Pen-and-paper or online? An academic writing course to teacher-trainees

Abstract This study attempted to examine the effectiveness of a writing course designed with the process genre approach and a blended learning approach (BLA) for teacher trainees, who are at the same time English as a foreign language learners. An action research was designed to implement the course content and collect data in relation to the students’ performances in pen-and-paper and online modes of teaching and learning. The results indicated that the participants’ performances in essay writing improved significantly both during pen-and-paper work and online work. However, when their performances were compared in the two modes of teaching, it appeared that the participants produced better written products when they were taught using the BLA.


Introduction
The teaching and learning of foreign language writing has been investigated in many studies (Badger & White, 2000;Scivener, 2011). Nonetheless, engaging learners in the process of both learning and teaching writing simultaneously, and giving and receiving written constructive feedback has not yet been focused on. In northern Cyprus, where English is a foreign language, the pre-service English Hanife Bensen Bostanci and Çise Çavuşoğlu ABOUT THE AUTHORS Hanife Bensen Bostancı's research interests range from error analysis, written corrective feedback in English as a foreign language settings, code-switching, syllabus designs, and innovative language teaching methods.
Çise Çavuşoğlu's research interests cover the relationship between ethnicity, culture, everyday language use, and attitudes of participants toward the learning and teaching of English in multicultural settings. The research reported in the article regarding the effects of pen-and-paper and online work in an academic writing course is directly related to completing writing tasks in class and online. Completing writing tasks online addresses the issue of time and syllabus constraints which are specifically highlighted in foreign language contexts. Such issues will be abolished with the incorporation of technology and online work.

PUBLIC INTEREST STATEMENT
The study sought to reveal the effectiveness of a writing course designed implementing pen-andpaper portfolio work and online work to produce essays. Sixteen English as a foreign language teacher-trainees acted as the participants. The participants produced two different types of essays for both pen-and-paper work and online work, wrote two drafts for each essay, and received feedback before submitting their final products during a period of 16 weeks. The 16 weeks were divided into two according to the two modes of teaching then compared at the end of the course and the findings revealed that both modes improved the students' writing performances. However, when the two modes of teaching were compared, it was revealed that the students preformed better in the online mode. Integrating online work in writing courses seems to be more beneficial and effective than pen-and-paper work.
language teachers are English as a foreign language (EFL) students themselves. Therefore, in this context, English language teachers are usually non-native speakers. This poses a significant challenge in their training as they need to improve their writing skills as learners before they can learn the methods of teaching writing. Learners in this study have one-to-one experience with regard to both learning how to academically write, give and receive written corrective feedback from different agents, and teach a specific text type which will contribute to their upcoming classroom teachings.
With regard to the teaching of writing, students in the English Language Teaching (ELT) departments take two courses related to the teaching of the main skills and writing is covered in an integrated manner, which does not specifically focus on all approaches to EFL writing, giving, and receiving written constructive feedback and writing assessment. A writing course that would address the needs of EFL prospective English language teachers was accordingly designed. There is a general consensus among English language teachers as well as among lecturers in the ELT departments that students are weak in writing courses (Bensen & Çavuşoğlu, 2017). This problem is invariably reflected on students' performances in other courses, which may not be directly related to the writing skill itself, as their requirements usually entail demonstration of good writing skills both in examinations and assignments. It is presumed that the learners' academic backgrounds have a great influence on their poor writing products.
