The effect of task-based language teaching on analytic writing in EFL classrooms

Abstract As task-based language teaching focuses on real word tasks and the learners need to complete these tasks in the process of learning a foreign or second language, it helps target language fluency and student confidence. That is why second and foreign language teachers and researchers have shown interest in TBLT. This study attempts to investigate the effects of employing task-based writing instruction on Iranian EFL learners’ writing competence. The participants included 69 Iranian EFL learners at the intermediate level and they were placed randomly into a control group and an experimental group. The students in the experimental group performed writing tasks using task-based language teaching techniques, while those in the control group practiced writing skills using traditional writing exercises. To collect the pre-test and post-test data, the researchers administered the writing sections of two paper-based TOEFL tests and analyzed the data through Statistical Package for Social Sciences using descriptive statistics, t-test, and analysis of variance. The results showed significant improvements in the writing ability of the Iranian EFL learners who practiced writing skills using task-based language teaching techniques. Besides, using task-based writing techniques improved the Iranian EFL learners’ ability significantly in terms of different aspects of the writing competence, including sentence mechanics, language use, vocabulary, content, and organization.

Abstract: As task-based language teaching focuses on real word tasks and the learners need to complete these tasks in the process of learning a foreign or second language, it helps target language fluency and student confidence. That is why second and foreign language teachers and researchers have shown interest in TBLT. This study attempts to investigate the effects of employing task-based writing instruction on Iranian EFL learners' writing competence. The participants included 69 Iranian EFL learners at the intermediate level and they were placed randomly into a control group and an experimental group. The students in the experimental group performed writing tasks using task-based language teaching techniques, while those in the control group practiced writing skills using traditional writing exercises. To collect the pre-test and post-test data, the researchers administered the writing sections of two paper-based TOEFL tests and analyzed the data through Statistical Package for Social Sciences using descriptive statistics, t-test, and analysis of variance. The results showed significant improvements in the writing ability of the Iranian EFL learners who practiced writing skills using task-based language teaching techniques. Besides, using task-based writing techniques improved the Iranian EFL learners' ability significantly in terms of different aspects of the writing competence, including sentence mechanics, language use, vocabulary, content, and organization. Reza Kafipour ABOUT THE AUTHORS Reza Kafipour is a PhD holder in TESL (Teaching English as a Second Language). He has more than 8 years teaching experience in state and private universities in Iran. He has published several books and more than 25 articles in reputable journals. His interests are Applied Linguistics, vocabulary acquisition, Vocabulary and language learning strategies, and error analysis.

PUBLIC INTEREST STATEMENT
In this study, we shifted from a traditional approach of teaching writing to a more studentcentred approach. Instead of giving our learners a topic to write about and revising their drafts afterwards, we presented our learners with a topic and asked them various questions to motivate them to discuss it. In this Pre-task stage, we used scripts, charts and pictures to elicit learners' ideas. In this stage, learners had an opportunity to work in pairs or groups to discuss the topic and produce some language that could be useful for their task. Then, during a Task stage, learners had a chance to write their essays while we stepped in to guide them occasionally. During the Post-task stage, one or two students presented their work so that cohesive devices, grammar, and content were discussed. This approach, known as Task Based Language Teaching, helped us to improve our students' writing proficiency significantly.

Introduction
Over the last decades, the emergence and decline of various teaching methods from grammartranslation method to more communication-oriented approaches have characterized languageteaching profession. The success and failure of a method is dependent on how instructors use it, for what purpose they use it, who the trainees are, and what type of the outcomes instructors expect of it, not to mention other factors, such as learners' personality characteristics, their learning styles and motivational level, and the purpose for which they are learning the language. Therefore, no one can claim a language-teaching method as a total success or a real fiasco. Another point is that a particular teaching method or approach no longer dominates the language teaching profession. Furthermore, with a shift from teacher-centered approaches to more learnercentered approaches, what is of high significance in language classrooms is the learners' needs and their ability to use language in natural real-life situations (Richards & Rodgers, 2001).

