CALL , Prewriting Strategies , and EFL Writing Quantity

This study sought to explore the effect of teaching prewriting strategies through different methods of input delivery (i.e. conventional, web-based, and hybrid) on EFL learners’ writing quantity. In its quasi-experimental study, the researchersrecruited 98 available sophomores, and assigned them to three experimental groups (conventional, web-based, and hybrid groups) and a control group. The control group received product-oriented writing instruction regarding paragraph writing, whereas the experimental groups were taught, though under different learning conditions, to use prewriting strategies pertinent to generating ideas through freewriting, brainstorming, listing, clustering, and asking wh-questions during the planning stage of writing. The learners’ writing quantity was measured, both prior to and after the implementation of the experiment, via counting the number of idea units in the writing tasks which were administered to the learners as the pretest and posttest. One-Way between-groups ANOVA was utilized to analyze and compare the collected dataand to discern whether and to what extent the participants in each group could gain benefit from their specific type of instruction. The results of this study revealed that writing quantity was indeed under the influence of different treatments, with the learners in the hybrid group outperforming their counterparts in all the other groups.


Introduction
Writing, based on Hacker, Keener, andKircher, (2009, p. 154) is "the production of thought for oneself or others under the direction of one's goal-oriented metacognitive monitoring and control, and the translation of that thought into an external symbolic representation".It is a preponderant productive skill that can be used in learning other receptive and productive skills (Zhu, 2004).Writing encourages thinking and learning, motivates communication, and makes thought available for reflection (Mekheimer, 2005).When thought is written down, ideas can be examined, reconsidered, superseded, rearranged, and changed.The paramount importance of this indispensable skill is stated by Olshtain (2001, p. 207) who maintains that "…the skill of writing enjoys special status-it is via writing that a person can communicate a variety of messages to close or distant known or unknown readers".
Notwithstanding this widely-acknowledged importance, students of writing, and their teachers, too, experience a variety of unfavorable emotions, ranging from fear, to trepidation to elation which potentially hinder the development of their writing skills (Byrad, 2010).Such is the case probably because writing, be it in one's first or second language, is one of the most demanding skills that our students have to learn (Rivers, 1981).The onerous nature of writing has recurrently been acknowledged by experts in the field.For instance, Nunan (1989) maintains "it has been argued that learning to write frequently and expressively is the most difficult of the motor skills for all language users regardless of whether the language in question is a first, second or foreign language" (p.35).Elsewhere, the same author contends that "…producing a coherent, fluent, extended piece of writing is probably the most difficult thing there is to do in a language" (Nunan, 1996, p. 271).Richards and Renandya (2002) take a more unwavering stance in this regard and hold that " [t]here is no doubt that writing is the most difficult skill for L2 learners to master" (p.271).
Similarly, Dujsik (2008) remarks that second language writing is an overwhelmingly complex process since it poses such challenges to learners as generating ideas on the content knowledge relevant to a writing topic, having concerns over diction and grammar, organizing sentences into paragraphs and paragraphs into a larger piece of discourse with appropriate organizational patterns, considering the writing purpose, audience, and so forth.Each of these challenges calls for a certain type of consideration and treatment.However, as Shafiee (2010) holds, in the writing process, the first step is always the hardest; a large number of ESL students have been observed to be afflicted by an utter state of bewilderment while trying to embark on a writing task.They simply do not know what to write and how to write it.This problem usually leads to a failure at developing subsequent stages in order to complete a writing task and, as such, requires special attention.
In academic settings, preparing EFL students to meet academic requirements which entail writing is a laborious undertaking, the success or failure of which hinges upon special methodology and resources (Ellis, 2008).One methodological approach which has gained widespread popularity over the past decades is strategies-based instruction.Also, among the plethora of instructional resources available these days, computer-assisted language learning (CALL) is one which is increasingly winning the support of language teachers and learners (Ebrahimi, Tabatabaee, & Shafiee, 2012).The present study aimed at implementing a prewriting strategy (which refer, based on Dujsik, to conscious thoughts, actions, or behaviors used by writers in the planning phase of writing) under different conditions-conventional classroom, web-based environment, and hybrid context-to determine the effect of the interventions under these different conditions on Iranian EFL learners' writing quantity.To put it differently, the performances of EFL learners in four different groups were compared in this research study: a control group, which received a product-oriented writing instruction, an experimental group which underwent a process-oriented prewriting strategies instruction in a conventional classroom setting, another experimental group which was exposed to the process-oriented instruction of prewriting strategies in a web-based environment, and finally another experimental group which was taught prewriting strategies in a process-oriented hybrid (i.e.combining conventional and web-based) context.
Moreover, an incentive which led to the implementation of the present study was the fact that strategies-based L2 instruction had been shown to be conducive to success in the four skill areas of language learning.However, the source of difficulty for most (if not all) L2 learners is either obliviousness or misapplication of language learning/language use strategies.A meticulously measured training of language learning strategies is thus apt to lead to the amelioration of L2 learners' performance in the area under investigation.Moreover, a close scrutiny of research on writing strategies reveals that much of L2 writing-strategy research is descriptive and most findings remain inconclusive (Cohen, 2006), calling for a need to conduct further experimental investigations in this area.In addition, few L2 writing studies have dealt with prewriting strategies and their impact on EFL learners' written productions.More importantly, very few L2 researchers have taken the advantage of computer technology, a potentially valuable supporting tool for language acquisition, to deliver strategy instruction in L2 writing-strategy research.That is why the present study attempted to investigate the differential effects of conventional, web-based, and hybrid instruction of prewriting strategies on EFL learners' writing quantity.Therefore, this study is guided by the following research questions: Does teaching prewriting strategies have any impact on the quantity of Iranian EFL learners' written productions?If so, which mode of instruction (i.e.conventional, web-based, and hybrid) could exert such an effect?

