Languaging and Writing Skill : The Effect of Collaborative Writing on EFL Students ’ Writing Performance

The purpose of this study is to explore the effect of using collaborative techniques and activities on EFL students’ writing performance. A total of 35 low-intermediate EFL students ranging from 15 to 18 years-ofage participated in this investigation. These participants were assigned into two groups: An experimental group (N=17) in which writing skill was practiced through a collaborative writing syllabus; and a control group (N=18) in which writing skill exercised individually in the classroom. In this study a pretest/post-test was run, also a paragraph rating scale was used for obtaining students’ overall writing performance and their performance on different components of writing such as content, organization, grammar, vocabulary, and mechanics. The findings of this study revealed that using collaborative techniques and activities had a positive effect (p=.001) on overall writing performance of EFL students, and on writing components such as content (p=.003), organization (p=.001), grammar (p=.001), vocabulary (p=.008), and mechanics (p=.001). The results of this study shed light on the importance of using collaborative techniques and activities in L2 writing classrooms, which bears some implication for teachers and curriculum planners.


Introduction
states that learning is a socially constructed phenomenon in which interaction among peers and members of the society leads one to learn and to construct his or her knowledge.In this process of constructing knowledge, collaborating with "expert" learners helps the "novice" learners to be capable of preceding this process of accruing knowledge (Dobao, 2012).According to some researchers (Storch, 2002;Swain & Lapkin, 1998) when the impact of applying sociocultural theory on L2 learning was investigated, the results always showed the effectiveness of this theory on the L2 learning.
There is always much about collaborative skills in research, yet collaborative writing is paid attention with little thorough considerations.Mancho´n (2011, 46), argues that the ''rationale for the language learning potential of writing derives from various influential theoretical strands of SLA research''.These are Skill Learning Theory (DeKeyser, 2007), Focus on Form (Long, 1996), the Noticing Hypothesis (Schmit, 1990), and the Output Hypothesis (Swain, 1985).In relation to collaborative writing, what becomes important is a theoretical construct which reflects the increasing realization of the relevance of the social context of language learning.Consideration of social context is frequently underpinned by the adoption of a culture of society theoretical approach which provides a means to understand and elucidate the learning process.It is within this construct that collaborative writing has been used to explore how social interaction contributes to learning, feedback, and our understanding of, and insights into, both of these.
The process of collaboration to assist each other in learning L2 is called scaffolding.Based on the researches conducted by some (Alegrı´a de la Colina & Garcı´a Mayo, 2007;Kim & McDonough, 2008;Nassaji & Tian,2010;Storch,2002;Swain, 2010), the process of scaffolding can also occur in an L2 context among peers when they are working on a task in pair.Some research has been conducted to see the effect of different tasks on L2 learning.Ellis (2003) states that lots of research focuses their attention on task complexity on L2 production, yet he asserts that they also put their attention on the influence of task design on focus on form, collaborative dialogue, and feedback.In this regard, Swain (2001) provides the definition of collaborative tasks as a set of tasks in which involve learners "in comprehending, manipulating, producing, or interacting in the target language while their attention is principally focused on the meaning rather than form".One requirement of collaborative tasks is, as Dobao (2012) mentioned, they require "learners [to] work in pairs or groups to 204 produce one jointly written text".Exercising this fact, learners will be pushed to collaborate on their language that they have already used.
Lots of the previously conducted research on collaborative tasks scrutinizes "spoken discourse" more than "written discourse" (Shehadeh, 2011).This scarcity in number of written discourse research is one reason of conducting the current paper.Moreover, the previously conducted research has been done in ESL context and the number of paper explored the effectiveness of collaborative writing in EFL context is so scant.In this regard the context of this investigation is EFL.Some (Storch and Wigglesworth, 2007) conducted their study just by comparing between two sessions of instruction; one in pre-test time and the other in post-test time and just by comparing students' performance in these two sessions, so the possibility of intervening factors were ignored in their study.Consequently, the thrust of this study is to examine the effect of collaborative writing after 12 sessions in an EFL context and with 35 low-intermediate students.