Task-based language teaching
The first person who brought TBLT into teaching profession was Prabhu. According to Prabhu (1987), leaners might learn easier when they focus their minds on the task and not on the language forms. He defined a task as an activity where learners use the process of thinking to come up with an outcome from the inputs they receive. TBLT also assists teachers to control and regulate the thinking and learning process (Van Den Branden, 2006). According to this definition, reading a bus timetable and deciding which bus must be taken to get to a given destination on a given day and at a given time is an example of an appropriate classroom task that students can perform (Richards & Rodgers, 2001).
The main objective of TBLT is to engage language learners in authentic language use through carrying out a series of tasks while interacting with other learners. It also assists students to learn new linguistic knowledge and organize their existing knowledge (Ellis, 2003). This approach also offers an alternative for language teachers. That is to say, teachers do not predetermine what language forms they practice in the classroom and the lesson revolves around how a central task performs, so what happens when the learners are performing tasks dominates the language forms (Frost, 2004).
Those who advocate TBLT argue that language exercises are unnecessary to learn languages. When language learners use languages to express meaning and transmit messages across situations that resemble real-world conditions, language instruction will be effective and successful (Nunan, 1989). Therefore, we should create and use only those tasks with functions, context, and unified language skills similar to those skills needed for establishing real communications in everyday situations. In TBLT, according to Nunan (1989), there is a focus shift from the outcomes of instruction (learners acquire linguistic knowledge or skills) towards the processes of learning (the activities learners need to do to learn more efficiently). In the same way, Kumaravadivelu (1993;cited in Ellis, 2006) pointed out that methodology is the central element of task-based instruction as the main goal in this approach is to enable students to find their own paths to learning. However, a fact worth mentioning is that a task-based curriculum requires teachers to make decisions about content, include the tasks in the syllabus and methodology, and perform these tasks in the classroom by learners.
Generally, the researchers and instructors agree that the task-based instruction is potentially motivating, stimulating, empowering, and challenging. These characteristics help learners to gain autonomy in language learning. Besides, task-based language teaching can present language learning in classrooms in a more natural way and may trigger a higher rate of language acquisition. Since the TBLT methodology assist learners to have a clear goal for real communication, to facilitate language learning, there is a need for interaction and the comprehensive input (Wang, 2006).
Task-based learning is helpful to language learners since it is more learner-centered than other approaches. While the teacher may provide language forms needed for communication in the pretask phase, the students can use any form they want and this enables them to use all the language forms and structures they know, rather than a single predetermined form set by the teachers or the textbook. Moreover, as students become familiar with the tasks, they are encouraged to be engaged, and this may further encourage them in their language learning.
In TBLT, learners also focus on meaning rather than language forms. This is because students perform a group of communicative tasks instead of doing form-based discrete language drills. In addition, students need to express their own ideas, in either a verbal or a written mode, about the topic under discussion. They can also work on their desired topics. Teachers can assist the students to express their ideas via the analysis of meaning taken from communicative activities and tasks. According to Willis (1996), "within the TBL framework, tasks and texts combine to give students a rich exposure to language and also opportunities to use it themselves" (p. 101). As such, learners practice a certain grammatical feature in an inductive way when performing a set of tasks.
Research on TBLT shows pros and cons regarding the use of TBLT in language classrooms. Furthermore, it shows various aspects associated, such as methodology, classroom management, and practical issues including proficiency in English, the number of students, time needed for preparation and workload in the TBLT approach, knowledge for how to complete tasks, students' unawareness of this learning process, the complexity of tasks which the learners should perform, and the assessment of learning.
Concerning teachers' attitudes towards TBLT and classroom management, it's worth mentioning that EFL teachers are unwilling to use communicative approaches, such as TBLT, but they prefer the traditional Presentation-Practice-Production (PPP). In addition, they are worried about discipline and the role of the teacher while they give priority to textbook coverage (Carless, 2002); moreover, they prefer quiet and controlled classrooms to noisy activities (Carless, 2002). Some teachers also believe that TBLT would delay the students who are more competent (Jeon & Hahn, 2006). Nonetheless, some teachers support the efficiency of TBLT for small-group work, interactive tasks, and learning outcomes, such as motivation, independence, autonomy, and strategies developed in line with students' needs (Jeon & Hahn, 2006).
Practical issues are also among other concerns for teachers. For example, teachers may personally have limited proficiency in English or feel insecure in classes where they have to perform communicative tasks (Watson Todd, 2006). Classes with many of students are another factor of concern (Jeon & Hahn, 2006). Lack of preparation, shortage of time, and heavy workload in the TBLT approach are other issues (Carless, 2002;Jeon & Hahn;. In addition, teachers may have difficulty with the limited class time available for carrying out tasks and all the other activities in the TBLT cycles, such as task repetition or focusing on form (Carless, 2003;Lopes, 2004;McDonough & Chaikitmongkol, 2007).
Other problems may result from teachers' lack of systematic knowledge for implementing tasks or for shifting from a teacher-centered approach to a learner-centered approach, such as TBLT (Jeon & Hahn, 2006;McDonough & Chaikitmongkol, 2007;Xu, Liu, & Jiang, 2008). Other systematic challenges include students' unfamiliarity with this learning process (Jeon & Hahn, 2006), task complexity (Carless, 2009), links between tasks (Watson Todd, 2006), transitions between various materials, and unexpected responses to students through clear instructions and accurate feedback (McDonough & Chaikitmongkol, 2007). Assessment can also cause some practical concerns for teachers. For instance, some teachers do not also know how to measure learners' performance (Jeon & Hahn, 2006), whereas others are concerned with the teacher's tendency to mark students' performance on tasks subjectively (Watson Todd, 2006). Another major problem is the mismatch between tasks and traditional form-oriented exams (Carless, 2007).
Before implementing TBLT, instructors need to take into account the notion of communicative tasks. Willis (1996, p. 28) defines the term task as those "activities where the target language is used by the learner for a communicative purpose (goal) in order to achieve an outcome." Furthermore, tasks must meet certain requirements. First, tasks should have an appropriate complexity and difficulty level that matches students' level of proficiency. Second, the main goal of a task in an ESP context should target learners' needs for communication in real-life situations or at the workplace. In addition, tasks must incorporate authentic materials from written or oral texts without simplifying their level of difficulty. Willis (1996, p. 23) has also suggested that teachers need to design communicative tasks and activities with "a suitable degree of intellectual and linguistic challenge and promote learners' language development as efficiently as possible." Moreover, tasks should incorporate a given feature of language form for students to convey meaning, including how to use the passive voice, comparative and superlative forms of adjectives, or the conditional sentences. Ellis (2006) has proposed the theoretical framework used in the present study. According to Ellis (2006), the TBLT methodology involves two types of processes. Firstly, there are processes that determine how to change activities included in the syllabus into actual lessons (lesson design) and the processes that address the teachers' and learners' participation and interaction in the lessons (participatory structure). Lesson design addresses the stages or components of a lesson in which a task is the main component. Ellis (2006) believed that in TBLT a lesson design includes three stages: pre-task, during a task, and post-task. The first phase, pre-task, encompasses a number of activities that teachers and students need to perform before starting the task. An example would be the time given to the students to plan how to perform the task.

