Computer-assisted Writing: An Investigation of Metacognitive Awareness of Writing Strategies of Iranian EFL Learners

The purpose of this study was to compare the reported use of writing strategies of Iranian EFL learners in two different venues of technology-assisted and conventional writing classes. Overall, thirty two EFL students participated in this study, eighteen in a technology-assisted and fourteen in a conventional English class who responded to a 64-item Likert scale writing strategy inventory adapted from Wei and Chen (2004) and Griffiths and Jordan (2005). The results of the independent sample t-tests revealed that there was a statistically significant difference between the reported use of language-related writing strategies. The results of the paired-sample t-tests, in addition, demonstrated that just in the technology-assisted environment, there was a statistically significant difference for the constructs of language-related and organization strategies and writing experience . The results of this study could carry significant implications for technology-assisted writing instruction in EFL contexts.


INTRODUCTION
The increasing use of Information and Communication Technologies (ICT) in higher education has had a significant impact on the way we write and the genres we create (Hyland, 1996). Social theorists and literacy scholars strongly believe that the use of ICT including the Internet has changed the way in which literacy is defined in today's world (Bolter, 2001;Coiro, 2003;Isbell & Reinhardt, 2002;M arcoux&Loertscher, 2009; Singh, M angalaraj&Taneja, 2010). Literacy, now has gone beyond "the ability to read and write" (Horning, 2004, p. 10) and has in addition embraced the skills of "being able to make sense of and navigate through several forms of information, including images, sounds, animation, and ongoing discussion groups" (El-Hindi, 1998, p. 1). The contribution of the electronic world into literacy development according to Karlsson (2002) has been discussed in terms of at least four contextual parameters: The web has brought a new technology to text -generation which enables different text structures; blurred the borders between public and private (or rather, it makes all texts globally available, whether intended or not; changes the traditional roles of readers and writers and challenges the dominant view of what writing is, and it has put new systems of meaning-making into the hands of homepage owners. Perhaps the most immediately obvious implication of computer-based writing in a language classroom is the use of word processors. Word processing emphasizes the process rather than the product of writing, encouraging students to brainstorm ideas and move through a series of drafts to a completed version of a paper (Lam & Pennington, 1995). Hyland (1996) has summarized the potential advantages of word processors as greater development of content, improvements in quality, awareness of writing as a process, greater fluency and accuracy, increased experimentation with language, more revisions, removal of handwriting barriers, and greater motivation. The disadvantages of using word processors, nonetheless, according to Hyland (1996) include increased focus on surface features, local revisions rather than global, premature completion of work, increased plagiarism, quantity at the expense of quality, preoccupation with physical appearance, increased anxiety and decreased writer collaboration.
In addition to word processors, the increasing use of electronic mail (e-mail) in educational contexts has inspired numerous scholars to explore the implications of this specific feature of the Internet in writing courses. Generally speaking, e-mail can encourage students to use computers in realistic, authentic situations in order to develop communicative and thinking skills (Singhal, 1997). E-mail is a useful tool for writing instruction as it permits teachers to establish classroom interaction and long-distance exchanges, motivating students to concentrate on fluency and meaning while writing for a real audience and purpose (Hyland, 1996).
Several studies have illustrated the importance of careful introduction of e-mail communications into writing classes. In his case study of semester 4 students of Portuguese at a US university, for instance, Kelm (1998) described how students were required to write at least 50 lines of email messages in Portuguese every week, to anyone, about anything, but copied to the teacher. The teacher was then able to provide tactful feedback on language issues. Kelm claimed that the student work became less passive, more participatory and imaginative, and recommended further research focusing on which L2 skills students acquire while using email and which level of proficiency is needed to gain benefit. Li (2000) compared 132 e-mail texts produced by newly arrived ESL students in response to four different pre-writing task demands and rhetorical purposesnarrative, informative, persuasive, expressivetwo involved interaction with peers, two did not. The texts with audience interaction had longer sentences, a consist ently higher level of syntactic and lexical complexity but also a higher frequency of error.
One issue related to the notion of technology -enhanced EFL literacy development which requires to be further examined is the nature of strategy use in this particular venue. Research has demonstrated that EFL learners utilize different strategies for reading print and hypertexts (e.g., M arandi&M okhtarnia, 2008); whether the same is true for the writing skill in two domains of technology -enhanced and conventional EFL classes is the focus of the current study. Exploring the question of whether EFL learners employ different writing strategies in two contexts of technology -assisted and regular writing classes could promise valuable potentials to shed more light on the process of EFL literacy development in today's digital world. Therefore, the research question addressed in this study included


In what ways does the reported writing strategy use of Iranian EFL learners differ in two different environments of technology-assisted and conventional writing classes?

Participants
Two intact general English classes of sophomore students majoring in Translation Studies at Azad University of Karaj took part in this study. Precisely speaking, eighteen students consisting of four males and fourteen females were grouped as Class A who were assigned technology-assisted writing assignments; and fourteen students including one male and thirteen female students were grouped as Class B who were asigned regular paper-and-pencil writing projects. Categorizing these two classes as on-line and print was done randomly.

