Evaluating Afghanistan University students’ writing anxiety in English class: An empirical research

Abstract Writing anxiety leads to poor writing performance among learners as it hinders their writing. This study investigated the level, types, and causes of writing anxiety among Afghan EFL students. A total of 133 undergraduates was selected as the respondents. The study used a quantitative research method and the data was collected using a questionnaire derived from Second Language Writing Anxiety Inventory and Cause of Writing Anxiety Inventory. Both descriptive and inferential statistics of SPSS were used to analyze the data. The findings of the study indicated a moderate level of writing anxiety, with cognitive anxiety as the dominant type of writing anxiety. In addition, the inferential analysis showed no significant differences in the level of writing anxiety across gender and their years of study. However, a statistically significant difference was found among students from different backgrounds who possessed different English proficiency levels (beginner, intermediate, high-intermediate, and advanced). Moreover, linguistic problems, time pressure, the pressure to be perfect, and the fear of teachers’ negative evaluation were discovered to be the leading causes of writing anxiety. Finally, it is believed that the findings of this study may provide several implications for practitioners in this field to be more aware of students’ writing anxiety in English class. It may alert them of the negative effects of writing anxiety and they may try to make the class as stress-free as possible to improve students’ performance in writing.

Ziauddin Quvanch is a senior lecturer in the English language and literature department of education faculty at Jawzjan University, Afghanistan. He Completed his Master's degree in TESL (Teaching English as a Second Language) at Universiti Teknologi Malaysia (UTM), Malaysia. His research interests are online teaching and learning, Technology Enhanced Language Learning (TELL), Computer Assisted Language Learning (CALL), Mobile Assisted Language Learning (MALL), e-learning, etc. He is also interested in investigating the impacts of anxiety on students' language skills. Kew Si Na is a senior lecturer at Language Academy, Faculty of Social Sciences and Humanities, Universiti Teknologi Malaysia, Malaysia. She received a bachelor in TESL at Universiti Teknologi Malaysia and completed her PhD studies in educational technology at Universiti Teknologi Malaysia.

PUBLIC INTEREST STATEMENT
Her research interests are educational technology, online teaching and learning, learning analytics (LA), teaching English as a second language (TESL), technology enhanced language learning (TELL), computer assisted language learning (CALL), etc. Her research group is Technology and Language Education (TLE).

Introduction
Writing is a significant skill and an important productive activity (Hussin et al., 2015). It has been an essential part of everybody's life in business, creative, and scholarly activities (Keshta & Harb, 2013). However, it is considered a more complex skill than other language skills (Akhtar et al., 2019;So & Lee, 2013). To illustrate, it "stimulates thinking, compels students to concentrate and organize their ideas, and cultivates their ability to summarize, analyze, and criticize" (Maghsoudi & Haririan, 2013, p. 60). It requires one to make compositions containing suitable structures and mechanics. Moreover, EFL learners are required to master academic writing (Giridharan & Robson, 2011), especially in Afghanistan. It means that producing academic compositions is essential for EFL learners. However, it is not an easy process, because it requires specific principles and persistent practices (Pritchard, 2008). For example, a composition has to be structured, logical, cohesive, and well-organized. Therefore, it has been the most challenging skill to learn in ESL/EFL contexts (Jusun & Yunus, 2017) including Afghanistan university students.
In fact, students with insufficient awareness of the process of writing and its components face difficulties in writing construction (Chan, 2010;Zhou, 2009), and it may cause them to become more apprehensive due to the complexity and requirements of writing (Ekmekçi, 2018). Consequently, these negative emotions and perceptions towards the process of writing have been identified as writing anxiety (Challob et al., 2016). It is crucial to come up with more alternatives to reduce the anxiety in writing among EFL students. For instance, using technology (Ebenezer et al., 2021;Kew et al., 2020) to assist teaching and learning practices. Feeling anxious when using the target language negatively affects feelings and impacts the performance and learning (Sadiq, 2017). Researches have long recognized anxiety as the primary source of problem and barrier affecting the learning of a foreign language (Kara, 2013). Besides that, anxiety happens in all primary skills (speaking, listening, writing, and listening) of a foreign language (Wahyuni & Umam, 2017). In particular, anxiety associated with writing outweighs that of the other language skills. As a consequence, it often leads to poor writing performance among learners (Challob et al., 2016). The same happens to Afghanistan university students. Therefore, it is significant to study more about the level of their writing anxiety and the factors contributing to it. Nonetheless, very limited studies have looked into this research area.
Furthermore, it has been identified that language learners feel anxious about writing at various levels of educational settings (Chiang, 2012;Huwari & Abd Aziz, 2011). In agreement with this, Baez (2005) and Horwitz (2001) asserted that anxiety is a widespread phenomenon among university students with different English proficiency levels and even one-third of those who are majoring in English as a foreign language experienced a moderate level of apprehension. Similarly, Miri and Joia (2018), reported low writing proficiency and writing anxiety among EFL students at schools and universities in Afghanistan. However, Afghan EFL students' weakness in writing have always been attributed to reasons other than anxiety, for examples, an absence of good writing skills and a lack of motivation. Hence, anxiety and its effects in writing should be paid more attention in order to help Afghan EFL learners. In conjunction with this, this study seeks to determine the level, types and factors of writing anxiety for the purpose of providing useful insights and implications to researchers and practitioners in this field. The following research questions were addressed to fulfill the purposes of the study (1) What is the level of writing anxiety of Afghan EFL university students?
(2) What are the relationships between writing anxiety and the backgrounds (gender, year of study and English proficiency level) of Afghan EFL university students?
(3) What are the types of writing anxiety faced by Afghan EFL university students?
(4) What are the factors that contribute to students' writing anxiety?