Theoretical model of the study
The second phase, during a task, is associated with the task itself and includes some instructional alternatives; for instance, it specifies that the learners have to perform the task under the time-pressure or not. The final phase, post-task, includes follow-up activities associated with the task performance and focuses on language forms. Indeed, only the task performance phase is compulsory in TBLT, while the two other stages are optional. Therefore, the main component of the TBLT methodology is the task performance phase (see Table 1): As this study was carried out in Iran, Najjari (2014) believes TBLT methodology has been welcomed by Iranian scholars and teachers although its adoption has been exercised with caution. According to Carless (2012), task-supported language teaching (TSLT) is a more moderate version of TBLT which can have more positive outcomes especially because of the challenges faced by East Table 1. A framework for designing task-based lessons (Ellis, 2006, p. 80) Phases Examples of options A. Pre-task * Framing the activity (e.g. establishing the outcome of the task) * Planning time * Doing a similar task B. Task performance * Time pressure C. Post-task * Number of participants * Learner report * Conscious-raising * Repeat task Asian teachers who are forced to adopt the strong, more theoretical version of TBLT. Despite this, Lai (2015) believes it is not the approval of one methodology over the other, but the adaptation of the tasks that learners confronted with and the roles language learners perform in the existing language pedagogy. Thomas and Reinders (2015) further added another determinant, the social context, in which the language learners perform the tasks. That is why, more or less, we can say that TBLT is more concerned with the theory of learning rather than the theory of language. Integrated with the theory of learning, TBLT even encourages not competent language learners and make them more confident to activate and use whatever language they already have in the process of completing a task without being afraid of making mistakes (Motlagh, Sharif Jafari, & Yazdani, 2014). Indeed, TBLT is designed to inspire language learners to enhance their language competence by being involved in completing meaningful tasks.

