The effect of structured versus unstructured collaborative pre-writing task on writing skills of the Iranian EFL students

Writing is considered a very complicated task for many EFL students and Iranian EFL learners have many problems in this regard. Present study through a mixed method design, attempted to investigate the effect of Structured Collaborative (SC) pre-writing task versus Unstructured Collaborative (USC) pre-writing task on the writing ability of Iranian EFL students. Among a population of 300 freshmen learners of English language translation of Islamic Azad University, 169 students were selected based on criterion sampling. The criteria for selecting the sample were English language proficiency of the learners, age of the participants and major of their study. On the other hand, 30 teachers were selected based on convenience sampling for the interview part of the study. Then the students were assigned to Unstructured Collaborative (USC), and Structured Collaborative (SC) pre-writing groups. These two clusters were considered as experimental group and as control group. The study was implemented over a period of 16 weeks and involved pre and post-tests. Results of the quantitative data analysis shows that students’ writing proficiency in both groups were improved. However, the outcomes of the study reveals that the (USC) group outperformed the (SC) group. The qualitative data analysis through classroom observation and teachers interview reveals that teachers of (IAU) practiced the collaborative tasks at the three stages of writing in their classes and they also favored and applied the five components of Cooperative Learning (CL). In this study, the researchers took the initiative to make a distinction between structured and unstructured collaborative pre-writing tasks. Consequently, the results of the current study are of benefit for several groups of people, namely language practitioners, university students and educational administrators.


Introduction
It was mainly after the communicative movement that writing found its true position in language teaching (Rashtchi & Keyvanfar, 2010). Writing is considered a difficult skill which requires learners to use their related skills and sub-skills. To write an English essay, learners require knowledge of grammar and terminology regarding the associated theme (Fitz & Glasgow, 2009). The capability for writing efficiently is very important in several academic settings. Furthermore, developing the writing capacity of the students in different parts of the world is assuming a significant portion in second language studies (Ghoorchaei, Tavakoli, & &Nejad Ansari, D., 2010). In English as a Foreign Language (EFL) settings, writing has always been regarded as a main skill, since it inspires rationality and forces learners to focus and shape their thoughts (Maghsoudi & Haririan, 2013), thus Second Language (SL) or Foreign Language (FL) learners must definitely pursue means with which develop their writing ability to deal with the demands of real-life (Sadiku, 2015). Regarding this issue, in the Iranian context, EFL learners have shown an increasing apprehension towards commencing communication and applying educational and occupational occasions by means of written communication with people all over the world (Gholaminejad, Moinzadeh, Youhanaee, & Ghobadirad, 2013). Regarding the dominant role of writing in today's world, the essence of increasing writing ability in the arena of SL/FL education no longer stimulates argument among academics (Merkel, 2018). On the other hand, it is an agreed-upon subject that writing skill has a vital role in education and English language development (Steinlen, 2018).
Collaborative tasks generate learning opportunities when students interchange meaning, suggest feedback and offer enhanced output for communicating the meaning (Gass, 2003;Mackey, 2012). Collaborative tasks through learning together is a proper choice for EFL learners and highlights dynamic interaction between learners with diverse capabilities and background knowledge. Additionally, it would result in more positive outcome in learners' social behavior and academic achievement (Azizinezhad, Hashemi, & Darvishi, 2013). According to Storch (2005) collaborative writing promotes the excellence of text and improves learner inspiration (Swain & Lapkin, 1998), improves their information (Donato, 1994), and enhances the focus on discourse, grammar and vocabulary use (Swain & Lapkin, 1998). It is believed that group doings amend students' engagement over group activities (Fernández Dobao, 2012;Shehadeh, 2011).
In this line (Fernández Dobao, 2012;Wiggleworth & Storch, 2009) evaluated the effectiveness of collaborative writing activities. Their outcomes have shown that students who were involved in group activities produced more precise manuscripts compared to the learners who performed individual writing tasks. Some studies observed the efficiency of corrective feedback on second language learners' texts (e.g. Storch & Wigglesworth, 2010;Yahyazadeh Jelodar & Farvardin, 2019). Results of their study showed that learners were affected by the feedbacks which they received in pairs. Moreover, the findings revealed significant consequences regarding the collaborative pre-writing activities over the participants' fluency and accuracy of their writings.
Besides, Mazdayasna and Zaini's (2015) study revealed that students who were engaged in collaborative pre-writing activities performed better than the students who worked individually. The findings confirmed the importance of pre-writing activities in developing writing ability of Iranian EFL learners. In the same vein (Ameri-Golestan & Dousti, 2015;Jafari & Nejad Ansari, 2012) studied the consequence of collaborative task on students' writing skills. The result showed that the students in the collaborative writing group outdid the other students, which highlighted the significant role of collaboration among Iranian EFL writing classes. In contrast, Hashempour, Rostampour, and Behjat (2015) investigated the influence of using brainstorming as a pre-writing plan. Outcomes of their study showed no significant correlation between brainstorming, its subgroups and EFL students' improvements over their writings.
Writing process requires the application of many cognitive and linguistic strategies. In Iranian EFL context, several learners complain about the absence of ideas and they claim that they cannot think of anything motivating or important to write about it. Majority of Iranian EFL instructors in academic contexts are frequently confused by the difficulties over their writing courses and could not discover an operative method to set learners minds working (Maghsoudi & Haririan, 2013).
Based on the extensive literature which has been conducted by the researchers of the study, and the need for a research in the Iran EFL setting, the researchers of the current investigation took the initiative to focus on pre-writing collaborative tasks and their effects on Iranian EFL learners. To this aim, the researchers of the current investigation used two innovative terms for pre-writing tasks, namely structured collaborative prewriting task and unstructured collaborative pre-writing tasks.
Furthermore, in the current study, not purely collaborative tasks which have been done by the help and involvement of language instructors have been named unstructured collaborative tasks which involved students-students' interaction along with teacher-students' interaction (e.g. Biria & Karimi, 2015;Lee, 2013;Maarof, Yamat, & Li, 2011;Shi, 1998;Yahyazadeh Jelodar & Farvardin, 2019). The notion of students-students' along with teacher-students' interaction supports Vygotsky's Zone of Proximal Development (ZPD) and the notion of scaffolding. Scaffolding in teaching considers (ZPD) as a transfer of accountability for the task implementation to the student. This perspective emphasizes the crucial role of collaboration between novice and expert students or the instructor and the students in knowledge construction (Daniels, 2001). In line with the aforementioned theories, Meeuwisse, Severiens, and Born (2010) and Schallert et al. (2015) highlighted the positive effect of the students-students' and teacher-students' interaction in academic contexts.
To address the gap, this study aimed to investigate whether structured collaborative pre-writing tasks or unstructured collaborative pre-writing tasks could contribute to improving EFL learners' writing proficiency. The outcomes, as expected by the investigators, might present a substitute method for writing instruction.
To achieve the goals of the current investigation, following questions are examined by the researchers: RQ1. Can structured collaborative pre-writing tasks improve Iranian EFL learners' essay writing skills? RQ2. Can unstructured collaborative pre-writing tasks improve Iranian EFL learners' essay writing skills? RQ3. What are the teachers' perceptions about structured collaborative pre-writing tasks and unstructured collaborative pre-writing tasks?