Writing skill and its challenges
Writing is "an action, a process of discovering and organizing ideas, putting them on the paper and reshaping and revising them" (Moore & Murray, 2006, p. 36). It is a sequence of sentences connected in a specific way to make a whole unity (Sulisworo et al., 2016). It has been significant in many areas, especially in the modern workplace (Selvaraj & Aziz, 2019), and in the academic field (Kew & Tasir, 2021;Rudiyanto, 2017). Therefore, it is an important skill and essential productive activity for EFL students (Hussin et al., 2015), including Afghan EFL university students who are expected to master writing skills well. This is because mastering writing skills enables learners to express their feelings in a meaningful way (Jani & Mellinger, 2015), and improves independence, creativity, and fluency (Sulisworo et al., 2016). With good writing skills, they will possess the ability to accomplish writing tasks effectively, which is highly demanded in today's workforce and academic field. As a result, whether or not one can write well plays a fundamental role in Afghanistan. Nevertheless, according to Rudiyanto (2017) and Challob et al. (2016), it has been a difficult skill for EFL learners, contrary to the expectations that they are supposed to master effectual writing skills and become proficient EFL writers Cocuk, etal., (2016). It requires a number of steps which demonstrates the complexity of writing nature. Another challenge is that it takes appropriate structures, content, mechanics, and organizations to accomplish good writing and not many EFL writers understand and master them well. Therefore, learners face barriers when trying to grasp academic writing skills.
Moreover, writing is not just a sequence of words to construct a sentence (Keshta & Harb, 2013), but it requires generating and organizing thoughts into a coherent, informative, and precise way to form a well-organized composition. Without this essential knowledge, students will feel anxious when they are assigned writing tasks. In addition, Ramasamy and Aziz (2018) stated that the complexity of writing lies in its various complicated roles such as the arrangement and expansion of ideas, proper use of grammar, conventions, punctuations, and spellings. Correspondingly, Rudiyanto (2017) also highlighted that writing requires one to organize ideas and apply suitable vocabularies, grammar, and sentence structures. In this regard, due to the complicated nature of writing, it is regarded as a challenging skill even for students who are deemed good writers. Consequently, the multiple requirements in writing make students feel lost, demotivate them, and cause them to lose their interest. In general, this phenomenon is identified as writing anxiety.