Iranian learners
Since many people in the world use the English language for communication purposes, many Iranian students have started learning it as a foreign language from elementary schools and even from earlier times. Therefore, one might assume that these students are able to master four language skills, e.g. listening, speaking, reading, and writing when they are adults. However, Iranian learners master these skills with varying degrees. Some learners may have a good command of all four skills and some may be good at a particular skill at the expense of other skills.
Accordingly, writing in English is the skill that most Iranian EFL learners have many difficulties with. For instance, many learners who are able to express themselves verbally will run into difficulty when writing a simple paragraph or even a few sentences, so they will fail to write effectively in English. This may be mainly because of the way instructors teach writing using traditional ways in EFL classrooms in which they provide students with a topic and then ask the students to write one or more paragraphs about it at home. Finally, teachers evaluate students' writing assignments and give them some feedback. The problem is that writing tasks are not included or taken seriously in pedagogical activities until EFL learners reach higher levels. At lower levels, writing is limited to writing isolated sentences focusing on a particular grammatical structure. As a result, most students do not know how to brainstorm ideas, to organize them, to use cohesive devices in order to put together these ideas in the form of a piece of writing, to revise the writing draft, and to proofread the text.
Despite numerous studies conducted on the effectiveness of TBLT on Iranian language learners' skills (Fatehi Rad & Mohammad Jafari, 2013;Hayati & Jalilifar, 2010;Nourbakhsh Kolaei Yarahmadi, & Maghsoudi, 2013), only a few studies have addressed the impact of this method on writing skills of Iranian EFL learners (Marashi & Dadari, 2012). In addition, little research explored the effectiveness of TBLT on Iranian EFL learners' analytic writings. Accordingly, the present study aims to investigate how the employment of task-based language teaching methodology might affect Iranian EFL learners' analytic writing skills.

Research objectives
The main objective of the present study is to investigate the impact of task-based language teaching methodology on analytic writing skills of Iranian EFL learners.
Besides, the specific objectives of this study are as follows: (1) To find out if task-based language methodology of writing instruction can improve analytic writing skills of Iranian EFL learners in comparison with the conventional method.
(2) To determine if task-based language methodology can improve components of writing competence (content, organization, vocabulary, language use, and sentence mechanics) for Iranian EFL learners in comparison with the conventional method.

Design of the study
According to the objectives of the study stated earlier, researchers employed a quantitative experimental method to study two groups of participants with the intermediate level of English proficiency through an experimental group and a control group. Besides, the present study employed a quantitative experimental method in the sense that it used both experimental approaches and statistical analyses of quantitative data, including the comparison of experimental and control groups, and formal, systematic measurement of quantities, such as the numerical values assigned to the participants' writing ability and its different components.