Even though many approaches have been introduced to teach the writing skill in the EFL context (Badger & White, 2000;Scivener, 2011), an eclectic approach is considered the best when teaching any skill in EFL classrooms as it fits the specific needs of the students. However, choosing the "right" eclectic approach to fit the needs of the students is a challenging task for the teacher due to students having different linguistic and cultural backgrounds. With the emergence of the approaches that concentrate on the processes essential to be able to produce a written product, language teachers are encouraged to spend more time on the writing skill and thus pay more attention to the evaluation of work in progress (Badger & White, 2000). Assessment is also a crucial aspect when writing is concerned. Teachers in this context usually employ summative assessment focusing on results in examinations and quizzes due to the large amount of students in university classes together with the teaching approach employed mentioned earlier. Even though formative assessment is seen to improve students' learning in the long run, it is difficult to be adopted in such classes. In addition, employing portfolios in writing courses "represent multiple measures of student's writing ability, providing more accurate assessments of competence across a range of genres, but also help students to understand more about the genres they have studied" (Hyland, 2007, p. 162). One of the main benefits of portfolios is reflection. Learners are able to compare different genres and writing experiences and consider their writing and the criteria employed for judging it (Hamp-Lyons & Condon, 2000). Furthermore, motivation should be taking into account when designing any course as it is considered an integral part of learning. When teaching any course, it is difficult to keep students motivated. For this reason, teachers especially EFL teachers implement various techniques and methods to keep their students motivated. Keeping learners motivated particularly in a writing course is an intricate task for the teacher. For the learner, writing is not only a daunting task but also rather boring. Unlike the other fundamental skills (listening, speaking, reading), writing is considered to be the most challenging skill as it requires the writer to possess extended knowledge in both lexis and syntax, and the principles of organization in a second or foreign language (L2) to produce a good piece of writing (Tangpermpoon, 2008).
The project from which the data in this article comes from aimed to overcome such issues and move away from traditional writing classrooms by adopting a blended-learning approach (BLA). A BLA is defined as a combination of face-to-face classroom activities and online work (Macdonald, 2006;Oliver & Trigwell, 2005;Staker & Horn, 2012) in teaching writing in an EFL context. Thus, this article aims to report the findings of an action research, where a new syllabus was designed with a BLA to teach academic writing to EFL learners, concentrating on the processes necessary in order to produce a good product. More specifically, the focus is on the effects of implementing a BLA in an EFL writing course designed with a process genre approach (PGA). The PGA focuses on the processes needed to produce a good product according to a specific genre. The learners in this approach produce several drafts and receive feedback before submitting their final products (Badger & White, 2000).

The process genre approach
In recent literature in relation to EFL writing, the PGA has been suggested over the product approach in helping students improve their writing skills in many ways (Chow Voon, 2007;Gupitasari, 2013). Therefore, it seemed necessary to employ such an approach into the academic writing course. As its name suggests, the PGA is the synthesis of the process and genre approaches to writing, which allows students to study the relationship between purpose and form for a particular genre as they use the recursive processes of prewriting, drafting, revision, and editing (Badger & White, 2000). By going through these steps, students develop their awareness of different text as well as the process of composing a text. Kim and Kim (2005) argued that following these steps, writing courses will both offer students opportunities to enjoy the creativity of writing and to become independent writers (as in process approaches). This also enables learners to understand the linguistic features of each genre and emphasize the discourse value of the structures they are using (as in genre approaches). Yang (2005) proposes three general guidelines for teachers when using the PGA. First, considering that writing is challenging for many students, the teacher should adopt the role of an assistant and a guide working closely with them to encourage them, to be able to offer constructive feedback and suggestions. It is crucial for teachers to offer positive and helpful advice on what students have written. Teachers can also make efforts to arouse curiosity and self-confidence by matching students interests to the writing topics, and they should be sensitive to any individual differences that arise in the writing process. Second, teachers should explicitly train students about different writing strategies. If teachers demonstrate how prewriting activates the schemata and outline strategies for the drafting and revision processes, then students will be more successful in composing written texts of their own. As Yau (1991) states, writing performance is as much a result of the students' use of strategies in various processes of writing, as it is of their handling of the language. Third, teachers should integrate the main language skills, namely, listening, speaking, and reading in their writing classes. Integrating the four language skills promotes the development of the students' overall language competence (Goodman, 1986). The PGA makes all of this feasible, as background material is read during prewriting activities, and speaking and listening take place during lectures and when giving or receiving feedback.