Foreign language anxiety and writing anxiety
Anxiety is a psychological construct commonly described as the subjective feeling of tension, a state of apprehension, a vague fear, nervousness, and worry associated with an arousal of the autonomic nervous system (Spielberger, 1983). It is characterized by a diffuse, unpleasant, vague sense of apprehension, often accompanied by autonomic symptoms, such as headache, perspiration, palpitations, tightness in the chest, and mild stomach discomfort (Kaplan & Sadock, 1996). The symptoms can negatively affect students' learning process and performance. In fact, language anxiety is a key influencing element in foreign language learning (Riasati, 2011), and it significantly affects learners' performance in learning and productive skills (Marwan, 2007). In short, it impacts and interferes with the flow of EFL learning (Kara, 2013;Wahyuni & Umam, 2017). Afghan EFL university students are no exception. On the flip side, it is reported that anxiety can have a positive effect. For instance, according to Wehner (2014), anxiety can be facilitating or debilitating. Facilitating anxiety refers to reactions resulting from uneasiness that raise a learner's motivation to improve his or her learning and achievement. It is something helpful and works as a regulation to avoid mistakes and errors (Ningsih & Kusumaningputri, 2015). In contrast, apprehensive reactions that cause low motivation is debilitating anxiety. It causes stress and worry in many people and will likely prevent them from reaching their goals, and affect their performance and achievement in writing. All in all, be it facilitating or debilitating, it is inevitable to involve stress and apprehension in the discussion of anxiety.
Besides that, the term foreign language anxiety (FLA) refers to "the feeling of tension and apprehension associated with second language contexts, including speaking, listening, and learning" (MacIntyre & Gardner, 1994). It was introduced by Chastain 1975 andScovel 1978 as an impactful factor that can hinder the process of learning a foreign language. Recently, anxiety in language learning has become a topic of growing interest among educators and practitioners as it has been a considerable barrier in foreign language learning. It usually happens in the learning environment, which occurs as a consequence of the complicated process of learning a language, all the more so when they are learning writing skills. Moreover, an association has been identified between anxiety and foreign language learning indicating that learners do suffer from negative emotions and reactions when learning a foreign language (Horwitz, 2001). The uneasiness they feel when they are exposed to a language learning setting may lead to poor performance. FLA has also been recognized as a barrier that negatively infuences students' level of proficiency and achievement (Dordinejad & Ahmadabad, 2014;Rudiyanto, 2017). It prevents the students from achieving their targets and keep them from performing in language skills, (Hashemi & Abbasi, 2013) which include writing. As it takes a lot of efforts and constant practices to master writing skills, learners ought to face challenges along the way and this conributes to writing anxiety.
Writing anxiety is a "psychological predisposition faced by an individual in writing task due to tendencies of overwhelming fear arising from a combination of feelings, beliefs, and behaviours affecting the individual's ability to write" (Al-Sawalha & Chow, 2012, p. 6). It has been introduced by Daly and Miller's (1975) influential work (Wahyuni et al., 2019). Since then, various studies on writing anxiety have sailed into the uncharted waters to explore its effects on writing performance in different contexts. Cheng (2004) stated that the studies of writing apprehension in EFL context started since the 1990s. To date, it is still an important research topic to be investigated as it has been identified as a core reason that affects students' performance (Genç & Yaylı, 2019;Miri & Joia, 2018;;Sadiq, 2017), and EFL learners have always been found to have inconveniency in composing (Negari & Rezaabadi, 2012). Anxious learners score lower marks in writing tests. Often, due to their anxiety, they struggle to organize their ideas, make more errors, produce short compositions, and use inappropriate grammar (Kırmızı & Kırmızı, 2015). They could neither concentrate properly nor produce cohesive and coherent compositions, feeling anxious, they are also not able to grasp the content of the input, and they even become forgetful. Writing anxiety also inhibits students' academic work, self-confidence and motivation (Martinez et al., 2011;Negari & Rezaabadi, 2012).
Nonetheless, it is argued that learners having a lower level of anxiety are able to produce wellorganized writing (Hassan, 2001). According to Liu and Ni (2015), moderate anxiety may enhance learners' writing skills. Regardless of whether they have a high or low level of anxiety, anxiety may affect students' performance either positively or negatively. Hence, it is important to understand learners' level of writing anxiety. Considering this, a significant number of studies have explored the perceptions, achievements, and performances of language learners in various ESL/EFL settings (Alfiansyah et al., 2017;Al-Sawalha & Chow, 2012;Dordinejad & Ahmadabad, 2014;Ekmekçi, 2018;Jebreil et al., 2015;Rezaei & Jafari, 2014). Most other studies have revealed that anxiety could have a negative influence on students' overall performance in language and writing skills. However, similar studies on writing anxiety in Afghanistan context are still vague.

Types of writing anxiety in second language learning
Writing anxiety in second language learning is classified into three types, namely, somatic anxiety, cognitive anxiety, and avoidance behaviour (Cheng, 2004). The term somatic anxiety refers to one's perception of the psychological effects brought about by the anxiety they experience. It is generally related to an increased level of negative emotions, like nervousness and tension, as if one is having butterflies in the stomach (Cheng, 2004;Rudiyanto, 2017). Based on Marzec-Stawiarska (2012), it affects language learners by causing them to suffer from upset stomachs, extreme sweating, headaches, heart palpitations and rapid breathing. Also, Alfiansyah et al. (2017) believe that it makes students panic and their mind will then go blank when they start writing. This type of anxiety makes learners feel nervous with high apprehension when they are unable to figure out any ideas under time constraints. It often occurs whenever students are required to finish a writing task within a limited time. On the other hand Morris et al. (1981, p. 541) define coginitive anxiety as "negative expectations and cognitive concerns about oneself, the situation at hand, and potential consequences". It refers to psychological states such as having negative expectations, feeling concerned about weak performances, and worrying about others' negative evaluations (Rezaei & Jafari, 2014;Rudiyanto, 2017;Wahyuni & Umam, 2017). Teachers' high expectations of students' performance may lead to apprehension among students. Students may focus more on fulfiling the teacher's expectations rather than the process of writing itself. To put it succinctly, anxiety-related concerns and tasks that are not related to cognitive activities might decrease students' performance. Moreover, avoidance behaviour associated with writing anxiety indicates the behavioral characteristics of avoiding writing (Cheng, 2004). It is commonly experienced by EFL learners (Rudiyanto, 2017),which causes them to avoid writing compositions or to avoid any situations that would engage them in writing. This is identified as a serious type of writing anxiety. and it affects students' performances in writing class. It is so because apprehensive learners are often distracted and restless, and they find it difficult to pay attention and comprehend their learning tasks. Rezaei and Jafari (2014) classified the possible contributing factors of writing anxiety into ten categories. The first is negative evaluation. It is the basis of writing apprehension among EFL learners (Alfiansyah et al., 2017).The learners worry about teachers' and peers' evaluations. Besides, most EFL learners feel apprehensive due to the evaluations given on their writing. It occurs when a student's composition does not suit the criteria given by the teacher. Furthermore, it also refers to worrying about being laughed at by others because of poor performance (Brown, 2004). In these conditions, learners invariably experience looking "foolish" and "dumb" when they are evaluated by peers. They also feel uneasy when their writing is compared to that of their peers (Zhang & Zhong, 2012). Next, the second category is fear of tests. It stems from the fear of failing in examinations. It arises when the learners' writing is evaluated based on the elaboration of ideas, coherence, cohesion, vocabulary and quality of idea arrangement in examinations (Ozturk & Cecen, 2007). Moreover, writing is a productive skill that is strongly influenced by time pressure; even the most competent learners make mistakes in test situations (Zhang, 2011).