Methods
In the current investigation, the researchers used mixed method approach to gather the required data. Triangulation, which examines the convergence of evidence from different methods through using more than one particular approach for achieving richer data was as also applied in this study. Also, the classroom observation scheme was used to investigate collaborative task implementation and semi-structured interviews were conducted to shed light on instructors' viewpoints towards collaborative task implementation in higher educational context. The observation which was used in the current investigation, was originally developed from the unpublished dissertation entitled; Megnafi (2016). Teaching writing through collaborative activities: The case of first-year LMD students at university of Tlemcen. Unpublished master's thesis: University of Tlemcen. In the original observation scheme, the researcher used three main steps of writing, namely pre-writing, while-writing and postwriting stages. In order to make observation scheme applicable in Iranian EFL context, the observation scheme was adapted based on the language teaching context of Iran. Furthermore, the tasks which Iranian teachers usually use in their classes were added. In order to evaluate validity of the observation scheme, five experienced university professors of TEFL were asked to check the validity of the instrument. Results of the face validity approved the face validity of the observation scheme which was 82.6%. Furthermore, the researchers examined inter-coder reliability of the observation scheme as well. To this aim, two encoders who were adequately proficient in collaborative tasks accompanied the researchers. The correlation coefficient was .94 which was acceptable. After examining and approving the reliability and validity of the observation scheme, it was used for data collection procedure. It should be noted that six classes were observed.

Participants
In order to select a sample as participants of the study, the researcher used criterion sampling which is a purposeful sampling of cases on fixed criteria, such as scores on an instrument. In the current investigation, among a population of 300 Iranian freshmen English translation students in Islamic Azad University (IAU), Southeast Tehran branch and North Tehran branch, 169 students were selected as participants of the study. The researchers considered Oxford Placement Test (OPT) score as well as Krejcie and Morgan's (1970) 's table 169 participants out of 300 between 19 and 27 years old have been selected as the participants of the current investigation. Moreover, 6 teachers' classes were observed during the educational period. Teachers who participated in the study were fully informed about the objectives of the study and the two key concepts which the researchers focused on them namely structured and unstructured pre-writing tasks.
Besides, in order to obtain a comprehensive perspective of the teachers' writing, their tentative problems as well as their attitudes concerning the Cooperative Learning (CL) application in their classes, the researchers used convenience sampling for the interview. To this aim, all of the writing teachers in the IAU of North Tehran Branch and Southeast Branch were asked to participate in the semi-structured interviews and out of 38 teachers, 30 of them agreed to participate in the interview part. The teachers who participated in the interview were PhD students and PhD holders in TEFL with more than 5 years' experience in writing instruction who were fully informed about the concepts of Cooperative Learning (CL) and collaborative task implementation.

Instruments
Oxford Placement Test (OPT) was used as the first research instrument (Allen, 2004). This test is a validated placement test published by Oxford University Press and it is usually used to evaluate respondents' language proficiency level. Second instrument of the study was a writing pre-test. Third instrument was a writing post-test. Written texts were corrected based on the correction scheme adapted from Hedgcock and Lefkowitz (1992). The fourth instrument was an observation scheme developed by the researchers and fifth instrument was semi-structured interviews.

Design and procedure
In the current investigation, the researchers used mixed method approach or triangulation technique (Ary, Jacobs, Sorensen, & A., 2010). To this aim, firstly quantitative data collected by using pre and post-tests, then qualitative data were gathered to see whether they support the quantitative results or not. Finally, by using triangulation technique, results of two parts were reported.
The researchers used quasi-experimental design and they had some criteria for selecting participants of the current study such as learners' language proficiency, their major of study and age range. Three major phases, namely pre-test, treatment, and post-test were employed in this study. The 300 students in IAU were given the Oxford Placement Test. Following the correction of the papers, the participants whose obtained scores were intermediate, were nominated as the respondents of the current investigation. Based on Krejcie and Morgan (1970) a sample of 169 subjects, which were around 6 classes, were selected. The classes met for one session a week with a 90-min duration over a period of 16 weeks. Participants studied "Academic Writing book from Paragraph to Essay" (Rumisek & Zemach, 2010), through which the learners studied about authentic texts, summary writing, grammar and text organization. Students also had 20 to 30 min of writing tasks based on determined topics. The learners were asked to write an essay of about 250 words under time limitation in each session. During practice sessions, students went through the stages of pre-writing planning, post-writing checking and after-writing reflection. During practice sessions, different types of writing tasks have been used namely, argumentative, cause and effect, opinion and descriptive tasks.

Writing pre-test and post-test
Writing topics were selected by the researchers. The pre-test writing was on "generation gap" and post-test writing was on "the secrets of longevity". Then, the classes were randomly assigned to structured and unstructured collaborative clusters. Similar to the practice sessions, the students in both groups wrote their compositions in 30 min. The compositions were examined according to the aforementioned correction system by the researchers and an experienced English language instructor. The average of the two scores for each writing considered as learners' final grade. The correlation between the two ratings was computed through Pearson product moment correlation coefficient formula and the finding revealed a high inter-rater reliability index for pretest(r = 0.90) and post-test (r = 0.83).

