STUDENTS’ WRITING PERFORMANCE IN SELECTED PREWRITING SETTINGS Alma Cita B. Soria-Calimbo1,*, Mayflor D. Prantilla-Arambala2, Amita B. Sususco3, Onnah Pierre P. Peralta-Talle4

ABSTRACT The study investigated the effects of three prewriting settings (individual, group face-to-face and group “chat” via synchronous computer -mediated-communication) on college freshmen's writing performance. Subjects were 203 college freshmen enrolled in five sections of freshman writing taught by four teachers. Students in each section were randomly assigned into one of the prewriting settings, and all three settings were facilitated by the same teacher per section. Students were exposed to four writing sessions; namely, expository, descriptive, narrative, and argumentative. Essays were rated using an adapted holistic rubric considering three features of the compositions; namely, content, organization, mechanics and grammar. Comparison of mean differences was performed using ANOVA followed by the Tukey HSD procedure. Although the prewriting settings had no significant effect on the content of the four essays and on the organization of most of the essays, it was found that the mean scores for mechanics in the group “chat” were significantly higher than those in the other two settings for descriptive and argumentative essays. Among the three prewriting settings, group “chat” was found to have a positive effect on students’ writing performance. Free writing and listing were found to be the most preferred prewriting techniques by students. Integration of computer technology into the writing process is then recommended. Key Words: computer-mediated-communication, prewriting, writing performance

1,*

Central Mindanao University, Musuan, Bukidnon, (Corresponding Author: [email protected]) Central Mindanao University, Musuan, Bukidnon, ([email protected]) 3 Central Mindanao University, San Jose, Malaybalay City, ([email protected]) 4 MSU-IIT, Iligan City, ([email protected]) 2

126

INTRODUCTION One of the main thrusts of any educational institution is to make its graduates globally competitive. Whether it will be in school, in the workplace, or in other social settings, success is dependent on how individuals perform activities, assignments or tasks which demand good writing capacities – writing essays, preparing research papers, taking essay exams, submitting reports, writing business letters, critiques or reaction papers and making written outlines for delivering oral reports or speeches (Wingersky, Boerner, & Holguin-Balogh, 1999; Langan, 2001).

Writing is a difficult skill. While many factors may contribute to this writing reality, experience and studies showed that to get started in writing is the most difficult part in the writing process. As Murray states, “Prewriting usually takes about 85% of the writer’s time” (in Priddy, 2009). However, such difficulty can be addressed by idea-stimulating prewriting activities. It is then imperative for language teachers to investigate what sort of prewriting activities work well with their students in their writing classes. Consequently, a number of studies have been conducted on the effects of some strategies of prewriting on students’ writing. Xianwei (2009) investigated the effects of prewriting discussion on the language quality of argumentative writings and found that those who participated in prewriting discussion “wrote more fluently than those who brainstormed individually.” Similarly, in their study Best Practices: Pre-Writing Collaboration Leads to Better Writing, Fitze and Glasgow (2009) posited the beneficial effects of prewriting in the group setting; i.e., the group face-to-face and group “chat” setting. They found that drafts written after group prewriting were interrelated to each other than those written after individual prewriting. Scheiker-Mara and Mara (2000) found that written expression was improved by the introduction of prewriting activities and that studentwriting anxiety could be reduced through a prewriting activity-based writing program. Such finding was confirmed by the study of Toshiko (1996) which showed the positive effects of prewriting activities in dealing with the writer’s block. Wijaya (2000) explored the notion of collaborative work in the prewriting stage and discovered that collaborative prewriting could help students generate interesting ideas, which would help them engage in dynamic writing.