Causes contributing to second language writing anxiety
Thirdly, insufficient writing practice can also lead to writing anxiety. It refers to circumstances in which the students are required to concentrate on the structure more than the content of the composition. However, they are unable to respond appropriately because of insufficient practice (Rezaei & Jafari, 2014). This situation makes learners anxious, especially for those who have not practised sufficiently and are unfamiliar with academic writing structures (Abu Shawish & Abdelraheem, 2010). It can also be seen when a composition requires learners to explore personal feelings, experiences, and attitudes (Kara, 2013). The fourth category is insufficient writing technique. To put it simply, it means students have poor skills in writing. Learners with inadequate knowledge and techniques in terms of writing experience apprehension (Wahyuni & Umam, 2017). In other words, learners who do not possess good writing techniques and proper composition skills suffer from poor writing performance. Fifthly, the problem could lie in the choice of topic. Inadequate topical knowledge leads to writing anxiety among learnes. In a manner of speaking, learners feel anxious when they know little about the writing topic given. In line with this, Zhang (2011) emphasized that language learners' prerequisite knowledge on a topic plays a major role in the accomplishment of its relevant writing task. Learners who are assigned to respond to writing topics which they are unfamiliar with face more anxiety than those who write on familiar topics.
Sixthly, linguistic difficulty can also contribute to writing anxiety. It refers to learners' incompetency in linguistic. It is recognized as a common challenge faced by learners when it comes to writing (Wahyuni & Umam, 2017). Having insufficient knowledge on linguistic components demotivates EFL learners as it takes an abundance of grammar rules to compose an essay and express their ideas accurately. The lack of vocabulary, grammar rules, and coherence of ideas makes learners anxious. Besides, not being able to master morphology, syntax and writing techniques also contributes to their anxiety (Ozturk & Cecen, 2007). Moving on, the seventh factor contributing to writing anxiety is the pressure to be perfect. Perfect writing requires tonnes of efforts as one needs to go through a complicated process and live up to the demanding standards. Also, it requires learners to compose and synthesize ideas whilst meeting the academic criteria. The high expectations cause them to be exceedingly anxious when they are asked to complete a writing task Conversely, Alfiansyah et al. (2017) opined that the pressure to seek perfect performance in writing could be advantageous. They believed that learners will benefit from the constant practices and exercises that are done to perfect their work. Eighthly time pressure can cause writing anxiety, too. It refers to the time limitation of a writing task and is often recognized as a source of language learners' anxiety (Rezaei & Jafari, 2014). A well-organized composition requires essentials steps such as planning, writing, and revising, which are usually time-consuming. Learners may lose confidence when they are unable to take their time concentrating on the process of writing.
Ninthly, frequent writing assignments may make learners anxious. EFL learners experience apprehension when the assignments in writing courses constitute a major percentage in their final grade (Shang, 2013)Since their performace will be measured, they feel anxious when writing. Finally, it affects their writing performance. Lastly, self-confidence plays a part in writing anxiety. Low self-confidence has always been noticed as a significant factor of anxiety in writing (Olanezhad, 2015). It impedes students' performance in writing. When they are not confident in themselves, they will not be able to share their feelings and ideas properly. On the contrary, a positive attitude helps learners engage in writing tasks and allows instructors to foster writing skills easily.

Relationship between writing anxiety and the background of second language learners
Numerous studies have been conducted to examine the relationship between writing anxiety and the background of second/foreign language learners as it is an important research area. It has been reported that writing apprehension affects students regardless of their gender (Karakaya & Ulper, 2011). Spielberger (1983) argued that "females are more emotionally stable than males in their reactions to highly stressful or relaxing circumstances" (p .19). In contrast, Machida (2001) stated that female students suffer from a higher level of anxiety as compared to male students. On the other hand, Jebreil, Azizifar and Gowhary (2015) found that male students experience a higher level of anxiety than female students do. As far as these previous researches are concerned, it is safe to conclude that,somehow rather, both male and female students experience a certain level of anxiety in writing (Salem & Al Dyiar, 2014). Besides, studies related to English proficiency level illustrate that ESL/EFL learners with low competency in English are more stressful than competent students (Daud et al., 2016). Similarly, Jebreil et al. (2015) reported that beginners in Iranian EFL context experience a higher level of anxiety. In contrast, Aljafen (2013) reported no statistically significant difference in the level of anxiety between beginner, intermediate and advanced students. He also stated that regardless of students' competency in English, all of them suffer from the same level of anxiety. Furthermore, a study on writing anxiety associated with learners' year of study shows that the level of anxiety decreases with the increment of year of study (Cheng, 2002). In line with this, Wahyuni et al. (2019)revealed that sophomores generally face a higher level of anxiety in writing as compared to juniors and seniors. As a result, it means that students who are in lower academic years are more susceptible to writing anxiety. In conclusion, since the findings from previous researches are diverse, it is of utmost significance to examine the relationship between writing anxiety and the background of second/foreign language learners in different contexts, including the Afghan university students.