The blended learning approach
Blended learning is seen to significantly improve the learning experience when implemented appropriately (Pena-Sanchez & Hicks, 2006;Staker & Horn, 2012;Stracke, 2007). According to Darrow (2012) from a teacher's perspective, the BLA is defined as: a pedagogical approach facilitated by a teacher where students have some control over their learning; and the teacher seamlessly incorporates the use of online learning tools (e.g. discussion boards, online collaboration, blogs, etc.), technology tools (computers, digital white boards, cameras, etc.), and face-to-face instruction so that instruction and learning can be accessed at any time by the student through multiple electronic devices. (para. 6) Moreover, from the students' perspective it is argued that blended learning is "a complement to classroom teaching" (Ruthven-Stuart, 2003, p. 170). There are many benefits of employing the BLA in the classroom environment as it provides a more individualized learning experience, provides more personalized learning support, supports and encourages independent and collaborative learning, increases student engagement in learning, accommodates a variety of learning styles, provides a place to practice the target language beyond the classroom, provides a less stressful practice environment for the target language, provides flexible study, anytime or anywhere, to meet learners' needs, helps students develop valuable and necessary twenty-first century learning skills (Marsh, 2012, pp. 4-5).
The experience of blended learning in the language classroom has significantly shifted with the emergence of the Internet which provided foreign language learners with immediate access to the worldwide community of English language speakers, and to authentic resources through its billions of interconnected web pages. The Internet, in particular the emergence of Web 2.0, represents a powerful medium for foreign language teaching and learning (Marsh, 2012). This rapid growth of learning technologies involving the Internet and web-based communication is in today's world inevitable, even though face-to-face foreign language teaching is still at core. Language teachers and learners have now wider opportunities to investigate and explore a suitable mix of teaching and learning styles for given tasks. Sharma and Barrett (2007) put forth three reasons for employing blended learning in the EFL context: (a) Learners' expectations-learners nowadays expect technology to be integrated into their language classes, (b) Flexibility-learners expect to be able to fit learning into their busy lives, especially professional adults and university students, and (c) Ministry of Education (or similar) directives-in some contexts teachers are expected to offer blended learning options. Other researchers argue that blended learning improves pedagogy, increased access/flexibility, increased cost effectiveness, convenience working in your own time and own pace, learner expectations, motivation, autonomy, collaboration, market reach, and experimental learning (Aborisade, 2013;Bo & O'Hare, 2013;Dudeney & Hockley, 2013;Eydelman, 2013).

Aim of the study
In the Turkish Cypriot higher education system, students mainly come from an exam-oriented background and they inevitably focus on improving their accuracy, that is, correct use of grammar, rather than skillful language production. Thus, an elective course that would target the writing skills of preservice English language teachers appeared necessary. The aims of this course were listed as helping preservice EFL teachers understand the parts of an essay as well as different types of essays; generating interest among preservice EFL teachers in writing essays using the PGA; integrating technologies that are available to pre-service EFL teachers on a daily basis in an academic writing course to improve their chances of getting involved in the processes of writing and giving/receiving written corrective feedback from different agents; and improving preservice EFL teachers' language abilities in general and writing skills in particular. An action research was designed to collect data regarding the students' performances in this course and to evaluate the results. The main aim of this article is to find out the effectiveness of implementing a BLA in a writing course in which EFL students are exposed to pen-and-paper and online work, which will both be designed according to the principles of the PGA. The following research question guided this study: (1) How do students' writings differ in pen-and-paper (portfolio) work and online work when both modes of writing follow a PGA?

Research design
An action research design was followed to answer the research questions for this study. In order to understand the effects of blended learning with a PGA to teaching writing, an advanced writing course was designed. The course had two sections; one section was taught using the PGA with pen and paper, that is, in class writing sessions, and the second part was carried out online using the same approach to writing. Data about the participants' performances were collected throughout the course and quantitatively evaluated to measure the individual changes in students' first drafts and their final products for their portfolio work and online work. In addition, data were collected regarding the processes involved, challenges faced and on-going changes made to the course design by keeping a research journal. Finally, structured interviews were also carried out with the participating students. The data presented in this article only concern the quantitative analysis of the effectiveness of the designed course by statistical analysis of students' scores in their first drafts and final products. Figure 1 illustrates the steps followed in collecting the data and administering the treatment.  The designed blended course, which was offered to the students of the Department of ELT as an elective course in their senior year, lasted for 16 weeks in total. This period included the mid-term and final examination weeks. Students were engaged in several writing tasks over a 30-h period, each class hour lasting for 50 minutes, which made up the treatment section of the study. Two types of essays were written both in the pen-paper mode and the online mode of the course and two drafts were written for each essay before the final product was completed. During this drafting period, feedback was provided by assigned peers, the lecturer, or collaboratively. The participants were all trained on how to give feedback to their peers. So, for the first draft the learner's peer gave feedback and for the second drafts the lecturer or collaborative feedback was given. For the collaborative feedback both the lecturer and the learners in the class discussed and pinpointed the errors set on the white board (before the class the written errors were highlighted by the lecturer) and clarifications and solutions were provided together.