127

Brodney, Reeves, and Kazelskis (1999) investigated the effects of selected prewriting treatments on students’ expository compositions. They found that the type of prewriting treatment significantly affected scores of students. Some studies were done to investigate the effects of computer-based prewriting activities on students’ writing. Eib and Cox (in Ulusoy, 2006) noted an increase in motivation and writing length and proficiency in the organization for students who integrated computer technology in the writing process. In the same vein, Sullivan (2006) ventured on the students’ perceptions regarding web-based prewriting activities which could be done in a synchronous or in an asynchronous manner and found the advantages of asynchronous on-line prewriting activity over the synchronous one. In addition, Sturm and Rankin-Erickson (2002) found that computermapping as a prewriting activity had significant positive effect on students’ attitude towards writing when compared to hand-mapping prewriting activity. Proske and Narciss (n.d.) investigated the effects of computerbased prewriting activities on students’ writing and concluded improved students’ motivation in writing. On the other hand, Fitze and Glasgow (2009) found that although students’ drafts were related to each other and students got more ideas in the group prewriting settings (face-to-face and computer “chat”), there was no significant difference between the quality of the arguments produced by students in the individual and group prewriting setting. Similarly, in the study of Huang and Hung (2008), although they noted students’ overwhelming preference of synchronous computer-mediatedcommunication, such mode did not have statistically significant effects on writing performance. In addition, Dujsik (2008) investigated the effects of computer-based prewriting strategies on students’ writing and found that although better scores were gained by students exposed to computer-based prewriting strategy than those not exposed to the said strategy, the difference between the two groups was not statistically significant. Lastly, Liu (2005) ventured on two prewriting strategies, face-to-face and synchronous electronic discussion, and found that although synchronous computer-mediated-communication received an overwhelming preference of students, there was no statistically significant difference between the average content scores of drafts written after face-to-face and synchronous CMC prewriting discussions. Indeed, researchers have varying results concerning the effects of different prewriting treatments on students’ writing which indicate that more research is needed to be conducted to settle such contradicting findings on the matter. 128

Conceptual Framework of the Study This study was anchored on the concept of cooperative or collaborative learning and LevVygotsky’s (in McLeod, 2007) social learning theory. Dr. Robert Slavin (in Arends, 2001), one of the founders of cooperative learning, believed that letting students work in groups could help them excel in academic tasks. Through tutoring, low-achieving members in a group could be helped by the high-achieving or the advanced members who in turn could also sharpen their cognitive capacities for the tutoring requires good thinking skills. More importantly, group work would encourage students to work interdependently for a common objective and in the long run could help them develop the skills of collaboration and cooperation which are so critical in the real world. Slavin’s belief was in consonance with Vygotsky’s (in McLeod, 2007) argument on the significant role of social interaction in cognitive development. He stressed that interactive and communicative environment could enhance and maximize learning. He suggested that learning environments must include activities which would allow learners to interact with each other. However, while cooperative learning model is found to be useful in certain academic tasks, it could be a challenge on the part of the teacher to apply such a model in writing tasks for writing is basically done solo. Figure 1 presents the conceptual framework of the study. Three prewriting settings were investigated whether or not they affected the writing performance of students.

Prewriting Settings

Writing Performance



Individual





Group-face-to-face



Group “chat”

Unified and wellsupported



Coherent



Error-free

Figure 1. Conceptual Framework of the Study 129

Objectives The study aimed to investigate the effect of varied prewriting settings on the writing performance of students. Specifically, it aimed to determine the effect of the selected prewriting settings on the quality of students’ compositions; to identify which prewriting setting results to better writing performance of students; and to identify what prewriting techniques were most preferred by students.