Previous related research
As writing anxiety has drawn much attention lately, several surveys have been carried out to investigate its levels, types, and contributing factors in ESL/EFL contexts (Wahyuni et al., 2019;Kusumaningputri et al., 2018;Ekmekçi, 2018;Genç & Yaylı, 2019;Rudiyanto, 2017;Jebreil et al., 2015). For instance, Rezaei and Jafari (2014) found out that EFL learners in Iran suffered from a high level of writing apprehension, with cognitive anxiety being the most dominant type. In a similar research, the EFL students in Iran were found to have high anxiety level in terms of writing skills with elementary learners were identified to have a higher level of anxiety (Jebreil et al., 2015). On the other hand, Min and Rahmat (2014) reported a high level of anxiety and somatic anxiety among their Malaysian respondents, particularly the males who were found to be more anxious than their female peers. Correspondingly, Wahyuni and Umam (2017) conducted a study in Indonesia and revealed that 54% of the students experienced a high level of writing anxiety, with cognitive writing anxiety being the dominant type and linguistic difficulties being the primary factor. Apart from that, a study conducted by Rudiyanto (2017) among Indonesian EFL learners revealed that students experienced a medium level of somatic anxiety and the primary factor was insufficient writing practice. In the same vein, Wahyuni et al. (2019) who investigated writing anxiety among Indonesian EFL learners discovered that that they had a moderate level of writing anxiety, with the choice of topic being the most challenging cause. Furthermore, Miri and Joia (2018) studied anxiety among EFL learners in Afghanistan and they managed to identify teachers' negative evaluation, inadequate grammatical knowledge and errors in composing essays as the causes of learners' writing anxiety. In summary, these studies have shown different research findings on the level of writing anxiety, its types, and the contributing elements of anxiety in ESL/EFL contexts. In fact, writing anxiety has been one of the essential issues in the academic world for a long time; however, not much attention pertinent to it has been given in Afghanistan. In particular, Afghan English learners have not been adequately studied. Literature shows that there are very limited studies on writing anxiety. There is only one qualitative study done by Miri and Joia (2018) in the Afghan EFL setting. It was conducted qualitatively from the perspective of 5 university students, and thus, the result cannot be generalized to the general population of Afghan EFL. Consequently, in line with the literature review and the lack of related studies on writing anxiety within the Afghan education setting, there is a need to conduct a study in a larger population.

Research design and sample
A quantitative approach with survey design was implemented to fulfill the aims of the study. An empirical research is based on observed and measured phenomena and it derives knowledge from actual experience. Flynn et al. (1990) defined the term "empirical" as "knowledge based on real world observation or experiment." In other words, an empirical research is used to describe field-based research, which uses data gathered from naturally occurring situations or experiments, rather than laboratory or simulation studies (Jasti & Kodali, 2014). Empirical studies are observational or experimental rather than theoretical. According to Flynn et al. (1990), survey design with questionnaires is the most commonly used approach in empirical researches. Similarly, empirical methods have been equated with the use of quantitative measures (e.g., content analyses, surveys) and primary collection and analysis of data (e.g., Bavelas, 1995). Odoh and Chinedum (2014) pointed out that a survey is the combination of distinct characteristics that are related to the way in which the information about the object of study is gathered. It is often used to collect information on a certain subject. In this respect, this study utilized a set of questionnaires to study the writing anxiety among English learners in Afghanistan. A purposive sampling method was used. It is a sampling technique whereby the researcher chooses a research population that conforms to specific criteria (Adams et al., 2007). A total of 133 students participated in this study. These participants were undergraduate EFL learners from English Language Department of the Faculy of Education in a famous Afghanistan university.

Research instruments
The current study utilized two survey questionnaires to collect data. The first instrument was the Second Language Writing Anxiety Inventory (SLWAI) developed by (Cheng, 2004). It was adopted to find out the level and the types of writing anxiety. It is a valuable inventory in terms of reliability and validity (Cheng, 2004;Ozturk & Cecen, 2007). Furthermore, the inventory involved 22 items and it was of a multidimensional scale that included subscales consisting of somatic, avoidance behaviour, and cognitive anxiety. Additionally, this inventory adopted a Likert scale with five options, namely, strongly disagree (1), disagree (2), uncertain (3), agree (4), and strongly agree (5). On the other hand, the second instrument was the Causes of Writing Anxiety Inventory (CWAI) involving 10 items developed based on the contributing causes of writing anxiety (Rezaei & Jafari, 2014). The questionnaire also employed a Likert scale with five options: strongly disagree (1), disagree (2), uncertain (3), agree (4), and strongly agree (5).

