFL teacher education programs could be specifically designed to help students become aware of academic literacies and encourage them to have a critical stance toward the topics they write about.

Limitations
The present study was limited to only 34 EFL graduate students, and the data were collected through snowball sampling.