Structured collaborative group (SCG)
Eighty-five participants of Structured Collaborative Group (SCG) were divided into 17 groups; each group included five members. In this group, students experienced structured collaborative pre-writing task. One week before the study, participants in this group were given some information about collaborative pre-writing tasks such as participants' roles, group dynamics and decision-making tasks. In the SCG, ten structured pre-writing tasks were created by the researchers which were adapted from McDonough and Neumann (2015). Participants wrote about ten similar topics which were used in both structured and unstructured collaborative groups. Each structured prewriting task had three parts. A declaration of the text theme and related issues over the book, a part for writing down thoughts, and a part for arranging a simple bullet plan. After students individually brainstormed some ideas and wrote those ideas, they could collaboratively work together to share their beliefs. The instructions directly specified that they must provide some feedback to their peers regarding the suitability of the thoughts concerning the writing assignment, and they were asked to include a column for writing and the evaluative explanations they received.
As they completed exchanging and evaluating their thoughts, the learners separately completed a bullet outline on their handouts. The last activity was to share their plans and to give feedback to each other. Then groups wrote individually in the classroom which took about 30 min. Finally, the instructor collected the compositions and corrected them according to the aforementioned correction scheme modified from Hedgcock and Lefkowitz (1992). Based on this correction scheme, the evaluation of content was as flows, 27-30 was excellent to very good, 22-26 was good to average, 17-21 was fair to poor, and 13-16 was very poor. Concerning text organization, the evaluation was as follows, 18-20 was excellent to very good, 14-17 was good to average, 10-13 was fair to poor and 7-9 was very poor. Besides, the evaluation of grammar was as follows, 22-25 was excellent to very good, 18-21 was good to average, 11-17 was fair to poor and 5-10 was very poor. Moreover, the evaluation of vocabulary was as follows, 18-20 was excellent to very good, 14-17 was good to average, 10-13 was fair to poor and 7-9 was very poor. Additionally, the evaluation of mechanics ranged from 5 which was excellent to very good, 4 was good to average, 3 was fair to poor and 2 was very poor. By considering the improvement of content, organization, grammar, vocabulary, and mechanics the instructor evaluated the essays and wrote her comments on various aspects of writing.

Unstructured collaborative group (USCG)
Eighty-five participants experienced unstructured collaborative pre-writing tasks. In this group, the Charette method of brain storming and Mind mapping was used as collaborative pre-writing tasks. In this group, before the writing activity, students were informed about topic of writing and Charette method of brain storming that is a stagebased method (Christmas, 2011) and Mind Map was applied. To this aim, the topic of writing was broken down into smaller questions. Then students were given 5 minutes to individually brainstorm ideas on the questions. Subsequently, they discussed their ideas in small groups. The class divided into groups with at least five members. Each group selected a reporter whose job was to summarize the thoughts generated collaboratively by that group. The reporters reported the ideas developed by the group members to the class. This stage was constructive and none of the ideas of the groups was rejected. Then, the instructor arranged the ideas according to the suggestions given by all of the students and drew a mind map on the board. Then, the students wrote about 30 min. In the USCG, the interaction was between the students-students and teacherstudents while in the SCG, the interaction was mainly between students-students. At the final stage, the instructor corrected the compositions based on the aforementioned criteria and wrote comments on various aspects of the students' compositions. In both groups the instructors graded the writing of each student. It should be noted that 2 weeks before study, teachers were given some information about structured and unstructured collaborative pre-writing task implementation in writing classes.

Results
The collected data were analyzed with statistical software SPSS version 22. Statistical methods applied in the data analysis process in this investigation comprised of paired sample t-test, independent sample t-test, descriptive statistic, standard deviation, frequency, percentage and mean index. Descriptive statistics specified that there was no statistically significant difference between the means of the two groups on the writing pre-tests, (SCG) (M = 16.49, Sd. = 1.32) and (USCG) (M = 16.22, Sd. = 1.36). To answer the research questions, paired samples t-test was run after the treatment. The descriptive statistics regarding the writing post-test specified that there was a statistically significant difference between mean of two groups on writing post-tests (SCG) (M = 17.91, Sd. = 1.27) t = 21.691 and (USCG) (M = 18.80, Sd. = .84), t = 26.078, p < .05. In fact, the unstructured collaborative pre-writing task was more effective than structured collaborative pre-writing activity in developing the respondents' writing skill.

Comparison between pre-test and post-test of the structured collaborative group (SCG)
In this section, the descriptive statistics regarding the pre-test and post-test of the SCG is shown.
To compare the pre-test and post-test writing scores of the participants' structured collaborative tasks, the mean of their pre-test and post-tests was compared. According to the data presented, the mean of the learners' scores in the pre-test of writing was 16.49 while their post-test mean was 17.91. Results of this section shows improvements in the writing performance of the students between the pre and post-tests.
A paired sample t-test was run to compare pre-test and post-test results of the SCG group. Based on the data presented in tables 1 and 2, there was a significant difference between unstructured (M=18.80, SD=0.84) and structured post-test (M=17.91, SD= 1.27), while t (167) = 5.36, P= 0.00, two-tailed, P < .05. Based on the data presented in Tables 3 and 4, the Mean Index of the post-test is higher than pre-test and the p-value was below the required cut-off scores of .05. Consequently, it can be concluded that there is a significant difference between the pre-test and the post-test.
Comparison between pre-test and post-test of the unstructured collaborative group (UCG) 'In this section, results of the study regarding the participants' performance in the (USCG) is presented in Table 5. To this aim, the results of the descriptive statistics of the pre-test and post-test as well as the paired sample t-test of their differences is presented.
According to the data analysis, the mean of writing pre-test of the USCG was 16.22, while the mean of the post-test was 18.80, which shows a significant improvement in their writing performance.
A paired sample t-test was run to compare participants' performance regarding the pre-test and post-test of unstructured collaborative writing ability. Based on the data obtained, there was a significant difference between pre-test (M = 16.22,   Table 6, the Mean Index of the post-test is higher than the pre-test, and the p-value is below the required cut-off scores of .05. Consequently, it can be concluded that there is a significant difference between the learners' performance concerning their pre-test and post-test of the unstructured collaborative writing.

