The Effect of Collaborative Writing on EFL Students’ Grammatical Accuracy

Writing is considered as an activity which is done individually and the role of thumb for its feedback is thought to be provided by teachers and instructors. Although there has been a growing body of research on using pair and small group activities in second or foreign language learning in relation to oral skill (e.g. Speaking), on writing there are only a few well-documented researches. This research describes a study exploring the effect of collaborative writing on EFL student’s grammatical accuracy in their writing. A total of 50 Iranian Advanced students of English all male and with the age range of 20 to 24 participated in this study. The subjects were engaged in tasks in which they worked on each other’s writing and gave feedback on grammatical points to each other. Obtaining corrective feedback from their fellows, enabled students to pinpoint their grammatical errors better and subsequently improve their grammatical accuracy in their upcoming writings. The results suggest that collaborative writing (CW) is beneficial in allowing EFL learners to make gain in grammatical accuracy.


INTRODUCTION
While much of the early research in the second language context examined the nature of collaborative talk and/or the role of the first language in second language learning (Swain & Lapkin, 2000), there has more recently been a focus on writing as a site for language learning, including collaborative writing activities. Mancho´n (2011, p. 46), argued that the ''rationale for the language learning potential of writing derives from various influential theoretical strands of SLA research'' (Williams, 2012). In relation to collaborative writing, of importance 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. From a theoretical perspective, the use of pair and group work in the L2 classroom is supported by the social constructivist perspective of learning. The social constructivist perspective of learning, originally based on the work of Vygotsky (1978), posits that human development is inherently attracting by social activity. In First Language (L1) contexts, the child's cognitive and linguistic development arises in social interaction with more able members of society (experts), who provide the novice with the appropriate level of assistance (Gillian  and Neomy ). In particular, these researchers have shown that DICTOGLASS tasks (tasks in which students reconstruct in pairs or groups a text read by the teacher as closely as possible to the original text) were successfully accomplished by learners as a collaborative or joint activity, and hat such jointly performed tasks enabled learners to solve linguistic problems that lied beyond their individual abilities. Further, these researchers have found on delayed posttests that there was a strong tendency for students to stick with the knowledge that they had constructed collaboratively, right or wrong (e.g., Swain, 1998; Swain & Lapkin,1998 as cited in Gillian and Neomy, 2012).

R E T R A C T E D R E T R A C T E D R E T R A C T E D R E T R A C T E D
On the pedagogical side, several researchers have emphasized the multiple benefits of collaborative pair and group work in L2 learning. For instance, McDonough (2004, p.208), citing evidence from pedagogically-oriented research, states that: Pair and small group activities provide learners with more time to speak the target language than teacher-fronted activities, promote learner autonomy and self-directed learning, and give instructors opportunities to work with individual learners. In addition, learners may feel less anxious and more confident when interacting with peers during pair or small group activities than during whole-class discussions.
As such, due to both theoretical and pedagogical considerations, it has been concluded that learners should be encouraged to participate in activities that foster collaboration in the L2 classroom. Now in the present study, we are heading for investigating the effect of collaborative activities in writing to EFL students to draw the finding in the case of grammatical accuracy. To put in another word, this research is heading to see the impact of CW on grammatical accuracy.
The interest in such research stems from both theoretical and pedagogical considerations.