For the portfolio work, students were expected to complete a portfolio, that is, an A4-sized case with the work done in class. The tasks completed every week included writing an introductory paragraph, the body, and the conclusion of an essay. These were placed into the portfolio together with the written corrective feedback given, showing step by step the development of a specific essay. Models of the specific essays contributed by the lecturer, transitions and linking words with definitions and clarifying examples, preplans of essays (jotted notes of ideas and tables), activities/ tasks, and readings completed in class and the assessment criteria were also added to the portfolios of the participants.
After the mid-term examination the BLA was adopted, during which the students were asked to complete writing tasks online, send and receive e-mails, and carry out research online. These online tasks involved watching and commenting on videos on YouTube.com, commenting and following task announcements on Facebook.com and reading an article and commenting on it on The Daily Mail's website. During class hours, written corrective feedback given by an appointed peer or the lecturer were discussed and elaborated on collaboratively. In addition to this, spontaneous issues that arouse during any task or work were clarified.

Participants
The participants of the present study were senior ELT students, that is, prospective English language teachers at a private university in north Cyprus. As the course was offered as an elective, 16 students registered for the course. All students accepted to take part in the study. Three participants' first language (L1) was English and 13 of the students learnt EFL. Despite the fact that the students were all in their final years of study, they had varying proficiency levels in English as they were all EFL learners. Taking into account that some students were better than others in English, the terms high and low proficiency levels will be used in this study to separate students' proficiency levels in English. That is to say, these terms will not reflect their actual proficiency levels but will rather be used relative to their class performances. Three of the participants were males and 13 were females. In this study, the participants' original identities are replaced by pseudonyms to ensure anonymity. Consent to carry out the study was received by the institution, head of department and the participants who took part.

Data collection and analysis procedures
Quantitative data were obtained from the essays that the participants wrote during the implementation of the portfolio work and online work mentioned earlier. Even though the students had written more than one essay for each mode of the course, that is, pen-andpaper and online work, only one essay for each mode was taken into account to be analyzed for this study. Each essay consisted of two drafts and a final product. The essay types that were focused on were the persuasive essay, the advantage and disadvantage essay, the argumentative essay, and the compare and contrast essay. The persuasive and the advantage and disadvantage essays were focused on during the pen-and-paper work until the mid-term examination week. Thus, the argumentative and compare and contrast essays were used during the on-line work. One essay from the portfolio and one essay from the on-line work were chosen to be analyzed in order to see individual changes in the first and final essays of the portfolio work and on-line work and the similarities and differences concerning marks given for the essay products of both pen-and-paper and on-line work. During the process of both modes, students were encouraged to produce their best work for the drafts as well as the products. The last essays for each work were chosen as data for this study due to the fact that students were more familiar with the procedures involved in drafting and feedbacking and hence better comparable work in their second essays for each approach. These essays were marked according to essay writing criteria set by the department and the scores of the students were compared to see whether there were any significant differences in their performances both from their drafts to their final products and between pen-and-paper work and online work.
To be able to understand the nature of progress, an error analysis was employed in which students were assessed, given corrective feedback and marks according to five aspects: content, paragraph and essay organization (cohesion/coherence), language (grammar), vocabulary, and mechanical accuracy (punctuation/capitalization) following the writing criteria offered by the institution.