Comparison between the post-tests of structured versus unstructured collaborative groups
In order to compare the effectiveness of the structured versus unstructured collaborative pre-writing task, an independent sample t-test was run to compare their related post-test results. In this section, the results of this part are presented. As revealed in the data analysis, the mean of the unstructured cluster regarding the pre-writing collaborative task was 18.80, while the mean of the structured group is 17.91. The results of this section reflect a significant difference between the post-tests of the structured versus unstructured group, in which unstructured group had a better performance.
Based on the data presented in Table 1, there was a significant difference between unstructured (M = 18.80, SD = 0.84) and structured post-test (M = 17.91, SD = 1.27), while t (167) = 5.36, P = 0.00, two-tailed, P < .05. Based on the data obtained from the participants of the current investigation, there is a significant difference between the consequences of the structured and unstructured collaborative post-writing tasks. The results signify that students had a better significant performance on the post-test of unstructured tasks group and this type of task helped them to improve their writing competence.

Results of classroom observation
After reviewing some established observation schemes and noting some classes, a classroom observation scheme was planned based on literature on collaborative theory and process writing. The purpose of the observations was to examine to what extent and how the instructors employed collaborative task and process writing in their writing classes. Field notes aided the goal of recording how and when the instructors follow three stages of writing and how to apply collaborative strategy during writing tasks. The data gathered by means of observation scheme were examined descriptively determining the extent of significance placed on each stage of teaching writing. Each observation was conducted during a 90-min instructional period at the IAU, North and Southeast branches, both of which are located in Tehran, Iran. The students in both universities major in English language translation. The observers were three English instructors from IAU North Tehran branch. To observe classes in each university, one of the researchers was accompanied by two experienced instructors. Regarding confidentiality issue, the teachers' last names were not used in the present study. In total, six instructors observed the classes, which were observed over a four-week period on two separate occasions. The observation over a fourweek period was scheduled with the intent to decrease the observant effect that there may have been on the classroom context, and is aimed at making the participants feel comfortable and perform more naturally in the classroom. Throughout the observation, the classes were recorded and observers used field notes. The analysis consisted of three steps: topic identification, category identification, and theme identification.  Instructors' attitudes regarding the collaborative tasks In the present study, the researcher applied thematic analysis to categorize themes with the data. Based on Ezzy (2002), coding is referred to as the process of "disassembling and reassembling the data" (p. 94). In this practice, the data is broken into small pieces of texts. Then, the researcher restructures the units by classifying them to elucidate the data. Researchers identified three major themes; "pre-writing stage", "while writing stage" and "post writing stage". ' Table 7 shows instructors' instructional practices during the three stages of writing through application of collaborative tasks.
According to the findings of the study, instructional practices of the instructors were in line with three stages of writing based on collaborative strategy. The writing stage that received the highest rank among the three stages of writing was related to postwriting stage (M = 2.86), followed by while-writing stage (M = 2.69) and the pre-writing stage (M = 2.53).
Pre-writing stage ' Table 8 shows descriptive statistics for the observation at the stage of pre-writing.' Based on the data analysis regarding item 1 of observation scheme, the observers reported that most of the participants (66.7%) clarified objectives through providing task completion guideline, promoting decision making strategy and supporting learners' ideas before writing task through unstructured collaborative tasks.
For example, teacher 5 (Sara) from Southeast Branch first elaborated the advantages of group work in class setting for 5 min, then asked students to respect group members' ideas during collective work. An additional 15 min was allocated to practice thinking and sharing ideas. During this practice she trained students to focus on main ideas and consider sub categories related to main ideas.
In line with the results of observation, all of the interviewees (100%) believed in the advantages of group work in academic context. For example, teacher 5 (Sara) from Southeast Branch believed; "Group work can be effective method to stimulate students, encourage learning, and communication. But without planning, group work can frustrate students." The result of interview and observation regarding item 1 echoes Kagan (1995) outlook which highlighted that through collaborative task implementation, students' output would be improved and acquisition could be elevated if it happens in a supportive, motivating and communicative context.
Regarding Item 2 of observation, 50% of instructors encouraged group dynamic and students' interaction, established team goals, assigned responsibility and defined roles before writing task through structured collaborative tasks. On the other hand, 50% of instructors hardly ever practiced this strategy. In present study, such cooperative procedures in writing classes occurred through group members' feedback, where students  negotiated with their group members as contributors and receivers of feedback, during which joint problem solving was established.
In this regard, teacher 3(Fatima) from Southeast Branch established team goals through explaining the group behavior, teacher 3(Fatima) first talked about group common goals for 5 min then divided the class into groups. After that, she assigned responsibility to each member of the group. She asked one of the students in each group to be a reporter of the group activity and the others shared ideas together for 10 min at pre-writing stage. After sharing ideas, the instructor asked the reporter of each group to report back what members talked together. The instructor determined students' roles and asked each student to follow specific task. She motivated students for further participation to share new ideas. She moved through the class and talked with all groups in the class and praised their interaction through expressive words such as "thank you" "Excellent". Students interacted with each other before writing activity for 15 min and then shared their ideas with the instructor. Afterward, the instructor wrote all related ideas on the board.
In line with the results of observation, most of the interviewees (90%) believed in group work as an effective method to motivate students and decision-making skills. Additionally, in line with observation findings regarding encouraging students' interaction and providing interactive context in class, most of interviewees (80%) highlighted the dominant role of interactive based classes through group discussions.
For example, teacher 3(Fatima) from Southeast Branch mentioned; "Group activities as a basic portion of every classes will enhance mutual interaction among peers and inspire students for class participation and decision-making skills." The result of observation and interview regarding Item 2 echoes some researchers' perspective (Swain, 2010;Yeh, 2015) who elaborated the activation of learners' reasoning through students' interaction in classes' social groups. They believe that collaborative talk leads to better presentation during cooperative assignment enactment.
Concerning Item 3 of observation for pre-writing tasks, 50% of the instructors used group brainstorming (e.g. mind map, Charett method, think pair share and clustering) techniques through unstructured collaborative tasks, more than the other instructors.
For example, teacher 6 (Elham) from North Branch at the beginning of the class explained process writing for 10 min. She brought some copies to the class. The paper was about specific framework for writing tasks. It had the topic of writing and some group procedures that students had to follow. The instructor divided the class into groups, gave paper to all students and asked them to follow the written procedure. Then students read the topic and completed the provided summary table about the description of the writing topic, changes that occurred related to that issue and the consequences of the changes. After 5 min, the group discussed the writing topic and the provided summary table on paper together. They shared group ideas and completed the last part of aforementioned table based on the group members' feedback. Then each student decided which information will be mentioned in his/her writing and based on which order. Then each individual student made an outline and wrote them on the space provided on that paper within 10 min. After that they shared their works with each other and received their group members' feedback. Additionally, as they listened to their group members' outline, students gave feedback about whether the outline is well organized. During this activity, which was about 10 min, instructor walked among groups and encouraged them for further practice through applauding individuals' active participation in group task.
In line with the results of observation, most of the interviewees (75%) believed in brainstorming, clustering and summarizing as the dominant cooperative tasks in writing classes.
For example, teacher 6 (Elham) from North Branch mentioned; "Brainstorming as a collaborative activity inspires creative thinking and helps students in developing ideas. Additionally, implementing clustering through putting ideas into groups over students' interaction is effective strategy for group work in classes".
The result of interview and observation regarding item 3 supports Colantone (1998) viewpoint about pre-writing stage for strengthening students' thinking skills. Additionally, some researchers (Alkhatib, 2012;Maghsoudi & Haririan, 2013) considered brainstorming as an effective factor which helps students' interaction in generating good ideas.
Regarding Item 4 of observation about application of authentic educational resources such as textbook or other realia, majority of instructors (83.3%) used authentic resources during the educational time of the class as a warm up at the stage of prewriting.
Teacher 1 (Leyla) from North Branch used modified text in her class. She used an excerpt of a magazine about "saving environment". She read the text herself for 5 min, then asked students to discuss the topic and write an outline related to the clustered ideas within 15 min. After that students comment on their work for about 5 min.
Concerning Item 5 of observation, majority of instructors (66.7%) encouraged organization of ideas and individual learners' accountability through structured collaborative tasks at the stage of pre-writing.
Teacher 3 (Fatima) from Southeast Branch provided several opportunities for students practice together through planning their writing task to deal with their writing problems. She explained brainstorming in the class for 15 min. First, she asked students to think of a storm then explained that storm is thousands of drops of rain, all coming down together. Now, imagine thousands of ideas "raining" down onto your paper. Then, she asked students to take a look at the title of their textbook, which was about "advertisement", and asked them to brainstorm ideas about advertising. She highlighted that when you brainstorm, write down every idea that comes to you. She elaborated that they should not worry whether the ideas are good or silly, useful or not. Then, she asked students to work together in groups of five and gather as many ideas as they could.
In line with the results of observation, 30% of interviewees considered encouraging personal accountability through providing mixed skill group to enhance students' motivation in writing classes.
For example, teacher 3 (Fatima) from Southeast Branch mentioned; "Although together learning creates positive viewpoint towards learning, students should practice taking responsibility during group task implementation." The result of interview and observation concerning item 5 is in line with (Mizuki, 2003) which emphasized sharing responsibility and gradually shifting responsibility from the teacher to learners during educational time of the class and the benefit of this modification for language learning in EFL setting.