2. Review of Literature
Compared to research that examined the benefits of collaborative work for the spoken discourse, research investigating the benefits of collaborative work for the written discourse in L2, especially collaborativ writing (CW), is scant (Storch, 2005;Storch & Wigglesworth, 2007). For instance, Storch ( 2005, p.153) states that ''although pair and group work are commonly used in language classrooms, very few studies have investigated the nature of such collaboration when students produce a jointly written text.'' Storch points out that most past studies on collaborative work in the L2 classroom ''have examined learners' attitudes to group/pair work in general, rather than to the activity of collaborative writing'' (p. 155). More important for the purpose of this study, Storch also stresses the novelty represented by the pedagogical strategy of having students composing in pairs (p. 168).
Researches  In a cross-sectional study that concentrated on collaborative dialogues, dictogloss, and text reconstruction tasks, Kuiken and Vedder,2002b investigated the role of group interaction in L2 writing. They tested the hypothesis that ''text quality in L2 is positively affected by collaborative dialogue: when learners are given the opportunity to reconstruct together a text, which has been read to them by the teacher, their joint product will be better than an individual reconstruction'' (p.169). The investigators collected data from 40 intermediate proficiency level learners of Dutch, English, and Italian as a second language. They focused on the syntactic and lexical quality of the text produced and how it is affected by the degree to which learners interact with each other and the kind of metacognitive, linguistic, and interaction strategies they used (Gillian and Neomy, 2012). The investigators found that there was a strong relationship between interaction among writers on meta-linguistic awareness and text quality in L2. That is, learners' reflection on and discussion of language forms, content, and the writing process itself resulted in noticing and, as a consequence, better knowledge of certain grammatical and lexical forms. Kuiken and Vedder argued that these findings show that collaborative language production can prompt learners to deepen their awareness of linguistic rules and trigger cognitive processes that might both generate new linguistic knowledge and consolidate existing knowledge. They also argued that meta-talk can help learners understand the relation between form and meaning, and positively affect acquisition of L2 knowledge. Storch,2005 investigated the process and product of CW and students' views on it. She collected data from 23 adult ESL students completing degree courses at a large Australian university. Students were given a choice to write in pairs or individually. Eighteen students chose to work in pairs and five chose to work individually. The study compared texts produced by pairs with those produced by individual learners. The study also elicited learners' reflections and views on the experience of CW. Storch found that pairs produced shorter but better texts in terms of task fulfillment, grammatical accuracy, and complexity, suggesting that pairs seem to fulfill the task more competently. She also found that most students were positive about the experience, although some did express reservations about CW. Storch concluded that collaboration afforded students the opportunity to pool ideas and provide each other with immediate feedback.
In a similar study, Storch and Wigglesworth (2007) compared the writings of individuals and pairs working on the same writing tasks. They also studied how pairs approached the task of writing and how they interacted as they were completing the activity.
The investigators collected data from 72 postgraduate students at a large Australian university.
The participants' proficiency level in English was advanced. Twenty-four of the participants completed two writing tasks individually, and 48 (24 pairs) completed the two tasks in pairs. Like Storch ( 2005), the investigators found that pairs tended to produce texts with greater accuracy than individual writers. They found that collaboration afforded students the opportunity to engage with and about language, and to work at a higher level of activity than the case where they were working alone. Storch and Wigglesworth concluded that:

International Letters of Social and Humanistic Sciences Vol. 11 49
collaboration afforded the students the opportunity to interact on different aspects of writing. In particular, it encouraged students to collaborate when generating ideas and afforded students the opportunity to give and receive immediate feedback on language, an opportunity missing when students write individually (p.172).
More recently, Storch and Wigglesworth ( 2010a, b) conducted a four-week period study. The study consisted of three sessions. In session 1 (Day 1), 48 advanced proficiency level learners worked in pairs (24 pairs) to compose a text based on a graphic prompt. In session 2 (Day 5), the learners reviewed the feedback they received from their teacher and jointly rewrote their text. In session 3 (Day 28), each of the learners composed a text individually using the same prompt as in session 1. The texts produced by the pairs after the feedback (Day 5) were analyzed for evidence of uptake of the feedback provided by the teacher (reformulations or direct feedback vs. editing symbols or indirect feedback), and texts produced be students individually in session 3 for evidence of retention. The investigators found that ''uptake and retention may be affected by a host of linguistic and affective factors, including the type of errors the learners make in their writing and, more importantly, learners' attitudes, beliefs, and goals'' (Storch & Wigglesworth, 2010a, p. 303).
As it can be seen from this review of literature, there aren't many researches in relation to CW and its effect on grammatical accuracy. So the first aim of the current research is to investigate this important matter.
Another important basis for the current study is the potential of writing for language learning in general, and in foreign language (FL) contexts in particular (see, e.g., Mancho´n 2007, 2010a, b). The ultimate aim of the current research is therefore to extend research from SL to FL contexts and investigate the potential of collaborative writing in these contexts.
The following research question is formulated for the purpose of this study: A) Does collaborative writing have any effect on improving grammatical accuracy on EFL students' writing?

1. Research Hypothesis
A null hypothesis was run in this research. Collaborative writing doesn't have any effect on improving grammatical accuracy on EFL students' writing.

Participants
A total of 50 advanced EFL students of English language learning all male and with age range of 20 to 24 were participating in this study. All of these students were Iranian and English was a foreign language for them. English was taught in an EFL context. They were recruited to participate voluntarily in an English composition program (ECP). The program was provided by the University of Raja in the spring of 2013. The aim of the program was to enable students to write in English and to improve their writing ability for the future course in ILSHS Volume 11 essay writing. The program started in April the 6 and finished in June the 6 of 2013. It took two month period and 24 sessions. Each week there were three sessions each took 1 and a half hour. For the purpose of the research students were divided into two groups. Prior to the starting of classes and in the registration day students were informed about a pretest in the second session.