To check the reliability of the marks given to the students for their first and final drafts, interrater reliability was employed. In addition to the lecturer of the course, each essay was scored separately by a second assessor using the same criteria and assigned scores out of 10. Pearson correlation coefficient was calculated to see if the two raters' scores were similar. The results of this analysis suggested a statistically significant and strong relationship between the two sets of results (r = .961, n = 16, p = .000), indicating that the marks assigned by the lecturer could be considered reliable. For this reason, it could be concluded that the assessment of the lecturer for the draft portfolio work was reliable (Kahn, 2010;Ravid, 2011).
The data obtained from the essays of the participants who took part in the writing course were quantitatively analyzed using the Wilcoxon Signed-ranked test to determine if students had any significant change in marks for the essays produced during the pen-and-paper and online work. Students' scores given by the lecturer for the drafts and products of the portfolio and online work were compared using a Wilcoxon Signed-rank test to see whether there were any statistically significant improvements in students' marks.

Findings and discussion
The initial analysis of the scores obtained by the students in the pen-and-paper and online modes showed that there was an observable improvement in the scores of students during online work. Table 1 illustrates these scores.
In order to investigate the effectiveness of the teaching approaches used in this study, the participants work were analyzed by comparing their scores obtained in both modes of teaching (see Table 2). The Wilcoxon Signed-rank test revealed that there is a statistically significant difference between the underlying distributions of the marks in the drafts and the final products of the portfolio work (Z = -3.464, p < .001). In other words, the participants individually showed progress from their drafts to their final products when they worked on their portfolios by writing with pen-and-paper (see Table 2).
Although the statistical analysis and content analysis of errors showed that students have definitely made progress in their writings during the first part of the study, the exact reasons for these improvements cannot be pinpointed through such quantitative analysis. There may be, however, some possible reasons behind such progress related to the way the course was taught. First of all, the peer and collaborative debates and discussions held in class enabled students to make adjustments and modifications to their work. This strategy also enabled students to understand the reasons behind the change. For example, while providing correction to the students' work, instead of directly correcting their errors, students were provided with error correction codes and expected to find the correct version. They were also given reasons and examples concerning the correct versions while they were self-correcting. For each error type, that is, grammar, punctuation, and so on, the lecturer retaught or revised previously learnt knowledge in class. That is to say, students were given extra feedback and tuition after each draft of their essays. Such focused correction seems to work better than general feedback or corrections (Bitchener & Knoch, 2008a). This could be one of the reasons for the apparent improvement in students' essay products. As a result, they were able to produce essays with less errors, resulting in higher marks. A study conducted by Ferris, Liu, Sinha, and Senna (2013), which took into account students' perspectives, also revealed that focused feedback paired with discussion activities has strong potential to be helpful in three ways: (1) It is relevant because it is tied to students' own texts; (2) It is clear because it is specific and because students have opportunities to ask questions and receive explanations; and (3) It is motivating because it actually gives students practical insights about their own writing challenges and knowledge that might help them solve those problems. (p. 323)  The second reason for the improvement in students' final products during pen-and-paper work could be due to the process of the PGA, which was used both in the first and the second part of the course. Students studied the relationship between purpose and form for a particular genre as they used the recursive processes of prewriting, drafting, revision, and editing. By going through these steps students were able to develop their awareness of different text types and at the same time the composing process (Badger & White, 2000). Taking into account Yang's (2005) suggestions, the lecturer adopted the role of an assistant and a guide, who worked closely with the students to encourage them, offering helpful feedback and suggestions. Moreover, the lecturer offered positive and constructive advice with regard to students' essays. Students' interests were also taken into account when dealing with the topics. The topics given as examples to illustrate the organization of the essays were specifically adopted to arouse students' curiosity, self-confidence, and interests. Another suggestion taken into consideration regarded directly training students about writing strategies (Badger & White, 2000). In connection to how prewriting activates the schemata, outline strategies for the drafting and revision processes were demonstrated by the lecturer. What is more, the three other skills, that is, speaking, listening, and reading, were integrated into the course in order to promote the expansion of the student's overall language competence (Banados, 2006;Goodman, 1986). Background materials, such as model essays were read during the prewriting activities, and listening and speaking occurred during brainstorming, stating opinions, classroom discussions, and during feedback sessions. To sum up, the approach employed during the portfolio work could be one of the main reasons why students showed individual progress from their drafts to their final products of their essays.