Review of Literature
Sociocultural theory may be considered as the main theoretical framework for collaborative learning (Donato, 1994;Ohta, 2000).Using sociocultural theory as the framework of their research, Storch (2002), and Swain and Lapkin (1998) indicated that applying sociocultural theory in L2 learning had a positive effect on L2 learning due to the fact that for each learner there could be some strengths and weaknesses that when learners worked with each other they could compensate for each other's weaknesses by their strengths (Ohta, 2001).It should be stated that there is a difference between what is called collaborative work and what is called cooperative work.According to Strauss (2007) it is just collaborative work that leads to L2 learning development and not cooperative work.The rationale for this difference is that learners cover each other's deficiencies in building the final knowledge in collaborative work, but it is not the matter of focus in cooperative work.
The existing research on pair and group work in L2 writing can be divided into two sections based on their specificity.First, the study of Connor andAsenavage (1994), andRollinson (2004) showed the effectiveness of pair and group work since they provided feedback that involved in the process of learning.Second, with regard to issues relating to group dynamics, various types of group formations, and how groups function in peer review tasks (Connor and Asons, 2002) indicated collaborative work effectiveness.Swain (2010) declared that "the learners' deliberations about how to solve problems concerning language use" called "languaging".When students are working in a language problem by their own "languaging" they are self directed, for instance when learners conduct silent reading to find out the gist of the passage.However, when learners involve in each other learning and work together, for example when they talk about a social event "languaging" refer to a social construction of meaning.According to Swain (2006) the later kind of languaging is the one which helps learners during the writing process because it contributes to language learners to put their focus on the writing problems together with their pairs and develop a deeper understanding of the language.
In a study on 72 postgraduate students in an Australian university, Storch and Wigglesworth (2007) compared the writings composed by individuals and those composed in pairs.The findings of their study showed that there is a significant difference between the two groups when accuracy does matter.However, the results indicated no significant difference between the two groups regarding complexity and fluency.In another section of their study, they revealed that 30% of the conversations among group members were about language issues that indicated the profitability of collaborative writing activities in contributing second language learners to be in an authentic-like situation to use the already learnt information.
Fernandez Doboa (2012) investigated the performance of Spanish learners who were at intermediate proficiency level and worked either in group, pair, or individually on written tasks that were about past tense grammar.In this study, students who worked in group and pair were asked to answer a joint written task while students who worked individually were asked to produce a piece of writing, each.The written texts were gathered and analyzed in the term of accuracy, fluency, and complexity (syntactic and lexical).The analysis revealed that with regard to accuracy, group-produced-text was ranked the most accurate, followed by the pair, and then individuals.It was also revealed that the group produced text were better in terms of fluency and complexity.The results also indicated that when students put their knowledge together it enabled them to do collaborative writing activities; and to be able to produce more accurate written text as a result of shared knowledge.
Collaborative writing in foreign language classroom was the focus of an investigation by Shehadeh (2011).In his study, Shehadeh used a mixed methods study in which he was not only interested in the effect of collaborative writing on the quality of learners' writing, but he also was interested in the learners' attitudes toward collaborative writings.He run his longitudinal study over sixteen week semester.His study had a pre-test post-test design.Shehadeh established two groups: an experimental group who worked on the writing tasks in pair and a control group that its members worked on the writing tasks individually.This study indicated some unexpected results.While in the previous studies the improvement in accuracy was reported in group work, in this study improvement was achieved in content, organization, and vocabulary, but not in accuracy.The justification of Shehadeh (2011) for the findings of this study was that because the learners who participated in this study were in low-proficiency level they were not be proficient enough to provide each other with the necessary language knowledge such as grammar.
The study conducted by Alegrı´a de la Colina and Garcı´a Mayo (2007) was also examined low-intermediate level learners who participated in a collaborative writing program.The findings of this study revealed that learners who had interaction in accomplishing their tasks (jigsaw, dictoglass, and text reconstruction) frequently arriving at the correct solution and also accurate answer in responding to the problems in the problem solving tasks and activities.In this regard, what Alegrı´a de la Colina and Garcı´a Mayo (2007) conducted showed that even in the low-intermediate level learners could benefit from each other feedback, and knowledge.Williams (2012) stated that the findings of these studies revealed that by using collaborative writing activates, learners could benefit from the mechanism necessary for learning through a verity of language processes.He states that collaboration which involved in these tasks is the important factor of improvement in different writing components including accuracy, fluency, and complexity.