METHODOLOGY The study employed quasi-experimental research design with the level of significance (alpha) set at 0.05. It was conducted to the first year college students of Central Mindanao University, Musuan, Bukidnon. The sample comprised of students enrolled in English 12 (Writing in the Discipline) during the second semester of S.Y. 2010-2011. For the purpose of convenience, only those writing classes handled by the researchers were considered. A total of 203 students from five intact English 12 classes participated in the study. The study utilized essay writing tests and a survey questionnaire as instruments. Using the prewriting sheets and the outline which evolved from their prewriting activity, students developed a five-paragraph essay based on a subject agreed by the authors. Students were subjected to four writing sessions, namely: expository, descriptive, narrative and argumentative. Some students were not able to finish their drafts within the time allotted for drafting. Hence, a total of 763 completed compositions was considered (203 expository, 187 descriptive, 188 narrative and 185 argumentative). Out of this figure, 263 compositions evolved from individual prewriting, 255 compositions evolved from group face-to-face prewriting and 245 compositions evolved from group “chat” prewriting. A two-item survey questionnaire was developed to identify the prewriting techniques most preferred by students. In the first item, the common prewriting techniques were listed, and students were instructed to put a check beside the technique or techniques (multiple answers were considered) which they found most useful in generating ideas for their essays. In the second item, students wrote the observed advantages and disadvantages of the prewriting setting they were assigned to which could be useful in validating the data on the effect of the prewriting settings on their writing performance.

130

Students were given a diagnostic writing test at the start of the semester after which scores were ranked. Based on their ranks, students in each class were randomly assigned into one of the prewriting settings: individual, group face-to-face, and group “chat” via computer. All three prewriting settings were facilitated by the same instructor per class. Individual and small group face-to-face prewriting activities were done in the students’ regular writing classrooms during the usual class time. Computer “chat” prewriting activity was done in the computer laboratory room. Prewriting activities done in the classroom were allotted with 30 minutes, and computer “chat” was allotted 45 minutes. Electronic interaction could be a slow process for it requires students to think and to type what they want to say usually considering grammar and spelling at the same time (Liu, 2005). Draft writing was done in the classrooms and was allotted one hour. The compositions were rated by the instructors using an adapted holistic rubric (Langan, 2001) considering three features of the compositions: content (allotted 25 points), organization (allotted 25 points), and mechanics and grammar (allotted 30 points). Writing performance was measured through the average holistic scores of students. The data were summarized using frequencies, means, and standard deviations. Assumptions for the analysis of variance were checked before performing the analysis for treatment effects. Because there was at least 62 subjects in each group, a normality test for each group was deemed unnecessary. The Levene’s test for homogeneity of variances was performed for each ANOVA, and all tests revealed nonsignificant results for comparisons of the variances. A significant ANOVA was followed by the Tukey procedure for multiple comparisons of means. RESULTS AND DISCUSSION Effect of the Selected Prewriting Settings on the Quality of Students’ Compositions For expository writing, the average scores of students were not statistically different on all of the three components of the rubric. The average scores for content were not different across prewriting settings, F (2,200) = 1.69, p= .19. This result confirmed the findings of Dujsik (2008) that different prewriting strategies had no significant impact on students’ writing.

131

The same observation was noticed for organization, F (2,200) = 1.95, p= .14 and for mechanics, F (2,200) = .20, p=82 which was in consonance with Liu’s (2005) finding. Investigating the effect of face-toface and synchronous computer-mediated-communication, she found that the two prewriting settings had no significant impact on the students’ drafts. This finding was in contrast with the number of studies proving the positive impact of prewriting strategies on students’ writing (Xianwei, 2009; Fitze & Glasgow, 2009; Eib & Cox, in Ulusoy, 2006; Sturm & Rankin-Erickson, 2002; Scheiker-Mara & Mara, 2000; Wijaya, 2000; Brodney, Reeves & Kazelskis, 1999; Toshiko, 1996; Proske & Narciss, n.d.). This finding would seem to suggest that the quality of the student essays were not solely influenced by prewriting which is in consonance with what Dujsik (2008) found in his study that students’ writing development was influenced by the complex interplay of factors “including writing strategy use, writing processes, writing tasks, task conditions, their past writing experience, and their language proficiency.” Table 1. Mean Scores (Standard Deviations) of Students across Different Prewriting Settings for Expository Writing Feature

Prewriting Setting Individual

Face to face 15.6 (5.0)