While-writing stage
' Table 9 shows descriptive statistics for the observation regarding while-writing stage.' Concerning item 6 of observation, at the stage of while-writing about implementing collaborative discussion, 50% of instructors considered students' interest and provided opportunity for discussion through mutual and active interaction based on structured collaborative tasks. Additionally, 50% of the participants applied this strategy through unstructured collaborative tasks.
For example, teacher 2 (Atiye) from North Branch conducted seminar-based class within 15 min. She asked students to take turns actively and participate in the collective discussions. Then through on-task interaction, at the stage of while-writing, students cooperated together relevant to the task at hand. The instructor did not participate in the discussion and only interrupted when necessary. She asked students during preparing a draft of writing in order to deal with writing problems and to ask their partners. If the group was not able to deal with the problems, they would ask instructor.
In line with the results of observation, 35% of interviewees believed in allocating resources and providing opportunity for effective communication and students' active participation.
For example, teacher 2 (Atiye) from North Branch stated; "Active participation strategies increase student engagement in the classroom. Students who are actively engaged will attend to the presented material and retain information being offered." The results of interviews and observations regarding item 6 echoes Berry (2003) perspective which highlighted cooperative task implementation as an approach to enhance students' learning, active participation, academic achievement and oral communication skills in class setting.
Regarding item 7 of observation, majority of instructors (66.7%) inspired learners rely on others' ideas and take responsibility for peer assessment through structured collaborative tasks.
For example, teacher 4 (Sahar) from North Branch encouraged a threat-free context to enhance positive competition. She talked with each group for 5 min and stood beside them and provided students with simple strategies to reduce their anxiety, she praised students for every new idea related to writing tasks. She was also involved in positive self-talk with those anxious students and stimulated pairing. Then, she helped groups to create a study schedule through defining task at the stage of pre-writing which took about 15 min. After that she gave students time to discover new ideas and to develop new information through note taking strategy.
In line with the results of observation, most of interviewees (75%) believed that group work leads to better correction and recall.
For example, teacher 4 (Sahar) from North Branch declared; "Through pooling of abilities when students work together through peer assessment they recall more." The results of interviews and observations concerning item 7 related to peer and selfevaluation are in line with Natri (2007) study, which suggested practical views of peer and self-evaluation in classroom context.
Regarding item 8 of observation at the stage of while-writing, observers highlighted that majority of participants (66.7%) faced very few challenges regarding gap between students due to their communication ability and the time limit for dealing with writing tasks through unstructured collaborative task implementation.
Teacher 2 (Atiye) from North Branch faced challenge regarding students' unequal participation during writing activity. During group task her class was noisy. She monitored groups and found that 3 students did not accompany group tasks and they did not follow group procedure. Then, she asked them self-select your partners and work with them. After that they changed their groups and followed group tasks with their new partners. Additionally, in order to manage chaos in class, teacher 2 (Atiye) talked with each group and let students ask their questions for 15 min. During task implementation, she asked the students several times to trust each other to achieve outstanding group work.
In line with the results of observation, 30% of interviewees considered implementing CL in class context may lead to chaos and conflict among group members in class.
For example, teacher 2 (Atiye) from North Branch revealed; "Application of CL is time consuming and it is difficult for timid and freshmen students." The results of interview and observation regarding item 8 support Ghaith (2018) study which highlighted teacher knowledge, crowdedness of the class, time management and school procedure as challenges related to cooperative task implementation in academic contexts.
Regarding item 9 of observation concerning cultivating open communication and keeping students on target, 33.3% of instructors applied this strategy based on unstructured collaborative task by supporting learners to deal with writing issues and 33.3% of them applied this strategy through structured collaborative tasks and the rest of participants (33.3%) practiced this strategy very little.
Teacher 4 (Sahar) from North Branch applied a friendly comment during writing task implementation and clarified how students can fix their writing problems through enhancing students' knowledge of the writing activity. She motivated students never worry about second drafts and plan for their writing. She prescribed free-writing strategy. For those who faced problems in her class, she asked them to write whatever comes to their head about the topic without stopping, which was completed within 20 min. She highlighted that freewriting would improve their writing fluency and mentioned that during free-writing don't stop if you make a mistake and just keep writing.
In line with the results of observation, 70% of interviewees believed in establishing team goals to keep students on target.
For instance, teacher 4 (Sahar) from North Branch said; "Setting goals for implementing group task is definitely important. Setting specific goals stimulates higher performance than setting none and can help us stay on our purpose." Concerning item 10 of observation related to removing the instructors' scaffolding phase by phase, 50% of instructors encouraged individual learners to plan the content of the writing task individually after consulting with their group members and eliminated their further support at the stage of while-writing through structured collaborative task.
For example, teacher 5 (Sara) from Southeast Branch valued students struggle as an acceptable procedure and encouraged students' interaction during task accomplishment which took about 10 min. She requested students to ask questions collectively and do joint problem solving during first draft preparation. She highlighted that if you face any problem you can ask me. The students were participants in their own learning, rather than observers of the instructor's knowledge. After task implementation within 10 min the instructor proudly displayed each group work throughout the classroom and said a thank you to each group task which was well done and completed.
Post-writing stage ' Table 10 shows descriptive statistics for the observation regarding post-writing stage.' Regarding item 11 of observation at the last phase of writing, none of the participants was motivated to evaluate each other's writing by means of error correction or feedback procedure (unstructured tasks). After students finished a writing assignment, the instructor hardly ever asked students work together to give comments to their group members.
For example, teacher 5 (Sara) from Southeast Branch after writing task, collected all papers and selected one of them without calling student's name and read that essay in the class. She added her comment on mechanics, cohesion, coherence and content organization within 15 min.
Regarding item 12 of observation or reassuring learners' self-reflection on how well they wrote based on their initial plans and how well they corrected their errors through proofreading their own writing tasks, 50% of instructors applied this strategy through unstructured collaborative task and 50% of the participants applied this technique through structured collaborative task less than the others.
For example, teacher 1(Leyla) from North Branch at the last phase of writing, made generic examples of writing essays and commented on that writing task for 15 min. She provided chance for students to compare the public examples to their own work and determine how well they dealt with their writing task. Additionally, she stimulated the students' self-reflexivity through asking students to think and reflect about their writing tasks. She highlighted that students think critically about what they have done within 10 min.
In line with the results of observation, majority of interviewees (90%) believed in students and teachers interaction as an effective method to develop critical thinking and reflection of students through decision making strategy.
For example, teacher 1(Leyla) from North Branch declared; "Reflection is a vigorous process of education which leads to better student performance in writing classes." The results of interviews and observations regarding item 12 echo Schraw's (2002) perspective on instructors monitoring process and their dominant role as a facilitator in promoting student's learning through improving learners' cognition of their own learning ability and feedback in writing classes.
Concerning item 13 of observation, all instructors noticed the learners' relevant need for further learning (unstructured tasks).
Regarding this issue, teacher 2 (Atiye) from North Branch also tried to share her goals before new lesson. She discussed learning intentions and criteria for success. Additionally, at the end of a class, she gave students a blank paper and asked them write responses to the following two questions: What was the most significant thing you learned today? What problem is still primary in your mind?
Additionally, she asked students to debrief their experiences with problem-solving during writing task implementation within 10 min.
In line with the results of observation, all participants (100%) believed in learner centered classes.
For example, teacher 2 (Atiye) from North Branch stated; "Through promoting trust, communication and collaboration between me and my students, a learner centered context is created in my class." The results of interviews and observations regarding item 13 support Krashen's (1982) Affective Filter Hypothesis concerning anxiety-free setting for learning which can promote student-centered classes and stimulate learners to produce more output and step towards superior presentation.
Regarding item 14 of observation, promoting connecting ideas to make a coherent piece of writing, majority of participants, (66.7%) applied this procedure in their classes through structured collaborative task.
Teacher 1 (Leyla) from NB, first defined coherence as a crucial quality for good academic writing. She highlighted that in academic writing, the flow of ideas from one sentence to the next should be smooth and logical. Without cohesion, your reader will not comprehend the main points of writing. She wrote some examples of cohesive devices on the board and provided some examples within 15 min. Then, she asked students to review their writing task and edit the cohesion between sentences.
Regarding item 15 of observation, about monitoring and evaluating the effectiveness of writing strategies and transferring strategies to new tasks, 50% of instructors provided learners with oral and written feedback through unstructured collaborative tasks and 50% of them applied this through structured collaborative tasks.
As an example, Teacher 3 (Fatima) from Southeast Branch facilitated learning process through providing written feedback on learners' final draft. Through written feedback she provided a record of what students did well, what needs improvement and suggested next steps. She wrote A+/A/B or C on students written task and when she was thrilled with a student's new on-task behaviors applauded that student's effort in class.
In line with the results of observation, majority of participants (70%) believed in group tasks as a main part of CL which leads to group feedback in an interactive context.
As an example, teacher 3 (Fatima) from Southeast Branch declared; "Group feedback will provide anxiety-free environment for learning." The results of interviews and observations regarding item 15 echo Küçükali (2017) study which revealed the positive effect of oral and written feedback on EFL learners essay writing skills.
Teachers' perceptions regarding structured versus unstructured collaborative pre-writing tasks According to the interview results, most of the teachers (76.6%) preferred unstructured collaborative pre-writing task while minority of them (23.4%) preferred structured collaborative pre-writing task.
For example, teacher 3 (Fatima) from Southeast Branch declared; "From my students' feedback it was clear that in unstructured group, task planning through the learners' interaction with my help, improved students' text organization. The students wrote longer text with more complex structures. They used implicit and explicit transitional devices. Besides, the hook and thesis statements were better developed and they could easily create a main impression about the topic of writing." Additionally, teacher 2 (Atiye) from North Branch highlighted the effectiveness of unstructured collaborative task; "Based on my experience, using unstructured collaborative task improved students' writing fluency and complexity. I also experienced when students put their knowledge together, they produce more accurate text due to collective knowledge. In their writing they considered what the audience knows and needs and provided supports. They mainly focused on the methods of development, word choice and punctuation. They also used better controlling ideas and categories to narrow down the topic of writing. Every sentence in their task was related to the main idea which has provided better unity in their writing. I personally believe that this task improved students' knowledge of grammar and organization." Additionally, teacher 1 (Leyla) from North Branch elaborated the effects of this task on different components of writing; "Concerning different components of writing I believe that unstructured collaborative pre-writing tasks can improve content, organization, grammar, vocabulary and mechanics of learners' writing skills. In my class I experienced that through implementing this task, the students used wider range of vocabulary and different devices such as quick examples, narrative examples, facts, statistics, details and analogies to develop ideas in their writing task and they built the blocks of their essay properly. I found that my students had better counterargument and problem solving. Besides, they considered opposing views in their text and had enthusiasm and energy from start to finish. I think that this task developed their understanding of vocabulary and mechanics of writing." In the same vein, teacher 5 Sara from Southeast Branch mentioned that unstructured collaborative task motivates students and engage them in learning process; "I believe that participating in group activity through unstructured collaborative task in Foreign Language teaching and learning, motivates students engage in their learning process and work together to attain their goals. It also encourages students' responsibility, relational skills and critical thinking skills in writing class. From my students' feedback it was clear that in unstructured group, all three phases of generating ideas, choosing the most suitable method of development and checking the errors were accomplished appropriately." The results of this part echoes some related previous investigations (e.g. Fatehirad et al.,2017;Fernández Dobao, 2012;Yahyazadeh Jelodar & Farvardin, 2019) about the positive effect of student-student interaction along with the interactions with their teachers in developing students writing skills.