Participants
To choose the participants for this study, English majors studying at Islamic Azad University, ShahrekordBranch, and Safahan Institute of Higher Education who were taking the Advanced Writing Course as part of their Bachelor's degree were asked to take part in the study.Altogether, they formed an aggregate of 107 male and female Iranian EFL learners, with an age range of 19 to 36.After explaining the aim of the study to the students, the researchers randomly chose one of the classes at Shahrekord University to be the control group (CtrlG), while the other class served as the hybrid group (HG).The writing class at Safahan University was the conventional group (ConG).From these three classes 26 students (15 students from Shahrekord University and 11 students from Safahan University) voluntarily agreed not to attend regular classes and to be placed in the web-based group (WBG).The weblog, along with its purpose as the course component, was separately introduced to the web-based and hybrid groups.A homogenizing writing pretest was given to these 107 students in order to exclude from the sample those whose writing abilities were far below or above the average students.Those who obtained a score between two standard deviations below and above the mean were chosen as the participants of this study.Not surprisingly, nearly all the students were in this range since most of them had not received any formal writing instruction before.

Materials
Prior to developing the four different instructional packages, the researchers observed several academic writing courses at different universities, examined textbooks for these courses, and discussed ideas related to the content of the packages with academic writing instructors and their colleagues.After the packages were developed, the researchers and two other colleagues reviewed them.They provided ample feedback for the revision of the instructional materials which were used in this study.

Materials for the Control Group
As for the control group, the book Paragraph Development by Arnaudet and Barret (1990) was used.This book was deliberately chosen for the control group since no prewriting strategy whatsoever is presented in this textbook.The class sessions and the instructional materials followed the order of book chapters, which discuss paragraph organization and give the students some practice on linguistic structures.During the five sessions of the study, up to 70 pages of the book (including such points as indentation, topic sentence, supporting details and their different types, etc.) were covered.

Materials for the Experimental Groups
The instructional package that was prepared for the experimental group was taken from the following textbooks: 1) Writing in Paragraphs by Zemach and Islam (2011): chapter 3, p. 22 on listing; chapter 4, p. 28 on word map (that is the same as clustering); and chapter 5, pp.36-37 on freewriting.
The selection of the instructional materials was the same for the conventional, web-based, and hybrid groups.The mode of input delivery, however, differed.That is, prewriting strategies were presented and practiced based on Grenfell and Harris's (1999) model of strategy instruction in class for the conventional group, online for the web-based group, and finally both in class and online for the hybrid group.