Scrutinizing the relevant review of literature, this study aims at shed light on some of the controversial parts of the previous studies such as the effect of collaborative writing on overall writing performance and also different writing components involving content, organization, language, vocabulary, and mechanics.Involving participants at low-proficiency level, this investigation studies the effect of collaborative writing in this level of proficiency which is the center of debate in the aforementioned literature.Also for expanding the literature of the effect of collaborative writing on EFL learners, this study will be conducted in an EFL context.The following research questions are at the forefront of the present study aims: 1. To what extent does collaborative writing help EFL students to improve their writing skill? 2. What is the effect of collaborative writing on EFL students' writing performance with regard to different writing components: content, organization, grammar, vocabulary, and mechanics?3. Methodology 3.1 Participants A total of 35 low-intermediate Iranian EFL students all male and with age range of 15 to 18 were participating in this study.All of these students were Iranian and English was a foreign language for them.They were studying English language at Sharif Language Center Institute; Tehran, Iran.In this institute, one-skill-program was provided for students in different L2 skills involving speaking and writing.After achieving their low-intermediate level, students were asked to participate in a one skill program in writing skill and the other L2 skills.The aim of this writing program was to familiarize students with the bases in English language writing and to improve their writing ability for the more advanced future courses in writing.The duration of the program was a month and a half.The program involved 12 sessions.Each session took an hour and a half.For the purpose of this research students were divided into two groups.While in the experimental group (N=17) different collaborative tasks were done, in the control group (N=18) participants wrote their essays individually.It should be stated that the instructor for the both groups was the same (one of the researchers of the current study).He taught both groups for four sessions a week; two sessions in the experimental group and two sessions in the control group.The process of providing corrective feedback was also conducted by this instructor for both groups.Another point to be mentioned is that for fulfilling the inter-reliability of assessing students' writings a colleague of the instructor helped him in assessing students' writing.He was trained how to use the rating scale (Appendix 1) for assessing students' writing.

Instruments
There was a pretest-posttest design in this research.For this purpose, two pieces of writing were required.In the first session of the program the pretest was administrated.In this session students were required to write one paragraph on the topic of "Describe your dream home".This topic was the same for both the experimental and the control groups.The allotted time for writing of this 100 (at least) word essay was 30 minutes.In the 12 th session the posttest was run.The same procedure of the pre-test was run in the both groups.This time, students were required to write one paragraph with at least 100 words on the topic of "Where do you like to travel?Why?"In the post-test time, students were required to write their essays in 30 minutes.
There is a widely used paragraph rating scale that first proposed by Jacob, Zinkgraf, Wormuth, Hartfiel, and Hughey (1981) and in (1991) was modified by Hedgcock and Lefkowitz.This scale determines the differences in students' writing in the both groups (Appendix 1).According to Shehadeh (2011) the scale defines the following five component areas on s 0-100 point scale: 1. "Content: knowledge of subject; development of thesis; coverage of topic; relevance of details; substance; quantity of details. 2. Organization: fluency of expression; clarity in the statement of ideas; support; organization of ideas; sequencing and development of ideas.3. Grammar: use of sentence structures and constructions; accuracy and correctness in the use of agreement, number, tense, word order, articles, pronouns, prepositions, negation.4. Vocabulary: range; accuracy of word/idiom choice; mastery of word forms; appropriateness of register; effectiveness in the transmission of meaning. 5. Mechanics of writing: conventions of spelling, punctuation, capitalization, paragraph indentation, etc."There are four bands for each component of the scale: excellent to very good, good to average, fair to poor, and very poor.Finally for analyzing the obtained data a series of t test were run by the help of SPSS 19.

Procedure
The 35 low-intermediate Iranian EFL students were divided into two groups and assigned into two classes: class A (the experimental group) and class B (the control group).One syllabus was created for both classes.This syllabus was the same regarding different elements such as time, duration of each session, the material that be covered, the objective of each session, and the exercises that be done each session.The only difference between the two syllabuses was that in the experimental group students went through writing assignment in collaboration with their pairs, but in the control group students worked on the assigned activities individually.In class A, the principles of collaborative writing were in process of the syllabus.Each session, first the teacher taught new points and discussed the materials.These took 45 minutes.Then the teacher nominated a topic and assigned students into pairs.Next, they were asked to work on the topics with their group members.They were allotted 45 minutes to work on the topic and to write about it.During this period, students started to talk about the topic with their peers.They provided each other with grammatical points, vocabulary load, and punctuation correction.During the time of doing collaboration tasks among students, the teacher went around them and provided them with any necessary points about the writing procedures.After 40 minutes they had to stop writing.In the left time (five minutes) students were required to read their writings.At the end of the class the teacher announced a topic and students were required to write about it at home and bring it next session.At the beginning of next session the teacher first got back students' writing in which he provided them with some feedback and then started the aforementioned process of writing class.