Discussion
Several researchers focused on the effectiveness of collaborative pre-writing to improve students writing ability but researchers of the present study applied two innovative terms, structured collaborative pre-writing tasks and unstructured collaborative prewriting tasks. Through structured collaborative pre-writing tasks, we focused on students-students' interactions and in unstructured collaborative pre-writing tasks we considered students-students and teacher-students' interaction. In the current investigation, the researchers could not find any ideal expert to rate the writing tasks of the students because we used two innovative terms (structured collaborative pre-writing tasks & unstructured collaborative pre-writing tasks), and they were not familiar enough with the mentioned concepts. Consequently, the researchers decided to rate the tasks themselves which could be considered a limitation for the study.
The result of the study exposed that the unstructured collaborative pre-writing activity was more effective than structured collaborative pre-writing activity. However, the comparison of the pre and post-tests of writing means showed that both methods could improve participants' writing. The development of unstructured collaborative group's writing aptitude might be partly recognized to the practices of transferring and sharing ideas among students' collaboration with their instructors. Likewise, the Charette method of brainstorming and mind map appeared to facilitate writing process. In line with the previous investigations, collaborative pre-writing task seemed to facilitate learners' mental processes and aid learners to think and write more efficiently (Vygotsky as cited in Wertsch, 2007). Moreover, the student-centered technique could enhance autonomous learning process (Murphy, 2008). Group activities appeared to be dominant in evolving communication among the learners and delivered the chance to exchange meaning and work jointly before a writing activity. Findings of the study highlighted the role of brainstorming as a pre-writing task. It should be noted that brainstorming, as a cognitive toolkit at student disposal, allows learners to learn their peer groups' viewpoints and improve variety of thoughts (Christmas, 2011).
The outcome of the current investigation supported the opinions of the advocates who claimed the effectiveness of the students' interactions and their interactions with teacher in improving students' writing (e.g. Elola & Oskoz, 2010;Fernández Dobao, 2012;Maghsoudi & Haririan, 2013;Mazdayasna & Zaini, 2015;Storch, 2005;Yahyazadeh Jelodar & Farvardin, 2019). The findings are also in line with Shi (1998) which showed that prewriting negotiations provided shared contexts where the instructor scaffolded learners to intellectualize their thinking, or learners helped each other in clusters to generate various ideas for the writing activity. Furthermore, the findings supports McDonough and Neumann (2015) result who signified pre-writing activities assist learners to generate content and effectively establish their thoughts. Moreover, collaborative activities can lead to scaffolding among peers, in this case learners can trust their knowledgeable peers in writing which leads to better outcomes (Donato, 1994;Storch, 2005).
In line with quantitative results, 90% of interviewee believed in students-students and teacher-students' interaction along with students-students contribution, as an effective technique to develop critical thinking and reflection of students based on decision making strategy. Besides, 66.7% performance of teachers, regarding clarifying writing objectives over providing task completion guideline and supporting learners' ideas before writing task through unstructured collaborative tasks, were in line with quantitative results of the study.
The result of instructors' interview and class observation concerning group interactions at the stage of pre-writing echoes the view of Yeh (2015) who elaborated the activation of learners' reasoning through students' interactions in classes' social groups. According to Swain (2010) collaborative tasks engage students in joint problem solving and through mutual interaction, learners' cognition is stimulated which results in improved performance. In the present study, such cooperative procedures in writing classes occurred through group members' feedback, where students negotiated with their group members as contributors and receivers of feedback, during which joint problem solving was established.
The results of instructors' interview and class observations regarding individual learners' accountability during group task, are in line with studies which considered sharing responsibility in EFL setting (Cunningham & Carlton, 2003;Stephenson & Kohyama, 2003). These studies focused on progressively shifting accountability from the teacher to students during educational time of the class and highlighted the benefit of this modification for language learning. The results concerning self-evaluation are in line with earlier investigations that recommended practical tenets of peer and selfevaluation (Natri, 2007;Thomson, 1996).
Additionally, the results regarding instructors monitoring process and feedback in writing classes highlight the dominant role of instructors as an organizer in indorsing student's learning and cognition of their own learning ability which provides context aimed at sharing responsibility for better outcome through stimulating students having a responsible attitude towards their own learning process (Scharle & Szabó, 2000;Schraw, 2002). In this study, instructors stimulated students to monitor and apply selfregulation strategy regarding their thinking practices and raised students' capability to become their own evaluators, which clearly weight the learners' metacognition in writing classes through collaborative tasks (Poehner & Lantolf, 2003).