Instrumentation
To collect data for this study, three instruments were utilized: Data regarding the participants' reported use of writing strategies was collected through a 64-item questionnaire adapted from Wei and Chen (2004) and Griffiths and Jordan (2005). This instrument consisted nine constructs of researching (W1), planning (W2), D e c e m b e r 0 3 , 2 0 1 5 supporting (W3), revision (W4), language (W5), organization (W6), content (W7), attitude (W8), and writing experience (W9). These items were concerned with the following issues.
Researching strategies (items 2, 7, 9, 40) aimed at tapping into the conscious decisions learners employed for seeking information which were required for completing their writing tasks. Language-related strategies (items 10,11,13,25), in addition, indicated techniques used for manipulating language elements such as grammar rules, and word choice to finish the writing tasks. Organization strategies (items 8,12,27,31,32,33,39,42,43,45,49), additionally included techniques required for establishing coherence in the written text. Content-related strategies (items20, 21, 37), moreover, aimed at examining whether students attended to the content of their writing or not.
Furthermore, attitude questions (items 51 to 64) referred to the orientation of the participants towards writing in English in general. Finally, the questions of the construct of writing experience (items 26, 35, 41) pointed to whether students wrote in English for purposes other than classroom assignments or not.
This questionnaire was piloted to a small number of students similar to the main participants of this study in order to be examined for its quality and relevance. Furthermore, the content validity of the questionnaire was assessed by discussing the items with two applied linguists who were expert in teaching writing. The Cronbach alpha coefficient for this instrument in the pre-test and post-test were 0.79 and 0.77 respectively indicating acceptable reliability estimates for this instrument.
In addition, the degree of computer familiarity of the participants was determined based on their responses to a questionnaire which was originally developed and used by Eignor, Taylor, Kirsch, and Jamieson (1998, as cited in Hedayaty, 2005) to assess the degree of computer familiarity among TOEFL examinees. The questionnaire which contained 20 items concentrated on four areas of access (item1), self-assessment of attitude and ability (items 3 through 7 and 9), and use of and experience with computers (items 2, 8, 12, 13 and 15 through 20), and use of and experience with related technologies (items 10, 11, and 14). The Cronbach alpha of this instrument at current study was 0.86.
Furthermore, In order to examine whether the participants of this study were initially homogeneous in terms of English proficiency, the grammar section of a M ichigan Test of English Language Proficiency which consisted of 40 items on various grammatical points was given to the participants before the introduction of the treatment.

Data Collection
Students of two intact general English classes took part in this study to allow the researchers to compare their reported use of writing strategies in two different venues of technology-assisted and conventional writing classes. For this purpose, class A was assigned ICTassisted writing projects, employing M icrosoft Word Processor to complete their writing projects and electronic mail to send the projects to their teacher and receive feedback from their her, whereas the other class, i.e., class B, was required to submit regular writing tasks and receive conventional print feedback. Overall, 10 sessions were held during the term where students were exp ected to submit 7 writing assignments on numerous conventional writing topics. On about 5 of these assignments, students were given feedback on grammatical mistakes/errors they had in their written texts. In addition, at the beginning and end of the term, a writing strategy inventory was given to the participants to examine whether significant changes occurred to their reported use of writing strategy or not following the introduction of the treatment. In addition,prior to administering the Writing Strategy Inventory, a Computer Familiarity Inventory and a grammar section of a M ichigan Test of English Language Proficiency were also administered to determine their level of computer familiarity and grammar proficiency of the participants.

RESULTS
The research question in the present study was directed towards examining the probable differences inthe reported use of writing strategies of EFL learners in two venues of conventional and technology-assisted writing classes.
In order to establish the homogeneity of the participants, a series of independent samples t-tests were conducted to examine whether the participants in two classes of technology -assisted and conventional writing were initially significantly different from each other in terms of computer familiarity and grammar proficiency. As presented in Table 1, no statistically significant difference was found between these two groups; therefore, it could be concluded that these groups were homogenous.
To address the question of the current study, first, the descriptive statistics of the participants' scores on both pre-tests and post-tests of the Writing Strategy Inventory in both technology-assisted and conventional writing classes are presented to provide preliminary information on the reported writing strategies. As Table 2  It was observed that prior to the treatment, participants in the conventional class reported a slightly higher use of writing strategies for all nine constructs of researching, planning, support, revision, language-related, organization, and content-related strategies as well as attitude and writing experience.

Table 2. Descriptive statistics of the writing strategy inventory for both technology-assisted and conventional classes
Following the treatment, the reported mean of the writing strategiesin the technology-assisted class (See Table 2  These items belonged to categories of planning strategies, language-related strategies, organization strategies, and writing experience respectively. The results have been presented in detail in Table 5. To compare the reported use of the writing strategies in two technology-assisted and conventional classes, a series of independent ttests were conducted between the obtained scores.   Nevertheless, for the rest of the constructs, i.e., Researching, Planning, Support, Revision, Organization, and Content -related strategies and Attitude and Writing Experience no significant difference was observed.