Group Chat 17.0 (4.2)

Content

16.5 (4.2)

Organization Mechanics

p .19

16.5 (3.5)

15.3 (4.2)

16.3 (3.8)

.14

18.6 (4.9)

18.1 (4.6)

18.5 (5.0)

.82

For descriptive writing, the average scores for content were not different across the prewriting settings, F (2,188) = 0.27, p= .76. This was also true for organization, F (2,188) = 0.26, p= .77. However, the average scores for mechanics were found to be different across prewriting settings, F (2,188) = 3.22, p= .04. The mean score for the group chat (M= 18.1, SD = 4.3, n= 66) was significantly higher than that of the individual setting (M= 16.1, SD= 5.0, n=62) but not significantly higher than the face-to-face setting (M= 17.5, SD= 3.9, n=63). Mean scores for the individual and face-to-face settings were not found to be significantly different from each other. 132

This result confirmed the finding of Liu (2005). In her study, she investigated the effect of face-to-face and synchronous electronic discussion prewriting activities on students’ writing and found that fewer spelling and grammar mistakes were made by students who discussed through the computer. This is because electronic discussion resembles written communication for students “have to think of using the right words and grammatical sentences.” In addition, it also resembles face-toface discussion for students share ideas freely (Chun, in Liu, 2005). Liu (2005) stressed that synchronous computer-mediated-communication is a “good environment for fostering the use of more formal and complex language on both lexical and syntactical level.” This result would imply then that integration of computer technology in a writing instruction could help improve students’ grammar skills. Table 2. Mean Scores (Standard Deviations) of Students across Different Prewriting Settings for Descriptive Writing Feature

Prewriting Setting Individual

Face to face

Group Chat

p

Content

14.0 (4.8)

14.6 (4.4)

14.5 (4.7)

.76

Organization

14.4 (4.4)

15.0 (4.0)

14.5(4.1)

.77

Mechanics

16.1 (5.0)a

17.5 (3.9) ab

18.1 (4.3) b

.04

Note: Means sharing a common letter are not statistically different at = .05 using the Tukey HSD procedure.

For narrative compositions, average scores for content were not different across prewriting settings, F (2,190) = 2.47, p= .08. For organization, however, the average scores were found to be different across prewriting settings, F (2,190) = 3.51, p=.03. The mean score for the group “chat” (M=16.9, SD =3.9, n=64) was significantly higher than that of the group face-to-face setting (M=15.1, SD =3.6, n=63) but not significantly higher than the individual setting (M= 15.9, SD= 3.4, n= 66). Mean scores for the individual and face-to-face settings were not found to be significantly different from each other.

133

Also, the average scores for mechanics in narrative writing were found to be different across prewriting settings, F= (2,190) = 3.55, p=.03. The mean score of the group chat (M=19.0, SD =4.2, n=64) was significantly higher than that of the group face-to-face setting (M= 17.2, SD= 3.9, n=63) but not significantly higher than that of the individual setting (M= 17.8, SD=35, n=66). Mean scores for the individual and group face-to-face settings were not found to be significantly different from each other. These results were in consonance with the findings of the previous researches (Fitze & Glasgow, 2009; Liu, 2005) that synchronous computer-mediated-communication could enhance students’ ability to develop more coherent essays and could improve students’ grammar and sentence skills. Again, this finding poses a challenge to integrate computer technology in a writing class. Table 3. Mean Scores (Standard Deviations) of Students across Different Prewriting Settings for Narrative Writing Feature

Prewriting Setting

Individual

Face to face

Group Chat

p

Content

16.2 (3.7)

16.3 (4.1)

17.6 (4.1)

.08

Organization

15.9 (3.4)ab

15.1 (3.6)a

16.9 (3.9) b

.03

19.0 (4.2) b

.03

Mechanics

17.8 (3.5)ab

17.2 (3.9)a

Note: Means sharing a common letter are not statistically different at = .05 using the Tukey HSD procedure.