In class B, the control group, the syllabus was followed in the same way.The same principles and rules of writing were taught and the same material were covered and taught by the teacher.For this group the same topics as what were announced for the experimental group were provided.The cut edge between the two classes was that in the class B after the first 45 minutes of the class time the teacher nominated a topic and students were required to write about it not in pairs but individually.The process was so similar to the experimental group, in which students wrote about the topic but it was done individually.This means that they did not take advantage of their classmate's grammatical feedback, vocabulary load, and punctuation correction.They depended on their knowledge and what the teacher taught them during the class.While students wrote about the topic in the class B individually, the instructor went around them and provided them with any necessary points regarding writing affairs.Then, in the same way like class A after 40 minutes students stopped writing and they read their essays.For the following writing session the instructor nominated a topic for the students to write about it.The students write about the topic at home and next session they brought their writings to deliver them to the teacher.Each session at the beginning of the session the teacher got back students' previous writings in which he provided them with feedback regarding different components of writing.Some other aspects of the procedures run in this program are worth mentioning.At the beginning of the class and after the instructor got back the students' writings with the provided feedback on them, he started to give his instruction according to the syllabus.He used power point slides to explain new principles, rules, and materials about writing.After the instruction was done, students would be provided with the printed version of the power point slides.They were asked to review the printed power point slides and if they had any question about any parts of them raised their question in the classroom.In this program some chapters of the low-intermediate book of North Star (2), and Reading & Writing 3rd edition were covered.It should be stated that the role of the books were not as powerful as the role of the teacher provided materials.By the teacher provided materials it is meant that the instructor of this program himself compiled a pamphlet based on the syllabus of the class and went through it during the 12 sessions.In the 12 th session the post-test was administrated.It was a composition writing test in which students were required to write a paragraph with at least 100 words.The topic for this post-test was nominated as "Where do you like to travel?Why?" Students in the both classes were allotted 30 minutes to compose their writings.The pre-test and post-test papers among both groups were collected and prepared for the scoring phase.In this phase of the study the writing scale was used and different writing components of the students' writing were assessed.For the matter of inter-ratter reliability, the essays were rated by two assessors according to the rating scale.In this section, the instructor's colleague helped him to assess students' writings.He gave a clear description of the scale for his colleague and they exercised on the scale by assessing two essays.When the instructor was sure that his colleague was well aware of the rating scale procedure, the assessing of students' writings were started.

Results
With regard to administrating the pre-test, participants in the both groups (experimental group and control group) were required to write about a topic on the title of "Describe your dream home".In the next stage, their essays were assessed with the help of the rating scale that first proposed by Jacob, Zinkgraf, Wormuth, Hartfiel, and Hughey (1981) by two assessors who were well-familiar with the writing scale.The correlation between the two assessors was .84.As it was expected, there was no significant difference between the two groups' mean score.The results of the Table ( 1) indicate that participants in the both groups were in the same level of proficiency at the beginning of the program.It is not only true with the total obtained mean scores by the participants but also true about the different components.Comparing different components of writing, one can find out that participants are at the same level of proficiency.
After 12 sessions of instruction, students participated in the final test which was considered as post-test.The same procedure was run in the post-test time.Students were required to write a-one hundred-word paragraph about a topic with the title of "Where do you like to travel?Why?" Like their pre-test, they were allotted 30 minutes to write this paragraph on the specified topic.Again, the two instructors were in charge of assessing the essays with the help of the described writing scale.Once again and due to the purpose of being sure about the inter-reliability of the scores another correlation between the scores gave by the both teachers was run.The correlation between the scores was .88.It can be inferred that due to the practice instructors exercised in the pre-test time, they would be more expert in assessing students' writing by using the rating scale in the post-test.Table (2) shows the results of total and the components descriptive statistics.An independent-sample t test comparing the total mean scores of the experimental and control group found a significant difference between the means of two groups (t (28) =.168, p < .05).Table (3) shows the results of independent-sample t test.
Table 3. Independent-sample t test It shows that the means are significantly different in the two groups.The mean of the experimental group was significantly higher (M = 68.26,SD = 5.06) than the mean of the control group (M = 60.53,SD =3.97).An independent-sample t test was conducted with regard to different writing components including content, organization, grammar, vocabulary, and mechanics.The aim of Table ( 4) is to clear the fact of whether the mean score differences among writing components are significant or not.The indexes in the Table ( 4) indicate that the performance of the two groups with regard to different writing components including content, organization, grammar, vocabulary, and mechanics are significantly different.