Participants
The population of the study included all Iranian learners who were studying English as a foreign language at the intermediate level. To select the participants in this study, the researchers applied availability-sampling procedure from two English language institutes in Shiraz. To this end, they administered Oxford Quick Placement Test to the participants and selected those with a score of one standard deviation above and below the mean. Accordingly, the research sample included 80 EFL learners. Then, they placed the participants randomly into two groups as the control group and the experimental group. The randomization aimed to assign the participants to both groups without using any predetermined criterion; so that, they would be homogeneous as much as possible with regard to a given variable, their English proficiency. The number of the participants randomly assigned to the control group was 40 EFL learners (20 males and 20 females); similarly, the same number of the participants, 40 EFL learners (20 males and 20 females), were assigned to the treatment group. In addition, the participants' age was 15-20 and their native language was Persian.

Instrumentation
The instruments used to collect the data were a placement test and two comparable writing tests. Oxford Quick Placement Test contains 60 items that measure students' level of English proficiency. The test contains various subsections, including multiple-choice questions, cloze passages, and matching items, which checks testees' knowledge of vocabulary, grammar, and comprehension. Each correct answer provided by the participants received a score of one while each incorrect response was scored zero.
The researchers included the writing sections of two paper-based TOEFL tests as the pre-test and post-test. Each writing test included two writing tasks in which the participants received a topic to write a paragraph about. The total score for the writing test was 30. The participants' scores on the TOEFL writing test acted as a unit of measurement for their English writing ability to determine if taskbased language methodology and traditional approach led to any difference in writing skills of Iranian EFL learners. Before the administration of these two tests, two EFL professors reviewed the tests to remove any problems with the tests and to ensure their validity. Besides, the researchers applied Cronbach's alpha coefficient to test the reliability of the instruments.
The Analytic Rubric (Jacobs's et al. composition profile) was the applied framework to score the participants' writing samples. The scale assesses the candidates' writing ability using five traits, including content, organization, vocabulary, language use, and mechanics. The scores allocated to each trait were as follows: Content = 30, Organization = 20, Vocabulary = 20, Language use = 25, and Mechanics = 5. The total mark was 100 points.
In addition, to provide a rather objective assessment of each written passage in both pretest and posttest, two raters scored the students' writing assignments on the pretest and posttest based on the analytic rubric. To measure inter-rater reliability, the researchers applied Pearson product-moment correlation, based on average ratings on scores for the five individual components of each writing task.

Procedure
At the beginning of the study, the researchers administered the Oxford Quick Placement Test to the students in two English language institutes and selected the students who got scores one standard deviation above and below the mean. Then, they placed the selected students into two groups as the control group and the experimental group. Afterward, both groups took a writing test (pre-test) to measure their writing ability and to ensure that they were homogeneous in terms of their writing skills. The instructors taught the writing skills to participants in the experimental group using Willis ' (1996) TBLT Framework in which language teaching involves a task with three main stages: pre-task that involves an introduction to the topic, task preparation, task cycle that includes task performance, planning and report, and post-task (language focus) that deals with language analysis and practice.
To this end, the researchers divided the class time into three phases, including pre-task, task cycle, and post-task. During the pre-task phase, the instructor presented the topic and (one of the researchers) encouraged the participants to activate the related schemata and the background knowledge on the topic. She asked the students to brainstorm ideas and write freely about the assigned task without having concern for form. In addition, the instructor used scripts, charts, maps, and pictures to clarify concepts and elicit the participants' ideas. The task cycle included three stages: task, planning, and report. During the first stage, the instructor asked the participants to organize their ideas and write about the assigned topic. The participants worked in pairs or in groups based on the difficulty of the task and get assistance from their peers. The instructor also monitored and helped students to write what they intended to say without intervening to correct errors of form. During the planning stage, the participants rewrote and drafted their writing using analytic essay principles (simplicity, concision, directness, and objectivity). In addition, the instructor let the students benefit from peer feedback and dictionary during the writing phase. In the report stage, the instructor asked one or two groups to read their essays in class. During the posttask phase, the instructor discussed the structure and organization of analytic writing with the students as the language focus phase (post-task phase) with practice on the cohesive devices, grammar, content, fluency of ideas, and word choice. The treatment lasted eight weeks, each week two sessions (16 sessions in total). In addition, each session lasted 90 min.
In contrast, the participants in the control group learned writing skills through the use of conventional methods of teaching writing. For example, the instructor assigned a topic to students and asked each student to write a passage about. Then, the instructor checked the written assignments, gave the students some feedback, and finally, assigned a score. The instruction course for the participants in the control group also lasted eight weeks, each week two sessions (16 sessions).
Upon the completion of the treatment, the instructor administered a writing test as the posttest to the participants in both the control group and the treatment group to find out how their writing ability has improved during the instruction and if there was any difference between the writing competence of the two groups. The participants in both groups did the test in 20 min.