For argumentative writing, the average scores for content were not different across prewriting settings, F= (2,194) = 1.91, p=.15 which was in consonance with Fitze & Glasgow’s (2009) finding that there was no significant difference between the individual and group settings although they reported that students who did the face-to-face group setting produced better arguments than students in chat and individual setting. The finding was also true for organization, F= (2,194)=0.71, p=.49 which is in contrast with what Glasgow (2009) found that drafts written after collaborative prewriting were more related to each other and that arguments produced after individual setting were “occasionally irrelevant or factually questionable.” 134

However, the average scores for mechanics in argumentative writing were found to be different across prewriting settings, F(2,194)= 3.68, p=.03. The mean score for the group “chat” (M=18.3, SD=3.5, n=65) was significantly higher than that of the individual setting (M=16.4, SD 4.6, n=65) but not significantly higher than that of the group face-to-face setting (M=17.0, SD=4.1, n=67). Mean scores for the individual and faceto-face settings were not found to be significantly different from each other. This finding confirmed the result of the study conducted by Xianwei (2009) on the effects of individual and group prewriting activities on the language quality of argumentative writings which noted that students who were assigned in prewriting discussion wrote “more fluently than those who brainstormed individually.” Table 4. Mean Scores (Standard Deviations) of Students across Different Prewriting Settings for Argumentative Writing Feature

Content

Prewriting Setting Individual

Face to face

Group Chat

p

14.5 (4.1)

15.9 (4.8)

15.8 (4.3)

.15

Organization

14.6 (3.7)

14.8 (3.6)

15.3 (3.4)

.49

Mechanics

16.4 (4.6) a

17.0 (4.1) ab

18.3 (3.5) b

.03

Note: Means sharing a common letter are not statistically different at = .05 using the Tukey HSD procedure.

It could be noted that among the three features of the compositions, only the content was not affected by the prewriting settings. Although such finding was in consonance with some researches previously conducted (Liu, 2005; Huang & Hung, 2008), it was quite surprising because prewriting is basically intended for content generation. It was also surprising that grammar was affected by the prewriting settings for grammar and mechanics were not basically considered during prewriting stage.

135

The lack of statistically significant difference in the content of essays written after individual and collaborative prewriting activities (i. e., the group face-to-face and group computer “chat”) could be attributed to the nature of writing as a process. While it is true that more ideas can be generated in the group prewriting settings, students wrote their drafts independently, and during the drafting process, there would be some changes, and there would be new ideas coming to the minds of the writers which did not occur in prewriting (Langan, 2001). The written product then could be mainly an individual output which might be or might not be influenced by prewriting (Liu, 2005). It is on this accord that prewriting can be nicknamed the Cinderella of writing because despite that it is the most tiring part, “only the neat piece receives the merit” (Antoniazzi, 2005). Students were used to writing alone. Hence, although they might have gotten many ideas from collaborative prewriting, their writing habit and the drafting process which was done independently might have more influence in writing. Prewriting Setting which Results to Better Writing Performance of Students The average holistic scores for expository writing were not different across prewriting settings, F (2,200=1.53, p= .22). Also, the average holistic scores for descriptive writing were not different across prewriting settings, F(2,188=.90, p=.41). This finding was also true for the average holistic scores for argumentative writing, F(2, 194=2.28, p=.10). This result confirms Huang and Hung’s (2008) finding that synchronous computer-mediated-communication and face-to-face discussions had no statistically significant effects on students’ writing performance. On the other hand, the average holistic scores for narrative writing were found to be different across different prewriting settings, F (2,190=1.88, p=.01). The total mean score for the group chat (M= 53.6, SD=10.8, n=64) was significantly higher than those of the individual (M=49.0, SD= 9.7, n=66) and group face-to-face (M=48.6, SD=9.4, n=63) prewriting settings. Total mean scores for the individual and group faceto-face prewriting settings were not found to be significantly different from each other.