Discussion and Conclusion
The first research question of this investigation was "To what extent does collaborative writing help EFL students to improve their writing skill?"For answering this question; a set of descriptive statistics was run.Table (2) indicated that the overall writing performance of the students who were in the experimental group improved more than those students who were in the control group.The total mean scores of the students in the both groups were compared in the pre-test and post-test.The total mean score of the experimental group was 57.13 and this amount reached to 68.26 after the treatment sessions.On the other hand, the control group total mean score was 56.66 in the pre-test that reached ultimately to 60.53 in the post-test.These results showed some factual points with regard to the essence of this investigation.First of all, the total mean score of the both groups in the pre-test showed that the participants of this study were in the low-intermediate level of proficiency, in other word, it indicated the homogeneity of participants of this study.In this regard the internal validity of this investigation is proved to be accomplished and exercised effectively through the effective sampling of selecting participants.At the end of the instruction, the post-test revealed that participants of the experimental group improved their proficiency level in writing from low-intermediate to intermediate.This is indicating that by providing low-intermediate students with instruction in writing skill if the provided instruction goes through collaborative techniques it can be claimed that in the end of instruction shifting in the proficiency level is feasible.
To investigate this claim an independent-sample t test was run.The result showed that the mean scores in the both groups are significantly different (t (28) =.168, p < .05).It indicated that the treatment was effective in this program.
The first reason may be the effect of what is called social constructivism.According to Kuiken & Vedder (2002b) collaborative writing tasks contribute learners to construct meaning in their writing jointly.These tasks cause "meaningful and purposeful communication" and this can engage learners in a cognitive process in which L2 learning will happen.So, students collaborate their knowledge in writing, this causes a meaningful communication.This communication is made by the needs of students.While writing about a topic, students have different knowledge in different area.Some are good at grammar, other have good diction, so they have some needs that have to be fulfilled with their peers.When the needs are fulfilled their cognitive process start to react toward the new obtained materials and finally learning happens (Storch, 2005;Storch & Wigglesowrth, 2007, 2010 a, b).
With regard to the second research question which was "What is the effect of collaborative writing with regard to different writing components: content, organization, grammar, vocabulary, and mechanics?"when administrating pretest and post-test; each of these components' score obtained with the help of writing scale in the Appendix (1).The results showed that regarding content students in the experimental group outperformed those in the control group.The mean score of the experimental group was 18.06 and for the control group it was14.93.An independent-sample t test revealed that the mean score difference between the content component in the experimental and the control group was significant (t (28) = .383,p < .05).In the next stage, organization as one of writing components was computed.While the mean score was 17 for the experimental group, it was 13.46 for the control group.An independent sample t test indicated that the mean scores were different between the experimental and control group (t (28) = .68,p < .05).The next component was grammar, in this component the experimental group mean score was 17.67, the control group mean score was 15.86.An independent-sample t test was conducted to compare whether the mean score difference between the two groups was significant or not.It was revealed the mean score between the two groups was significantly different (t (28) = 2.23, p < .05).Vocabulary was another component of writing in which there was a difference in writing performance in the both groups.While the mean score was 12.20 for experimental group, the mean score of control group was 11.66.The independent sample t test showed the significant difference between the mean scores of the both groups (t (28) = 4.22, p < .05).The last writing component that an independent-sample t test was conducted to find out whether the difference was significant between the mean scores or not was mechanics.The results showed that the difference between the two groups was significant (t (28) = .06,p < .05).
The obtained data indicates that with regard to different writing components the experimental group in which writing tasks were performed through collaborative writing outperformed the control group in which writing activates performed individually.The first reason for that can be mentioned as using collaborative activities in a group may help a group to use the potential strength of all its members.While one of the members may be good at organizing the ideas, the other one may have good knowledge of vocabulary.In this regard, students learn from each other while they are working on different tasks.Mancho´n (2011) made a well-distinguished difference "learn to write" and "writing to learn".In this way, the former is to instruct your learners then require them to use those instructions to write about a topic.It is a product-oriented learning.If we consider instructing writing in this way, students will attend at writing classes in which the teacher provides the necessary instruction and then they are required to write individually.The latter, however, is writing in process with the use of instruction and other things.It is a process-oriented learning, in which learners learn something in the process of instruction not just by the help of the instruction.This fact is to great extent in the same way with collaborative writing.In collaborative writing classes students learn from their peers.According to Hirvela (2011) using this policy in writing, one can promote knowledge of language, and also knowledge of content.