Conclusion
Results of the study regarding the first research question revealed that structured collaborative pre-writing tasks improved Iranian EFL learners' essay writing skills. The results are in line with the views of the supporters who highlighted the effectiveness of the students' interactions in developing their writing skills (e.g. Elola & Oskoz, 2010;Kuyyogsuy, 2019;Mazdayasna & Zaini, 2015;Neumann & McDonough, 2014a;Nikoopour & Aminifarsani, 2010;Wiggleworth & Storch, 2009). Additionally, results of the study concerning the second research question revealed that unstructured collaborative pre-writing tasks improved Iranian EFL learners' essay writing skills. Findings of the second research question are in line with the views of the advocates who highlighted the efficiency of the students' interactions and their collaboration with their teachers to improve their writing proficiency (e.g. Grami, 2010;Kristiansen, Burner, & Johnson, 2019;Lee, 2013;Maarof et al., 2011;Mizuki, 2003;Yahyazadeh Jelodar & Farvardin, 2019;Yeh, 2015;Zarei & Feizollahi, 2018).
Concerning the third research question, based on the interview results, majority of teachers preferred unstructured collaborative pre-writing task. The findings revealed that teachers had welcoming attitudes towards collaborative task implementation in higher educational contexts. Besides, teachers implemented collaborative tasks at three stages of writing. Findings support the outcomes of the researches that reflected the positive attitude of the teachers towards collaborative task in educational contexts (e.g. Alias, Hussin, Mohamed Adnan, Othman, & Hussin, 2018;Ghaith, 2018;Prieto Saborit, Fernandez-Rio, Cecchini, Méndez-Giménez, & Méndez-Alonso, 2016). Besides, finding of this part are in line with similar studies that focused on the collaborative task implementation over educational contexts (e.g. Gillies & Khan, 2009;Hämäläinen & Vähäsantanen, 2011;Van Leeuwen, Janssen, Erkens, & Brekelmans, 2013).
The study has an instant implication for instructors who try to find alternative techniques of writing instruction to EFL students. Both methods used in the current investigation can support students in generating ideas and organizing their writings effectively. Furthermore, the USC pre-writing activities looked more effective due to its interactive nature as a result of interaction between instructors and students. The study proposes alternate techniques in EFL writing courses. Collaborative pre-writing activities might lead to deeper visions into the themes that learners want to write. It provides more opportunity for practice (Ortega, 2007), students are provided with more effective linguistic feedback and valuable input from peers (Vygotsky, 1978). In this case, they are stimulated to produce higher output. The current investigation might encourage language teachers to do investigations regarding the effectiveness of USC pre-writing activities at several contexts. Students can benefit from collaborative pre-writing tasks where anxiety is reduced, self-confidence is improved and the class focus is not absorbed on a separate student but on entire cluster (Slavin & Karweit, 1981). Also, these collaborative pre-writing tasks can help instructors integrate these techniques into their classroom routines in order to reduce the burden of instructors in writing classes.