Procedure
After the learners were assigned to different groups, they were required to take a homogenizing writing test (which also served as the pretest of the study).This test consisted of writing an argumentative paragraph with the following prompt: "What are the most important characteristics in a manager?Why?"An open-ended question was also given to the participants in the experimental groups to find out whether any of the students was familiar with prewriting strategies or not.Those who were cognizant of the intended strategies were excluded from the research.While the control group underwent a product-oriented writing class where the students were required to write on different topics and provided with feedback on the content, structure, and organization of what they wrote, the conventional group was exposed to prewriting strategies instruction in class and the web-based group to online instruction of prewriting strategies, the hybrid group experienced prewriting strategies instruction in a hybrid environment where they were taught the lesson in class and had to do and discuss the follow-up activities online.These three experimental groups experienced a process-oriented strategies-based instruction based on Grenfell and Harris's (1999) model of strategic instruction.That is, the following steps were respectively taken in order to teach them prewriting strategies: awareness raising, modeling, general practice, action planning, focused practice, and evaluation.The intervention phase of the study was completed in five sessions, excluding the pretest and posttest sessions, since each of the prewriting strategies of freewriting, brainstorming, listing, clustering, and wh-questions were taught and practiced in one session.The instructional materials for both web-based and hybrid group were available at http://sajadshafiee.blogfa.com.Finally, a posttest (including another argumentative writing, as was the case with the pretest) was given to the learners in the four groups.The performances of the learners on the pretest, and also on the posttest, were compared across the four groups using a one-way ANOVA to help reach the aims of the study.

Results
The quantity of the writings was operationalized as the number of idea units used in each piece of writing.The idea units per writing were counted by the researchers, and then compared via a one-way ANOVA test.What follow is the result of the analysis for the pretest and posttest:

Pretest Results
Tables 1 and 2 illustrate the results of one-way ANOVA employed for the sake of comparing the performances of the four groups prior to instruction.The number of participants in each group, their mean scores, standard deviations, etc. is shown in Table 1.The mean scores for the CtrlG, ConG, WBG, and HG were 11. 29, 12.11, 12.92, and 12.86, respectively.This indicates that there was not a great difference among the performances of the participants in these groups.However, to establish the lack of such a difference, the one-way analysis of variance (ANOVA) table must be consulted.There was not a statistically significant difference at the p < .05level in the homogenizing writing pretest scores for the four groups: F (3.94) = 1.67, p = .17.That means the four groups were at roughly the same level of writing ability before the commencement of the experiment.The researchers could, then, safely implement the intervention and do the after-the-experiment comparisons subsequently.

Posttest Results
After the exposure of the control group to the placebo and the implementation of the relevant interventions for the experimental groups, one-way ANOVA was conducted anew to compare the performances of the four groups on the writing posttest.Tables 3, 4 and 5 depict the results of this analysis: The mean writing quantity scores for the CtrlG,ConG,WBG,and HG were 16.54,19.62,19.68,and 23.03 respectively.To figure out whether the differences between these mean scores were statistically meaningful or not, one needed to glance up the one-way ANOVA table below: The obtained results indicated a statistically significant difference at the p < .05level: F (3.94) = 14.426, p = .00.This means that (different modes of) teaching pre-writing strategies did have a sizeable effect on the students' writing quantity.The actual difference in the mean scores between the groups was also found to be large.The effect size, calculated using eta squared, was .31.Post-hoc comparisons using the Scheffe test (See Table 5) pinpointed the exact locations of the difference between the mean scores.Table 5 indicated that there was a difference between the control group on the one hand and the other three experimental groups on the other.Hence, teaching pre-writing strategies appeared to be a determining factor contributing to the writing quantity of the students.Moreover, among the three experimental groups, the difference between the conventional group and web-based group failed to reach statistical significance.On the other hand, the mean score for the hybrid group was significantly larger than the mean scores of both conventional and web-based groups.This would imply that all different modes of teaching pre-writing strategies in this study were conducive to writing quantity, and among these three methods, the hybrid way of input delivery overrode the others.