The help of additional materials might also have had an impact on students' improvement. During the pen-and-paper mode, students were distributed extra materials illustrating step by step how an essay is organized with supporting tasks which students had to complete. Students followed all the steps, which led them to writing their own advantage and disadvantage essay. These materials gave students the opportunity to gradually produce an essay dealing with chunks of the essay rather than the whole essay at once. In addition, the supporting tasks enabled students to practise the theory of essay writing.
With regard to the improvement of students' work in the online work, the results in Table 3 show that there is a statistically significant difference between the underlying distributions of the marks in the drafts and the final products of the online work (Z = -3.585, p < .000) as well. In other words, the participants individually showed progress from their drafts to their final products when they produced essays online.
The apparent improvement in the students' products in the online work can be attributed to a number of reasons. Since the same approach that was utilized when administering the pen-andpaper part of the study, that is, PGA, was employed during the blended mode as well, it could be argued that drafts, in other words the revision stage, helped them improve their accuracies in the final products (Chandler, 2003;Krashen, 2004). In addition, the participants had the opportunity to benefit from both direct and indirect corrective feedback types (Bitchener & Knoch, 2008b;Myles,   . One significant aspect of this study is that additional support and corrective feedback during both the pen-and-paper and online work were provided. During the implementation of the first phase, this additional support and corrective feedback was orally provided during classroom hours, both individually and collaboratively with peers. During the employment of the BLA, written feedback was provided online via e-mail and Facebook.com by private messaging and oral feedback was also given in office hours. Therefore, during online work, not only were the participants engaged in self, peer, and collaborative assessment involving both direct and indirect corrective feedback during classroom sessions, but they also had oral and written feedback and support. This study, therefore, highlights the impact of extra support and written feedback in writing classes, which seems to have a significant effect on students' writing development. Moreover, students were flexible to complete their tasks at any time or any place they wanted and would at that present moment ask and receive information from the lecturer. Baring in mind that the internet is now available on smart phones, students were able to send an instant message via Facebook.com private messaging or e-mail their lecturer throughout the process of the BLA. Thus, being able to contact and receive immediate responses from their lecturer outside of the classroom when they needed assistance was a significant advantage. The level of the language proficiency of the students also had an impact on self-correction (Blanche, 1988;Brown & Hudson, 2002;Davidson & Henning, 1985;Heilenmann, 1990;Janssen-Van Dieten, 1989). As previously stated, the participating students were senior ELT students who had enough language proficiency to be able to check their own work. Students were able to selfcorrect during both approaches but the BLA helped and fostered students' self-correction with the features that were presented in the Microsoft Word document, that is, grammar check, thesaurus, spelling, track changes, and so on. The self-correction technique employed in the BLA made a considerable contribution to the correction of students' errors.
The availability of materials outside the class contact hours appeared to be a very significant element, positively impacting the participants' performances. Issues such as missing classes and not being able to contact any of their classmates or the lecturer before the due date of assignments, did not cause any deductions in their marks as they were able to complete tasks online on their own pace. During the first phase of the course, for whatever reason, when students missed classes, their only opportunity was to see the lecturer in office hours, which again was sometimes difficult as they had other lessons to attend, clashing with the lecturer's office hours. Such problems did not appear in the second part of the semester when the online work was done.
Further analysis was carried out to investigate whether the difference in students' progress was greater in the portfolio work or in the online work to determine which of the approaches was the more effective one in this case. For this purpose, the final products of the two approaches were compared using the Wilcoxon Signed-rank test. This analysis showed that students' writings showed more progress in the online work compared to the pen-and-paper portfolio work.
The results in Table 4 indicate that the participants individually showed significantly more progress in the online work compared to the pen-and-paper work (Z = -3.307, p < .001). In other words, the employment of the BLA in the second part of the course appears to have had more of an impact on students' essays than the portfolio work in terms of improving their errors. There may be a couple of reasons for this. During the use of the BLA, students did not need as much clarification as the portfolio work probably because this approach was being employed after the pen-and-paper work, where some of their errors were already corrected. Another possible reason for less number of errors may be because they had the chance to check their work via Microsoft Word in the blended phase. For these reasons, it could be said that the electronic work integrated into the BLA encouraged self-correction and hence autonomous learning. Related literature has also put forth that blended learning fosters autonomous learning (Eydelman, 2013;Marsh, 2012). Being engaged in online work, where students prepare their essays using a Word processor, unfortunately showed its drawbacks in the students' essays, where students produced typing errors. Students became reliant on the proof-reader within the Word processor and did not check their work properly. During the feedback sessions, students' explained that the errors they had made were due to hastening and/or not revising their essays before submission.