Data analysis
In order to answer research questions, the researchers codified and analyzed the collected data by SPSS Software (Version 19). To answer the first research question (To find out if task-based language methodology can enhance analytic writing skills of Iranian EFL learners in comparison with the traditional method), the researchers applied descriptive statistics including mean and standard deviation and inferential statistics including independent samples t-test. The aim was to ensure the homogeneity of the participants concerning their writing skills before conducting the study and to find the effects of the treatment on the participants in the experimental group.
In order to answer the second research question (To determine if task-based language methodology can improve components of writing competence [content, organization, vocabulary, language use, and sentence mechanics] for Iranian EFL learners), in comparison with the traditional method, the researchers applied frequency analysis and ANOVA.

Results
This section presents the results of the data analysis to make it clear to what extent the implementation of task-based language teaching affected the writing skills of Iranian EFL learners.

The effect of task-based instruction on writing skills of Iranian EFL learners
The first objective of the present study was to find out if task-based language methodology and conventional method of writing instruction could lead to any possible differences in analytic writing skills of Iranian EFL learners. To this end, inter-rater reliability of the scores given by two raters were examined. The reliability coefficient for total score was .674 followed by .658, .611, .668, .666, and .752 respectively for components, including content, organization, vocabulary, language use, and sentence mechanics. This shows there was a relatively high consistency between the two raters concerning the way they scored the participants' pragmatic knowledge and its components. In addition, the correlation coefficients were all significant (P < 0.05), implying that the two raters were significantly consistent in scoring the writing tasks by the participants.
The mean score assigned by the two raters to the control group's performance on the writing pretest is 8.17 (out of 20) and the mean score assigned by the same raters to the experimental group's performance on the writing posttest is 8.10 with the standard deviation of 1.19 and 1.17 respectively. What these figures indicate is that both groups had a rather poor performance on the writing pretest. To see if this difference among mean scores of two groups is significant, the researchers applied independent samples t-test as presented in Table 2: As it can be seen in the above table, there is no significant difference between the performance of the control and experimental groups on the pretest (P > 0.05), suggesting that the writing ability of both groups was at the same level before conducting the treatment. Table 3 shows descriptive statistics for both control group and the experimental groups concerning the components of their writing samples on the pretest.
The mean scores of the two groups for different components of the writing samples range from 1.42 to 1.74. However, no significant differences seem to exist between the two groups in terms of their scores for different aspects of their writings in the pretest. To ensure this, the researchers applied the independent samples t-test as shown in Table 4: In the above table, as the significance levels are all greater than 0.05 (P > 0.05), the differences between the two groups in terms of the components of the participants' writing samples on the pretest (content, organization, vocabulary, language use, and sentence mechanics) are not significant. In other words, the Iranian EFL learners in the control and experimental groups were homogeneous in terms of both their writing ability and different aspects of their writing samples before conducting the treatment.
To investigate whether task-based writing in comparison with the conventional method had a stronger impact on the writing skills of Iranian EFL learners, the posttest mean scores of the participants in the control group and experimental group were compared. The experimental group (mean = 12.64, SD = 1.23) outperformed the control group (mean = 9.01, SD = 1.02) in posttest. To ensure if the better performance of the experimental group is statistically confirmed, independent sample t-test was conducted as shown in the following table.
As depicted in Table 5, there is a significant difference between the performance of the two groups on the writing posttest (P < 0.01), so the experimental group significantly showed higher writing competence in posttest than the control group.