Discussion and Conclusion
Previous research has paid scant attention to the investigation of the effects of prewriting strategies on EFL students' writing quantity.Little comparative research has also examined the promising prospects of hybrid instruction in this regard.This study was, hence, a right attempt to uncover the possible effect of teaching prewriting strategies under conventional, web-based, and hybrid conditions on the quantity of written productions by Iranian EFL Learners.
What can be construed from the results of this study is that the teaching of the intended prewriting strategies seemed to have a positive effect on the writing quantity of the participants in the experimental groups.While there was no marked difference between the groups at the outset of the study, the experimental groups performed better than the control group on the posttest, indicating that teaching prewriting strategies was indeed effective so far as the students' writing quantity was concerned.This finding seems to endorse the findings of the studies which give credit to the usefulness of writing strategies (Lally, 2000;Schrader, 2000), brainstorming (Matsumoto, 1995), and idea generation (Cava, 1999;Matsumoto, 1995).
On the importance of prewriting strategies, suffice it to mention that Matsumoto (1995) contends that skilled writers spend extensive time planning before beginning to write, and that they are willing to change their original plans as new ideas emerge.Cava (1999) also reports that previous studies have concluded that prewriting planning was a predictor of writing success.That is, more planning resulted in a better final product.On the face of it, Cava believes that it is not always the case that increased planning time results in a better piece of writing.She reports that in her experiment with four unsuccessful learners, the learners were given ample time; they were also aware of metacognitive strategies.Notwithstanding, they did very little prewriting planning since their barrier turned out to be the lack of effectiveness of application.Purpura (1999) also holds that imperative is how to use strategies effectively.In the present study, however, the students in the experimental groups outperformed those in the control group probably because the former were not merely made versed in metacognitive prewriting strategies.What they were subjected to was indeed the teacher's modeling of how to apply prewriting strategies.They were hence capable of using the targeted strategies and improving their writing ability.
More specifically, teaching prewriting strategies had a significant impact on the participants' writing quantity as measured by idea units.All the experimental groups of the study surpassed the control group as far as writing quantity was concerned.Among the experimental groups, there was a substantially significant difference between the performance of the hybrid group and those of web-based and conventional groups.The difference between web-based and conventional groups, however, was infinitesimal.Not surprisingly, the results of the current study comply with much of previous research that reported a positive impact of prewriting strategies instruction on writing quantity (e.g., Sullivan, 2006).
In conclusion, As Chamot (2008) holds, writing in a new language is arguably the most demanding of the modalities in which one can achieve communicative competence.Novice writers struggle with finding the words they need and remembering grammatical structures, whereas more advanced students find it difficult to link their ideas with coherence and to produce appropriate target language discourse.Given these difficulties, instruction in writing strategies could be beneficial for L2 learners.Writing strategies could pertain to the phase of planning (prewriting), drafting, and/or revising.Prewriting strategies, despite their potential value and fruitfulness, have not received due attention by both researchers and practitioners.While much of the writing strategies research has centered on such aspects as drafting and revising (Cresswell, 2000;Sengupta, 2000), writing processes in L1 and L2 (Oxford, 2000), comparing skilled versus unskilled writers (e.g.Raimes, 1989), studies dealing with strategies of the planning phase of writing have remained scant.
Teachability of language learning strategies in general (Larsen-Freeman, 1991;Weigle, 2005) and prewriting strategies in particular (Ojima, 2006) has been reported by EFL scholars.Such strategies as brainstorming and organizing ideas, among other ones, are conducive to improvements in EFL learners' writing ability, and they can provide a scaffold for EFL writers when they are planning what to write and how to write it.Since the prewriting phase typically occurs prior to the drafting phase, EFL writers, particularly novice writers, may benefit from the reduction of the information-processing burden of mental resources in the process of writing.This in turn makes it possible for them to focus their attention on other demanding requirements such as rhetorical features, text organization, text coherence, etc. (Lv & Chen, 2010;Schrader, 2000;Yuan, 2001).Based on this axiom, the present study posited that teaching prewriting strategies would yield positive effects on the learners' writing ability, which include writing quantity.To be more precise, a comparative enterprise was designed to unravel the effects of conventional, web-based, and hybrid methods of teaching prewriting strategies on writing quantity.It was revealed that the experimental groups' writing quantity indeed boosted in the wake of exposure to the related treatments.

Table 1 .
Descriptive statistics for comparing the writing quantity mean scores of the four groups on the pretest

Table 2 .
One-way between-groups ANOVA results forcomparing the pretest writing scores of the four groups

Table 3 .
Descriptive statistics for comparing the writing quantity mean scores of the four groups

Table 4 .
One-way between-groups ANOVA results for comparing the writing quantity scores of the four groups

Table 5 .
Post-hoc scheffe test results for comparing the writing quantity mean scores of the four groups