136

Although the group computer “chat” significantly affected one among the four types of essays, the data show that among the three prewriting settings, the group computer “chat” prewriting activity had a significant positive effect on students’ writing, which is in consonance with the results of the previous studies conducted investigating the role of computers in writing instruction (Proske & Narciss, n.d.; Eib & Cox, in Ulusoy, 2006; Sturm & Rankin-Erickson, 2002). Table 5. Mean Scores (Standard Deviations) of Students across Different Prewriting Settings for the Four Types of Essays Type of Essay

Prewriting Setting

Exposition

Individual 51.6 (11.3)

Face to face 48.6 (12.7)

Group Chat 51.9 (11.7)

p .22

Description

44.6 (13.4)

47.1 (10.8)

47.1 (12.0)

.41

Narration

49.0 (9.7) a

48.6 (9.4) a

53.6 (10.8) b

.01

Argumentation

45.6 (11.1)

48.0 (10.9)

49.5 (9.7)

.10

Note: Means sharing a common letter are not statistically different at = .05 using the Tukey HSD procedure.

The statistically significant positive effect of computer “chat” on students’ writing could be attributed to some factors crucial to effective writing. One is the motivational factor. In her study conducted to determine students’ preference between face-to-face and synchronous electronic discussion as prewriting activities, Liu (2005) found that 76% of the participants preferred electronic discussion for the reason that it “softens” stress which shy students usually experienced in face-to-face discussion. In another study having the same focus, it was found that although synchronous computer-mediated-communication “failed to produce any statistically significant effects” on students’ writing, these were met with overwhelming preference from students (Huang & Hung, 2008). Similarly, Proske and Narciss (n.d.) found that computer scaffolding improved students’ motivation in writing. Another is the convenience factor. Inputs from electronic discussion could be saved in computers which could be useful in the drafting stage, in contrast to the inputs from face-to-face discussion in which some important ideas might be lost or forgotten after prewriting. Also, electronic discussion is time-efficient than face-to-face discussion. Everybody could write at the same time through computer “chat” but not everybody could talk during face-to-face discussion. 137

This finding of the current study then confirms the relevance of integrating computer technology in writing instruction. Most Preferred Prewriting Techniques by Students Among the five most common prewriting techniques, free writing, and listing were the most preferred by students. While listing is generally useful for both individual and group prewriting, free writing is generally useful for individual prewriting. The data showed that in the individual and group prewriting setting, free writing consistently received the highest preference from students. This finding could be attributed to the fact that among the five prewriting techniques, these two techniques are the freest and the most unstructured ways of generating ideas (Trupe, 2001; Langan, 2001). Table 6. Prewriting Techniques Used by Students in the Prewriting Settings Prewriting Setting

Prewriting Techniques Free writing

Questioning

Listing/ Brainstorming

Mapping

44 (53.0 %)

7 (8.4%)

21 (25.3%)

3 (3.6%)

8 (9.6%)

27 (42.2 %)

11 (17.2%)

19 (29.7%)

1 (1.5%)

6 (9.4%)

Chat

34 (49.3%)

11 (15.0%)

19 (27.5%)

2 (3.0%)

3 (4.3%)

Total

105 (48.6%)

29 (13.4%)

59 (27.3%)

6 (2.8%)

17 (7.9%)

Individual Face-to-face

Outlining

CONCLUSIONS AND RECOMMENDATIONS In conclusion, the prewriting settings have no influence on the content of the students’ essays; however, these have a positive influence on the mechanics and grammar. Computer scaffolding results to better writing performance of students. Finally, unstructured prewriting techniques such as free writing and listing are the most useful in generating ideas. As computer “chat” was found to have significant positive effect on students’ writing performance, it is recommended that teachers should design writing activities integrating computer technology. Also, since prewriting is just among the stages in the writing process, another study can be conducted on how the other stages (drafting and evaluation) should be facilitated to improve writing performance of students. 138