DISCUSSION
The purpose of this study was to shed light on how writing strategies employed by Iranian EFL learners in a technology -assisted classroom differed from those in a conventional setting. The results revealed that in both electronic and print mediums, the participants demonstrated a moderate awareness of writing strategies with 54% of reported strategies with a mean score between 2.5 to 3.4.
Considering the subcategories of Writing Strategy Inventory in the post-tests, the participants reported the highest mean scores for the constructs of attitude and organization in both print and electronic environments with the strategies in the conventional setting enjoying higher values. The lowest reported mean in the post -tests, in addition belonged to researching strategies followed by the construct of writing experience in both mediums.
In the pre-tests, nevertheless, the highest means were reported for the constructs of attitude and content-related strategies in both contexts with the print setting enjoying higher means. The lowest means belonged to the constructs of researching and writing experience in both mediums with the print environment having the lower means. On the whole, the overall pattern of writing strategy use in both mediums was the same (Figure 1).

Fig 1: Comparison of the mean scores in print and electronic mediums in the post-tests
The results of the paired-sample t-tests conducted between the pretests and posttests in both environmentsrevealed that merely in the technology-assisted class, there was a statistically significant difference between not only the total mean scores but also the mean scores of the constructs of language-related and organization strategies and the construct of writing experience with the post-test manifesting higher values. In other words, the treatment in the technology-assisted writing class had a significant impact on enhancing the metacognitive awareness of language-related and organization strategies of the participants as well as their writing experience.
In addition, the results of the independent sample t-tests in order to compare the reported use of writing strategies in both technologyassisted and conventional venues demonstrated that there was a statistically significant difference just for the construct of languagerelated strategies. After the treatment, the participantsdemonstrated greater use of language-related strategies in the technology assisted writing class.
This finding is indeed in line with that of Sullivan and Pratt (1996) who found small but significant increase in the writing ability of their participants in their comparison of a computer-assisted writing class using networked computers for writing with a traditional class. These results also corroborate the findings of Robinson-Staveley and Cooper (1990) who reported significant performance differences for students who used word processing for their writing projects. Areas in which students showed gains included f ewer mechanical problems, greater average sentence length, greater number of complex sentence structures, and overall higher scores on final products.
This result also agreed with the findings of Fang (2010) who demonstrated that a majority of his students benefited from usin g computer-mediated feedback to revise their essays. M oreover, interview data of his study revealed that the computer-mediated feedback had a positive effect on the development of their writing skill, particularly in suggesting changes for form rather than content.
Some findings of this study, however, turned out to be contrary to similar studies that examined the effects of word processing on L2 writing. Although the revision capacitiesof M icrosoft word processorsuch as spell-and-grammar checking, block moving, block deleting,formatting, thesaurus, and storage of informationare maintained to have the potential to facilitate operations such as text generating, revising, and editing (Li, 2006), the results of the independent sample t -test conducted between the mean scores of the strategy of revision in both technology-assisted and conventional classes did not yield any significant differences. Though the mean score in the electronic environment (M = 30.93; SD = 5.05) appeared to be higher than that of print (M = 28.73; SD = 5.87). T his finding is in contrast to that of Snyder (1993) who asserted that reviews of research on word processing indicated an increase in revisions, fewer mistakes, and more correction.
The current study provided some insights into the nature of writing strategy use in both electronic and print mediumfor EFL learners. With the risinguse of ICT technologies in education, it is increasingly believed that the concept of literacy is going under dramatic changes (e.g., Reinking, 1995;Coiro, 2003). "Today, the definition of literacy has expanded from traditional notions of reading and writing to include the ability to learn, comprehend, and interact with technology in a meaningful way" (Pianfetti, 2001, p.256 found significant differences in the metacognitive awareness of overall, problem solving, and support reading strategies for hypertext and printed-text reading processes. However, no significant difference was found for the metacognitive awareness of global reading strategies. Regarding writing strategies, the results of theindependent t-testsrevealed significant differences just for language-related strategiesin two technology-assisted and conventional classes. This finding of in fact suggest that technology-assisted writing should not be treated the same way as the conventional writing; therefore, teachers are required to provide relevant assistance regarding the requirements of this medium; as Rimor and Kozminsky, (2002) asserted that electronic environment deserves special emphasis due to the fact that it is considered an "ill defined" environment in which the strategies employed are not explicit or predefined.
An additional implication of this study could pertain to the domain of computer-basedassessment of writing. With the increased influence of high-stake tests such as computer-based and Internet-based TOEFL on the academic and educational life of students, more care is needed to be paid to adequately prepare students for such important tests. According to M acArthur and Schneiderman (1986), factors such as the developmental stage of the writer, the previous exposure to computers, and t he amount of keyboarding proficiency, and access to word processors over an extended period of time bear a significant impact on the success of technology-assisted writing if it is accompanied by responsive instruction.
On the whole, the application of ICT technologies in language classrooms and more specifically writing classes, in spite of difficulties such as the availability and cost-effectiveness of new technologies including the Internet connection and the willingness of teachers to apply them, bears great potentials to enhance the quality of language education including the teaching of writing skill, provided that the educationalists and curriculum developers pay due attention to the pre-requisites of such paramount decisions.