The effect of task-based instruction on components of writing competence of iranian EFL learners
The second objective of the present study was to determine the effectiveness of the task-based instruction in improving the aspects of writing competence (content, organization, vocabulary, language use, and sentence mechanics) among Iranian EFL learners in comparison with the conventional method. To this end, the researchers compared the posttest mean scores of the participants in the two groups concerning different aspects of their writing samples as depicted in the following table.
As it is evident in Table 6, mean scores of the participants in the experimental group for different components of the writing samples are greater than mean scores of the participants in the control group. To understand if it can be claimed that the participants in the experimental group showed more improvements compared to the students in the control groups concerning different aspects of their writing samples, independent sample t-test was run as shown in Table 7.
The values of significance level in the above table are all less than 0.05 (P < 0.05). This means that differences between the two groups in terms of different aspects of writing ability as measured through the writing posttest are significant. Therefore, the various aspects of the writing competence of the Iranian EFL learners who practiced writing skills using task-based techniques improved significantly in comparison with the writing skills of the Iranian EFL learners who used the traditional writing exercises. This shows that the use of task-based writing techniques improved the Iranian EFL learners' ability in different aspects of the writing competence.

Discussion
The results of the study indicated that there were significant improvements in the writing ability of the participants in the treatment group after they practiced writing skills using TBLT techniques, showing TBLT effectiveness for performing writing tasks in comparison with performing conventional writing exercises in the classroom. This result supports findings of another study that indicated TBLT improved EFL learners' writing competence and language proficiency significantly because it had given students an opportunity to be more active and confident in using the target language in an enriching learning environment (Min, 2014). According to Al-Mahrooqi, Thakur, and Roscoe (2015), task-based writing far outweighs present-practice-produce pedagogies in that it benefits learners in terms of lexical phrases, collocations, and language forms.
It was also noted that the various aspects of the writing competence of the Iranian EFL learners who practiced writing skills using task-based techniques improved significantly in comparison with the writing skills of the Iranian EFL learners who used the traditional writing exercises. This finding is consistent with the results of previous studies (e.g.,   2013), which investigated the effectiveness of task-based language learning. These studies confirmed effectiveness of TBLT in improving learners' opinions toward reading-to-writing English instruction, enhancing language classroom interaction, developing English reading comprehension skills, elevating the students' writing and reading comprehension ability, increasing the students' potential capability in English learning, particularly their reading, writing, speaking, and listening, promoting listening self-efficacy, and etc. For example, by applying meta-analytic techniques for 50 current studies, Bryfonski and McKay (2017) revealed that TBLT not only have short time benefits in a variety of learning outcomes but it has also promising and significant effects in the long run. In another study carried out by Valli and Priya (2016), it was reported that out of all four basic language skills, writing is considered the most challenging one that can benefit from positive effects of TBLT in improving learners' writing abilities and language competence.
Confirming the usefulness of TBLT for improving the writing skills of the Iranian EFL learners, the results of this study are also generally in line with the results of the past research concerning the effectiveness of task-based writing. For instance, Birjandi and Malmir (2009) showed the effectiveness of TBLT in the narrative and expository writing of Iranian EFL learners in comparison with the traditional approach. In addition, Bantis (2010) observed the effectiveness of task-based writing methodology on learners' writing skills. In the same way, Marashi and Dadari (2012) demonstrated the positive effects of task-based writing on students' writing performance and creativity. Moreover, Ashari Tabar and Alavi (2013) showed the positive impact of task-based writing on intermediate EFL learners' writing performance in personal and decision-making tasks. Furthermore, Payman and Gorjian (2014) showed the improvement of writing skills of students majoring in translation through task-based strategies. In another study conducted by Kakh and Wan Mansor (2014), students who were exposed to audience-focused tasks (the ability of writers to address their audience needs) were interviewed after six months. The results showed that due to the extensive disciplinary-based writing tasks the students handled, students could manage to produce more well-organized discursive paragraphs and could also improve their writing in terms of cohesion regarding repetition of keywords, and the use of pronouns, and connectors.