Data collection and data analysis
Before collecting the actual data, a pilot study was carried out and evaluated using a reliability test in Statistical Package for Social Science (SPSS) software. The results proved that the instruments were of a reasonable range of reliability. Next, after obtaining the permission from the university, the questionnaires were administered to the respondents. The researcher described the guidelines to respondents. They were then given 20 minutes to answer the questionnaires. They were asked to read the items carefully and be sincere in their responses in order to obtain more accurate results. Finally, all responses were arranged and analyzed using SPSS Version 25. The data was analyzed using quantitative data analysis procedures. For RQ1, descriptive analysis was used to determine the level of writing anxiety. As for RQ2, inferential statistics were used to answer the formulated hypothesis in terms of respondents' gender, English proficiency level, and year of study. Also, for RQ2 and RQ3, descriptive analysis was applied to analyze and find out the types and factors of writing apprehension.
Other than that, the researcher employed a number of steps to analyze the data obtained from SLWAI, which will be explained as follows. Seven items (1, 4, 7, 17, 18, 21, and 22) which were negatively worded were scored reversely before calculating the total scores (Ekmekçi, 2018;El Rusyda, Raja & Yufrizal, 2017;Jagabalan & Nimehchisalem, 2016 and). After summing up the total score of each item, the mean values were classified into three categories (high, moderate, and low level) to determine the level of anxiety. The scores above 65 were placed in the category of highlevel anxiety whereas the scores below 50 were categorized as low-level anxiety. Meanwhile, the scores between 50-65 were classified as moderate-level anxiety. Table 1 highlights the mean classification.
Furthermore, systematic steps were implemented to analyze and determine the types of writing anxiety. Firstly, the items were grouped according to the types of anxiety, namely, cognitive anxiety (1, 3, 7, 9, 14, 20, and 21), somatic anxiety (2,6,8,11,13,15, and 19), and avoidancebehavior anxiety (4,5,10,12,16,18 and 22). The items (1, 4, 7, 17, 18, 21, and 22) were reversely scored before summing up the total score for each category. Finally, the formula below was applied to find the mean of each type of anxiety (Wahyuni & Umam, 2017). The one with the highest mean was recognized as the most commonly experienced type of writing anxiety. The figure 1 below shows the formula. On the other hand, to analyze CWAI, another calculation technique was applied (Wahyuni & Umam, 2017). Firstly, the scores of each item were summed up. Next, the total score of each item was divided by the total score of all items. After that, the obtained result was multiplied by 100% to get the percentage of each item. The ones constituting the highest percentage could then be identified as the primary contributing factor of writing anxiety. The figure 2 below shows the formula.

Analysis of respondents' background
The analysis of respondents' gender illustrated that 77 (57.9%) were males, and 54 (42.1%) were females ( Table 2). The majority of respondents 78 (58.6%) were 21-26 years old. 54 (40.6%) of them were between 15-20, and there was only one (0.8%) participant who belonged in the elder age group of 27-32. As for year of study, 31 (23.3%) of the respondents were freshmen and, 23 (17.3%) were sophomores. Juniors and seniors constituted 46 (34.6%), and 33 (24.8%) of the respondents respectively. In terms of English proficiency level,19 (14.3%) respondents possessed beginners' level, 47 (35.3%) of them were at the intermediate level, followed by 43 (32.3%) who were at high intermediate level, and 24 (18.0%) were advanced learners.    Table 3 indicates the descriptive analysis of the level of anxiety. There were three levels of anxiety; high, moderate, and low. The categorization of the level was based on the data (scores) obtained from SLWAI. The scores range from 22-110, but the scores in this study ranged from 34-90. In parallel with the calculation method, the result shows that 47.37% (N = 63) of the respondents experienced a high level of anxiety, 41.35% (N = 55) had moderate-level anxiety. In comparison, only 11.28% (N = 15) of the respondents experienced a low-level anxiety in writing.

Findings on writing anxiety level
Based on the outcome, the overall mean score was (Mean = 64.30 < 65). It specified that the learners had a moderate level of writing anxiety.

Findings on the relationship between writing anxiety level and gender
An independent t-test was directed to analyze the level of writing anxiety among different gender. The result of Leven's test indicated that the significant value between the two groups was 0.694, which was higher than the alpha value (0.05). Hence, there was no significant difference in the level of writing anxiety between male and female students. The t-test analysis for the second language writing anxiety level is shown in Table 4.
As shown in the table, the result indicated that the p-value/significance value (p = 0.542) was higher than alpha value, 0.05. Therefore, the researcher failed to reject the formulated null hypothesis (H0). It indicated that there was no significant difference in the level of writing anxiety among different gender (T (131) = 0.611, p > 0.05).

Findings on the relationship between writing anxiety level and year of study
The one-way ANOVA test showed that, the significant value of 0.062 was higher than the alpha value (0.05) (Table 5). Therefore, the researcher failed to reject the null hypothesis and concluded that there was no statistically significant difference in the level of anxiety among the respondents from different years of study (F (3,129) = 2.500, P > 0.05).