3. Materials
There was a pretest-posttest design in this research. For this purpose, two pieces of writings were required. In the second session of the program, the pretest was run and in the 24 th session the posttest was held. The topic for the pretest and posttest writings was chosen as "the needs for learning English in today's life". The participants were in advanced level of proficiency so writing about this topic wasn't hard for them. For scoring, an Analytic procedure was used which was once proposed by Brown and Bailey. In this scoring procedure, the students employ a marking scheme. Each rater reads the composition and assigns a grade from the specified range to each grammar point. This scale was also given to the students to do correction and give grade based on it. In the first session the teacher taught students how to use it. For analyzing data, SPSS 16 was used.

4. Procedure
The 50 EFL advanced students were divided into two classes: class A and class B. In class A, collaborative writing was the desired treatment. In each session from the second session on, the teacher gave a subject to the students to write about. In the following session students came back with their writings, but they didn't give them to their teacher to give corrective feedback on them. Instead, they gave their writings to their partners to give CF on them. They consulted the problematic parts with each other and did their best to make their

International Letters of Social and Humanistic Sciences Vol. 11
composition better. In each piece of paper there were two split places for scoring and feedback. One of these places was devoted to students' partner to give their feedback and utterly their score; on the other hand, there was a place for the teacher to assess the feedback which was given by partners to their friend's composition. To give a gist of it, each session students were working on each other's writings then the paper gathered and delivered to the teacher and he assessed the given feedback by students in the following session. 21 sessions had passed in this way. During these sessions, collaborative writing (CW) was quite observable due to the sense of humor and friendship that it caused. Students were working on each other's writing with a great passion. In class B; however, there wasn't any tasks in relation with collaborative writing.
The same procedure like class A was run in relation to pretest, and posttest. In the second, session the pretest was held and in the 24 th session the posttest was run. The teacher was the only assessor of students' writings in the class B. During each session, the students were reading their writings and the teacher gave CF on the spot of reading time. Then the papers were gathered and the teacher gave CF on them for the next session. There were no interactions among students. The atmosphere of the class B was quite boring. Something important to mention is that the first session in the class A was devoted into teach students how to use the comprehensive rating scale.

RESULTS
First, students' score in the both classes were gathered in pretest and posttest. Next, the obtained scores were put on SPSS to analyze. Table 2 is the descriptive statistics for class A and class B. By putting means along side of each other an interesting finding will ravel: it is class A that was able to improve its accuracy mean from 65.08 to 76.08. Improving 11 credits in mean score for class A clearly showed the significant effect of Collaborative Writing on improving grammar accuracy. On the other hand, we have also an increasing improvement in the means score of class B. Class B mean score has improved from 65.80 to 67.16. Fig. 1 and Fig. 2 also show the improvement of means score in both class A and class B. although both classes have improved their mean score in posttest, class A improved it in a sharper way that shows the significant of CW tasks in EFL classroom.

ILSHS Volume 11
The trajectory of class A is as sharp as about 11 that make the fact of improvement obvious in this class.

International Letters of Social and Humanistic Sciences Vol. 11 53
The trajectory of Fig. 2 that devoted to Class B is much less than Class A and is something about 1.36 and better to say 80 percent less than Class A. although by the illustrated statistics, it is clear that the effect of CW on improving grammatical accuracy is significant, we did use test of Between-Subjects Effect to answer the null hypothesis. Table 3 illustrates the obtained statistics. It is quite easy to survey the hypothesis with the statistics of ANOVA Test of Between-Subjects Effect. The p = .001 is showing that the CW was significant so the null hypothesis is rejected. The p = .63 also is certificate for this rejection; it shows that using other techniques aren't significant for improving grammatical accuracy.

CONCLUSIONS
The result of the present study support the hypothesis that Collaborative Writing has a significant effect on improving grammatical accuracy of the EFL students' writings, thus supporting the previous studies (Kuiken and Vedder, 2002b; Storch and Wigglesworth, 2007). It was found that improving in the knowledge of grammatical accuracy tended to be larger with the use of collaborative writing in the classroom. At one encounter, the participants in the CW class demonstrated large improves in grammatical accuracy (80 percent more than the other class); indicating that for advanced-level learners CW can increase the accuracy in grammatical points.
The results indicate, overall, that Collaborative Writing has a greater influence on grammatical accuracy of advanced EFL writers than the other traditional method which was to correct student's writings on the spot of their reading in the classroom. As a whole, the results are consistent with those of Storch and Wigglesworth (2007), who found that the use of CW tasks improved the grammatical accuracy.
It is important to note that the participants of this study were in their advanced level of proficiency and already have known about CW, and Analytic rating scale. Further research examining other proficiency level in accordance with the use of CW would be a useful follow up to this study.
This study sheds lights on the way that Collaborative Writing could improve the grammatical accuracy of advanced EFL writers so it can be an excellent mean to improve writing ability in EFL context. In this way, teachers and instructors may be able to take advantage of CW in their writing classes to improve their students' grammatical accuracy and finally writing ability.