Instructors' Interview questions
8. Do you think that CL makes the teaching-learning experience more dynamic and enjoyable? Why? Or Why not? 9. Can you think of any disadvantages of CL? 10. How do you prepare students for working in groups? 11. What is your idea regarding structured versus unstructured collaborative prewriting tasks?

Structured Collaborative Pre-writing Task
Unit 10 "The secrets of advertising". Writing Topic: You will write an essay in which you clarify how advertising has changed in your country over the last decade. In your essay, you will … … describe the change, … clarify reasons for this change. … explain consequences of this change.
Part 1: Generating & Evaluating Ideas a) Using the table below and take notes on how advertising has changed.
List some consequences of this change, including details of these consequences.

Description of change
Reasons for change Consequences of change feedback from group b) Tell your group about the change in advertising in your country. Explain the most main causes for this alteration and the most important consequences of this change. c) As you listen to members of your group, assess and explain whether they have selected good reasons and examples. As you receive feedback, record it in the table above.
Part 2: Selecting & Organizing Ideas a) Based on your group's feedback, choose what information you will include in your paragraph. What information will you mention in which order? Make an outline, and then share it with your group.
• • • b) As you listen to your group's writing plans, give them feedback about whether their outline is well prepared.
A sample of student's writing task in the unstructured collaborative group How can we develop educational system in this university?
Nowadays all the students are doing their best, they are strugling to be accepted in the best university. But when they have accepted they feel like this hard time is over and don't need to try anymore. For example, we have an entrance exam for universities in Iran which is one of the hardest process in each persons' lifes. Undergraduates today look to be spending less time on their courses. Thus, I think the educational system in university needs to be changed. The process and achieving educational evidence should be harder than just entering to one. In this regard, some factors such as empowering students' skills, educational facilities and learning environment contributes to educational development.
Recently, in our context some companies complain that many graduate students they employ lack basic skills such as writing and speaking in English. Additionally, in university, curriculum ignores these important goals for the undergraduate education and students can't write or speak properly in English. University can provide more interactive programs to help students improve their skills.
Another thing is that university student need more equipment and related facilities. For example, proper computers in laboratory setting seems crucial. Besides, all classes should be equipped with video projectors and laptops. Additionally, educational workshops are the other effective factor in university setting.
Another perspective is that if the classes be more exciting and enlivening, students would rather to attend the classes and they won't ever skip or miss any session. This is exactly influenced by teachers' teaching style and her/his class management. For example, they can do some group works or use different ways of teaching. Even teachers can apply stress-free practices.
All in all, the way our mother, grandparents and ancestors leaved should not be continued without minimum change. Most students come to university to increase their educational attainment and get better job. Thus, this system should prepare them for work through providing better educational system and as an English language A sample of student's writing task in the structured collaborative group Today heads of university efforts to improve the quality of teaching. Some of them tries to change their existing programs. I think that problems in educational system affects higher education thus the content of lessons, pleasant setting and facilities are important factor about this issue. Firstly the content of courses, facts and principle that are taught in specific courses should be changed. and university must update the curriculum based on students needs. It is suggest that heads of university considers skill development.
Secondly the environment is important too. University should consider pleasant setting for learning. A place for after class discusion with teachers. Such as talk show and after class meeting.
Thirdly facilities such as library and lab can improve educational system in university. It can help in teaching and learning. Students can use books do their projects and do their homeworks. Moreover university should prepare students a computer site for search. Students can use the computer site to take part in on line courses.
Finally recently university students spends more time on campus and interacts with their friend in university. I think that setting, facilities, content of lessons are importants for a university. Changes in learning and teaching are importants in educational system. *Errors and mistakes have been detected by the researchers.