Findings on the relationship between writing anxiety level and English proficiency level
The analysis showed significant differences in the level of writing anxiety among students with different English proficiency levels (Table 6).  According to the obtained result, the significant value of (0.003) was smaller than the alpha value of 0.05. So, the null hypothesis was rejected. It can be concluded that there was a statistically significant difference in the level of writing anxiety among students with different English proficiency levels (F (3,129) = 5.019, P < 0.05). Moreover, to determine the exact difference, a post hoc (HSD) test was applied. Table 7 presents the result Tukey (HSD).
The results indicated there was significant differences between respondents of beginner and intermediate levels (p = 0.033), respondents of beginner and high intermediate levels (p = 0.024) and respondents of beginner and advanced levels (p = 0.001). The null hypothesis was rejected. Overall, beginners suffered from highest level of anxiety whereas advanced students had the lowest level of anxiety.

Findings on the types of writing anxiety
To find out the dominant type of second language writing anxiety, the mean of each type of writing anxiety was calculated. The type that obtained the highest score was defined as the dominant type of writing anxiety. In other words, by calculating the gained score of each item related to each type of SLWAI, the dominant type of writing anxiety was identified. The mean and percentage of each type of writing anxiety are shown in Table 8.
Based on the results, the type of anxiety that obtained the highest mean value was cognitive anxiety (M = 23.42), somatic anxiety obtained a lower mean value of (M = 22.43) and avoidance-behavior anxiety scored the least mean value (M = 18.45). Therefore, cognitive anxiety was the dominant type of anxiety.

Findings on the causes of writing anxiety
To find each item's percentage, the scores of each item were summed up and then divided by the total score of all items. The one with the highest percentage was identified as the dominant cause of writing anxiety. Table 9 presents the calculation and percentages of each item of CWAI.
As shown in the table, item 6 marked the highest percentage (11.60%). Therefore, it can be claimed that the primary source of writing anxiety was linguistic difficulties such as insufficient mastery of vocabulary, substandard sentence construction, grammatical mistakes and so on.

The level of writing anxiety
The outcomes regarding the level of writing anxiety revealed that the average mean was 64.30, indicating a moderate level of writing anxiety. Pertaining to this issue, Aljafen (2013) considered the average level of writing anxiety normal. It is a naturally occurring feeling that will motivate them to work harder, improve their performances and become better writers. In addition, according to Liu and Ni (2015), moderate-level anxiety may boost learners' writing skills and help them produce wellorganized writing. The finding is similar to that of Ekmekçi (2018) who investigated writing anxiety in Turkey EFL context and found out that more than half of the participants had a moderate level of anxiety. Besides, Wahyuni et al. (2019) also discovered that the majority of Idonesian EFL learners in their study experienced a moderate level of anxiety. However, the finding of this study does not correspond with the researches done by Rezaei and Jafari (2014) and Jebreil et al. (2015) who reported a high-level writing anxiety among Iranian EFL university students. Moreover, Min and Rahmat (2014) similarly reported a high level of writing anxiety among the students studying in Universiti Putra Malaysia (UPM) affecting their writing performance.

Writing anxiety and the background of the students
The results revealed that there was no significant difference in terms of anxiety levels between males and females. It is supported by Salem and Al Dyiar (2014), who reported no statistically significant difference between respondents of different gender in terms of anxiety level. However, the findings of this study conflicts with that of Jebreil, Azizifar, and Gowhary (2015), who found statistically significant results which indicated that males experienced a higher level of anxiety as compared to females. Thus, it can be an ordinarily occuring situation for males and females to suffer from the same level of anxiety. In harmony with Salem and Al Dyiar (2014), it is normal for male and female learners to experience the same learning difficulties since they are exposed to the same learning environment, learning context, course and content at the university. In other words, being familiar with the learning environment may lead them to experience the same input.
Moving on, there was no significant difference in the level of anxiety among freshman, sophomore, junior and senior students. Cheng (2002) used to report the same outcome. However, they conflict with Wahyuni et al. (2019) study as they found statistically significant differences in a similar setting. In particular, they concluded that sophomores encountered a higher level of anxiety than juniors and seniors did. The findings of the present study could be attributed to similar background and English proficiency level among the respondents. Also, the same learning environment, curriculum and writing courses offered by practitioners might be another possible reason for students from all academic years to experience the same level of anxiety.
On the other hand, the results revealed statistically significant differences in writing anxiety among respondents with different English proficiency levels. The beginners were found to have a higher level of anxiety. In this regard, Daud et al. (2016) stated that learners with low proficiency level tend to be more anxious because of a lack in language skills. Based on this argument, students with low proficiency level are generally more anxious than their advanced counterparts. It is supported by, Jebreil et al. (2015) who found significant differences which indicated that beginner studenst suffered from a higher level of writing anxiety. Nevertheless, it contrasts with Aljafen (2013) who found no statistically significant differences as the participants suffered approximately the same level of anxiety. To sum up, in comparison with proficient English users, beginners with little language skills were more easily influenced by anxiety.