LITERATURE CITED Antoniazzi, M. I. C. (2005). Prewriting in EFL. Retrieved from www.facso.uchile.cl/publicaciones/enfoques/.../ Casado_N7_2005.pdf Arends, R. I. (2001). Learning to teach (5th ed.). NY: McGraw-Hill Co. Brodney, B., Reeves, C. & Kazelskis, R. (1999). Selected prewriting treatments: Effects on expository compositions written by fifthgrade students. The Journal of Experimental Education. Retrieved from http://www.istor.org/pss/20152612 Dujsik, D. (2008). The effects of pre-writing strategy training guided by computer-based procedural facilitation on ESL students’ strategy use, writing quantity, and writing quality. Retrieved from http:// scholarcommons.usf.edu/etd/221 Fitze, M. & Glasgow, R. (2009, February). Best practices: Pre-writing collaboration leads to better writing. Paper presented at the 1st Philippine International English Language Conference, SM Mall of Asia Complex Pasay City. Huang, H. T. D. & Hung, S. T. A. (2008). Chatting to write: Synchronous computer-mediated-communication in an EFL writing classroom. Retrieved from http://www.editlib.org/p/29873 Langan, J. (2001). College writing skills (5th ed.). Singapore: McGrawHill Co. Liu, M. (2005). Face-to-face and synchronous electronic discussion as prewriting activity in the foreign language class. CELEA Journal, 28 (4). Retrieved from http://www.celea.org.cn/teic/62/62-46.pdf McLeod, S.M. (2007). Lev V ygotsky. Retrieved from http:// www.simplypsychology.org/ vygotsky.html Priddy, J. (2009). W riting process: Pre-writing resources. Retrieved from http://www.writingfix.com/Process/PreWrite.htm Proske, A. & Narciss, S. (n.d.). Supporting prewriting activities in academic writing by computer-based scaffolds: Is more support more meaningful? Retrieved from http://www.springerlink.com/content/ xj270467j764v73q/ Scheiker-Mara, K.E. & Mara, W.T. (2000). Investigating the effects of prewriting activities on writing performance and anxiety of at-risk students. Retrieved from http://www:ingentaconnect.com/content/ routledg/urpy/2000, 00000021/ 00000002/ art 00002

139

Sturm, J. M & Rankin-Erickson, J. M. (2002). Effects of hand-drawn and computer-generated concept mapping on the expository writing of middle school students with language disabilities. Learning Disabilities Research and Practice, 17(2), 124-139. Retrieved from www.ets.org/Media/Research/pdf/RR-08-55.pdf Sullivan, K. B. A. (2006). ESL students’ perceptions of web-based prewriting activities. Retrieved from http://www.archive.org/stream/ eslstudentsperce00sulluoft/ eslstudentsperce00sulluoft-djvu.txt Toshiko, Y. (1996). Effects of prewriting activities on EFL students’ writing. Retrieved from http://www. csupomona.edu/~uc/tutor/tutman 41.5 html Trupe, A. L. (2001). Prewriting Practices. Retrieved from http://people.bridgewater.edu/~atrupe/ENG315/prewriting.htm Ulusoy, M. (2006). The role of computers in writing process. The Turkish Online Journal of Educational Technology, 5(4). Retrieved from http://www.tojet.net/articles/v5i4/548.pdf Wijaya, J. (2000). Collaborative work in the prewriting stage. Retrieved from http://puslit2.petra.ac.id/ejounal/index.php/ing/article/ ViewArticle/15458 Wingersky, J., Boerner, J. & Holguin-Balogh, D. (1999). W riting paragraphs and essays: Integrating reading, writing, and grammar skills. USA: Wadsworth Publishing Company. Xianwei, P. (2009). Effects of prewriting discussion on the language quality of argumentative writings. Teaching English in China-CELEA Journal, 32(5). Retrieved from www.celea.org.cn/teic/87/87-16.pdf

140