Despite everything, Hai-Yan (2014) conducted task-based teaching of writing for over-crowded classrooms in China. The results revealed that sometimes language learners' understanding of teachers' roles in task-based writing classes falls short of expectations. Especially in Asian settings in which teachers are still considered "sage on the stage", students prefer to be told what to do rather than assuming responsibility for their actions. This, it seems, was quite apparent in the traditional Chinese attitude towards teaching and teachers' role in Hai-yan' (2014) study. Carless (2012) believes in Confucian-heritage settings, that is TBLT needs certain adaptations based on the demands of the host context. Some hidden findings of the present study suggested that some aspects of the writing ability of the Iranian EFL learners in the control group, such as language use and sentence mechanics increased significantly as a result of practicing writing skills through the traditional writing exercises, while some other components including content, organization, and vocabulary did not show any significant improvement. One possible reason why the traditional methods of teaching writing led to the improvement of writing components, such as language use and sentence mechanics in this study is that the traditional writing classes mostly emphasize such features than other features. For instance, language teachers tend to give more weight to word order, verb tense, and subject-verb agreement (language use) and spelling, punctuation, capitalization, and paragraphs (sentence mechanics). On the other hand, such classes focus slightly on the notions of which ideas to include and how to arrange these ideas. In addition, the focus of instruction is mainly on language forms rather than the meaningful use of language for real communication purposes, which is at the heart of TBLT approaches (Ellis, 2006;Nunan, 1989). In contrast, all aspects of the writing competence of Iranian EFL learners, including sentence mechanics, language use, vocabulary, content, and organization improved significantly because of the implementation of TBLT techniques. In addition, language use and sentence mechanics were the two traits of the writing competence, which showed more improvements than other aspects for both the control group and the experimental group. Nevertheless, a comparison of the participants' posttest mean scores indicated that the same traits showed significantly more improvements among the Iranian learners in the TBLT group than those in the control group, demonstrating the effectiveness of TBLT for teaching these aspects compared to the traditional method. However, it's worth mentioning that previous studies did not manipulate different aspects or components of students' writing skills and thus did not show how TBLT could affect such aspects. Therefore, the present study is rather new in this regard.

Conclusion
As the results of the study indicated, the employment of task-based language learning techniques could enhance the writing ability of the Iranian EFL learners in comparison with those learners who had practiced writing skills using the traditional method of drafting a piece of writing and having it corrected by the instructor. This is because task-based instruction creates a natural language learning context and writing tasks may help to express meaning and transmit messages across situations that resemble real-world conditions (Nunan, 1989). In addition, writing tasks have clear functions and context with unified language skills that are similar to those skills needed for establishing real communications in everyday situations.
It was also observed that various aspects of the writing competence (content, organization, vocabulary, language use, and sentence mechanics) improved significantly among the Iranian EFL learners who practiced writing skills using TBLT techniques in comparison with the writing skills of the Iranian EFL learners who used the traditional writing exercises. In fact, tasks are meaningful activities that students do to communicate to each other. In addition, as they have a clear outcome, instructors and students know whether communication has taken place successfully or not. When the students try to perform a task, they need to work to understand each other and to express their own ideas. As it was stated by Willis (1966), task-based language learning includes the pre-task, the task cycle, and the post-task cycle that is the language focus in which students can pay close attention to language structures and task components.