Types of second language writing anxiety
In this study, cognitive anxiety was found to be the most dominant type of writing anxiety among the Afghan university students. It shows that these EFL students were mostly influenced by external factors such as high pressure brought about by negative evaluation, concerns on weak performance and tests. Also, students experienced cognitive anxiety due to the pressure of writing tests and assessments. As a consequence, they were worried and nervous; thus, they had problems concentrating on the writing task. Besides, they were also affected by the grading system. In a study conducted by Rahim et al. (2016), the respondents reported that their grades were given based on the quality of their writing and their work was assessed based on the efforts that they had put in writing. It is similar to the Afghanistan context, whereby the nature of learning is product-oriented. The only assessment method of students' products is the rigid grading system. It may affect their performance and lead to anxiety.
The outcome of this study is in line with previous research studies (e.g., Jebreil et al., 2015;Masriani et al., 2018;Rezaei & Jafari, 2014;Wahyuni & Umam, 2017) which correspondingly reported cognitive anxiety as the dominant type of writing anxiety. Nevertheless, it differs from a few other studies which identified somatic anxiety as the dominant type of writing anxiety (Alfiansyah et al., 2017;Ekmekçi, 2018;Min & Rahmat, 2014). The present study also does not comply to Jagabalan et al. (2016)who summarized avoidance behavior as the dominant type in the same aspect. To conclude, the findings from different studies show that learners with various social, cultural, geographical, and knowledge background experienced writing anxiety differently. However, this study confirmed that the Afghan university students experienced cognitive anxiety the most.

Causes of writing anxiety
Based on the findings, linguistic difficulties was found as the main cause of writing anxiety. The findings are in agreement with Wahyuni et al. (2019) in the Indonesian EFL context citing linguistic difficulties as the primary factor of anxiety in writing. However, some studies were contradicting, for instance, Rezaei and Jafari (2014) and Alfiansyah et al. (2017), who found teachers' negative comments as the top-ranked source of writing anxiety. On the other hand, Rahim et al. (2016) reported time pressure as the most crucial cause of writing anxiety. It is not surprising that linguistic difficulties is the dominant causes since it requires extra efforts and time. Additionally, it shows that Afghan EFL learners suffered from a lack of experience in writing, poor vocabulary, grammatical errors, and the confusion of sentence structures and spellings. Therefore, they might face anxiety in writing. To support this idea, Wahyuni and Umam (2017), stated that a lack of knowledge in linguistic components is a common challenge faced by EFL learners when they write compositions in English. It even makes them lazy to write because it is difficult for them to express their ideas appropriately and grammatically. On the other hand, linguistic components such as sentence structures, grammar rules, and spellings, have always been the essential instruments in English language teaching and learning. EFL learners feel anxious when writing compositions because they are unable to cope with the complex grammar rules to ease the flow of their ideas, and this impedes their writing performance. In this process lies the most significant reason that contributes to writing anxiety.

Conclusion and implication
The purpose of this study was to find out the level, types, and causes of writing anxiety among 133 EFL students in Afghan context as the literature shows that pertinent researches in Afghan context are limited. Conclusively, the findings revealed that these learners had a moderate level of anxiety, while the analysis of inferential statistics demonstrated that there was no significant difference in the level of writing anxiety across gender and respondents' years of study. However, interestingly, a statistically significant difference was found among students with different English proficiency levels. It was also found that the beginners suffered a higher level of anxiety. It means that more alternatives and assistances should be provided to these students to help them overcome their writing anxiety. Furthermore, cognitive anxiety was recognized as the dominant type of writing anxiety, and the leading cause of writing anxiety was linguistic problems. In this regard, in order to help Afghan students to perform better in English writing tasks, teachers should emphasize more on students' linguistic problems, and lower their expectations regarding student's performance. High expectations lead to cognitive anxiety when learners focus more on expectations rather than the writing process.
The findings of this study managed to provide several implications. Firstly, it helps teachers to be aware of students' writing anxiety. They can concentrate better on addressing students' difficulties in writing. In the meantime, students' awareness of anxiety is also important because it allows them to reflect on their problems and improve their writing deficiencies. Eventually, it is useful for syllabus designers to develop a proper writing syllabus which focuses on students' specific learning needs. Lastly, the implication for higher education institutes is that they can establish writing centers to provide free consultations on academic writing in order to reduce students' writing anxiety and to implement the technological tools and software to ease writing practice (Na & Kwan, 2019;Zheng & Na, 2021), especially in Afghan context.

Limitations and recommendation
Firstly, this study was limited to only one university in Afghanistan. So, it is recommended that future researchers study writing anxiety in more universities to get a more accurate picture of writing anxiety. The second limitation is the small sample size associated with the first limitation. As only one university was chosen the researcher did not manage to get a great deal of respondents. Therefore, it could be useful to carry out a similar study with a larger population majoring in the same field across universities in Afghanistan. Moreover, the researcher conducted a quantitative study using a set of questionnaires. It would be better to also adopt qualitative data collection techniques such as interviews and classroom observations to obtain in-depth data.

Questionnaire II-Causes of Writing Anxiety Inventory (CWAI)
Instructions: Read the following statements and tick the appropriate column that best describes your degree of agreement for each statement.
Please answer the questions using the scale: • SD: Strongly Disagree (1) I worry about negative comments and evaluation given by the teacher.
(1) I am afraid of writing tests. (