The second reason for promotion of writing skill in collaborative writing environment is that students working in collaborative groups can take advantage of group members for built-in peer review as they complete writing projects.
Each student may accomplish one section of assessing in regard with his/her area of strength.For instance, in a group one of the members has good grammar knowledge so he can help to assess the grammar component of the group and then other members may be able to use those instructions as up taking it as new knowledge.Yhen in the later writing situation they may use that new knowledge.
One more reason for the effectiveness of the collaborative writing is that it encourages students to consider their audiences.When there is a group and students pose their opinions in that group and receive feedback on those opinions, they come to the conclusion that how is others reaction toward their opinion.It helps them to consider it and try to change their content, their organizing manner and other writing components such as vocabulary.During this time they put their focus and their attention in different ways and from the feedback they receive they establish their new knowledge.
As a whole, the foremost theory behind the effectiveness of collaborative writing may be co-constructivism.When two learners, one "novice" in one writing aspect and the other "expert" in that area, joint to each other to write about a topic the knowledge that transmute between them helps to construct new knowledge of writing.In this regard, they will take advantage of their peers and creating their "uptake" knowledge, later on, by using the "uptake" knowledge they make their permanent knowledge, and finally learning does happen.
The results of the present study support the hypothesis that Collaborative Writing has a significant positive effect on improving writing skill of the EFL students, thus supporting the previous studies (Kuiken and Vedder,2002b;Shahedeh, 2011;Storch and Wigglesworth,2007).This study found that improving in the knowledge of writing tended to be larger with the use of collaborative writing in the classroom.Students in collaborative writing classrooms have more interaction with each other and in a co-construct way they construct new knowledge.In line to this the results of this study confirm what has been done by Ohta, (2000); Storch, (2002) in which they came to conclusion that by pairing different learners with different proficiency levels together caused the transferring of different kind of the knowledge among pairs and finally learners made their own knowledge from the obtained knowledge.It is worth mentioning that different writing components including component, organization, grammar, vocabulary, and mechanics tend to improve in the experimental group more than the control group.In this regard the results of this study is compatible with that of Dobao (2012), Wigglesworth and Storch (2012), but the results of the current study has conflict with that of Shahedeh (2011) in which he asserts that, as one of writing component, grammar does not improve in the collaborative writing classes, yet the results of this study indicates that grammar along with other writing components does improve in the collaborative writing.One rational reason is that some of the learners have got better grammar knowledge than others so in the process of collaboration they can share it with those who are weak in that component of the language.Even though they are in the low-intermediate level of proficiency, they have already known about some rules of English language that can share with each other.
The pedagogical perspective of this study make it clear that by using collaborative task in writing there may be established a positive condition among the students.This helps students to learn from each other in an unobtrusive way.One of the theoretical implications of this study is to make this fact clear that by using collaborative writing tasks in the writing classroom, not only the grammar, vocabulary, and mechanics improve but also the content and organization will improve.The last implication of this study might be using collaborative writing tasks helps students to improve their writing skill that shows if L2 instructors use collaborative task with regard to other skills such as reading, writing, and listening they may see improvement in these skills too.

Table ( 1
) illustrates the total and the componential mean scores between the two groups.

Table 2 .
Descriptive statistics of the total and the componential scores in the post-test Max.As can be seen in Table(2), the total mean score for the experimental group is 68.26, and it is 60.53 for the control group.Both groups improved their total mean score in the post-test but the experimental group improved it more than the control group.While the amount of increased total mean is 11.13 for the experimental group, it is 3.87 for the control group.There is 7.27 total mean score difference between the both groups.With regard to different component mean score, one can see that the amount of increased in mean score of components of the experimental group is more than the amount of increased mean score in components for the control group.In this way content improved from 14.46 to 18.06, organization from 13.73 to 17, grammar from 14.93 to 17.62, vocabulary from 10.40 to 12.20, and finally mechanics from 3.53 to 5.93 in the experimental group.On the other hand, content in the control group improved from 13.93 to 14.93, organization from 12.53 to 13.46, grammar from 15 to 15.86, vocabulary from 11.26 to 11.66, and mechanics from 3.80 to 4.70.

Table . 4
Independent-sample t test of different writing components