In addition to students being more motivated by mere inclusion of technology in their classes, the fact that students are continuously engaged in the online community on an everyday basis may be another reason for their increased motivation, and hence better performance. As Allan (2007) states, the BLA "offers the opportunity to combine the best of a number of worlds" (p. 8). In other words, students are able to see and meet different cultures through social networking sites, be engaged in different tasks online, use different tools presented on a computer, be engaged in peer and collaborative work both online and in class, assess peers and self-assess online, analyze and comment on a "real" essay and/or article, and at the same time do all of these in their own time (and space) within their daily routines. Thus, the BLA involves an effective combination of different modes of delivery, models of teaching, and styles of learning (Procter, 2003). In the current study, the participants were engaged in several tasks that required them to use websites such as YouTube and Facebook, which are part of their everyday lives. Due to this fact, the employment of such an approach enabled students to see academic work to be relevant to what they were doing every day, hence reducing classroom related anxieties and increasing performance.
Finally, during online work, students were still engaged in face-to-face classroom discussions and feedback. Similar to the pen-and-paper phase, discussions and debates elaborating on topics and students' opinions about a certain topic were still in practice. The model texts sent via e-mail to the lecturer were also discussed with regard to the organization, usage of transitions, conjunctions, and language, during class hours. So, students were engaged in almost the same procedure dealt with during the portfolio work supported by online work. This may be another reason why students had performed better during online work compared to the pen-and-paper work.

Conclusion
It is argued that the inclusion of technology in writing classes has a significant impact on students' development in writing in general. As put forth by Pena-Sanchez and Hicks (2006), Stracke (2005) and Stracke (2007), integrating face-to-face sessions with technology has the potential to improve learning significantly. Research drawing on learners' attitudes toward the employment of the BLA revealed that students preferred BLA to face-to-face classroom sessions initially due to the BLA being more motivating (Brett, 1996;Leakey & Ranchoux, 2006;Lin, 2003). Therefore, one of the possible reasons behind the significant improvement in students' performances in online work compared to pen-and-paper work could be motivational differences when being engaged in these modes.
The findings of this study appear to support the existing literature on the effectiveness of employing BLA in teaching in general (Aborisade, 2013;Bilgin, 2013;Marsh, 2012) and teaching writing in particular (Eydelman, 2013). Students appeared to have adopted some principles of autonomous learning and made less number of errors when working online. Despite the fact that the results revealed that the students showed progress in both pen-and-paper work and online work, the specific contribution of this study to the field is that the online work was designed using principles of PGA, that is, many drafts and feedbacking sessions were held during online work as well. Thus, this study shows that when administering online/blended courses for EFL learners, using a PGA is beneficial. It also revealed that additional feedback sessions, support and in-class discussions and debates during online work enabled students to have the freedom of producing essays in their own pace and time. With the help of the features presented in the Microsoft Word, students were able to produce significantly better products.
Based on the results of the current study, it is recommended that a deeper analysis of written errors in a writing course designed with the BLA could be carried out to see the impact of this approach on self-correction. Lecturers, teachers, instructors, and teacher trainers will have the opportunity to benefit from such a research as errors are an inevitable part of the 212 learning process (Krashen, 1987) and their treatment as well as the ways they are treated are crucial. In addition, an analysis of the perspectives of the students who participate in such a writing course can help course designers to understand the benefits and difficulties experienced by the learners. Such a study could also reveal perspectives in relation to writing using the PGA. Finally, this study can be replicated with students from different departments in English for Specific Purposes (ESP) settings to see if the results will be different when students from non-English related departments are taught using the BLA. These findings will contribute to the field of EFL and ESP in terms of teaching and learning writing.