1 Introduction

Task-based language teaching (TBLT), which emerged in the late 1970s as a result of Prabhu’s (1987) research, focuses on using authentic, task-based activities to engage learners in communicative interaction. Central to TBLT is the negotiation of meaning in real-life situations that tap into learners’ natural mechanisms for foreign language acquisition (Skehan 1998, Skehan 2003). The theoretical basis for using TBLT is that language is best learned through interaction (Pica et al. 1993; Samuda and Bygate 2008). As Skehan (1998) stated, “the task-based approach sees language acquisition as a process of learning through doing. Learners develop their interlanguage by attending to form in the context of meaning-focused activities” (p. 4). Tasks should be structured in a way that asks learners to use the language, “not for the sake of producing the language as an end itself, but as a means of sharing ideas and opinions, collaborating toward a single goal, or competing to achieve individual goals” (Pica et al. 1993: 10).

TBLT, depending on how the tasks are designed and processed, has been proven successful in improving the accuracy, complexity, and fluency of learners’ oral output (Robinson 2001), sharpening learners’ vocabulary building skills (Chacón 2012), enhancing learners’ ability to transfer what they learn in the classroom to the real world (Macías 2004), helping build a positive learning environment by engaging learners in completing tasks (McDonough and Chaikitmongkol 2007), boosting learners’ confidence (Lopes 2004), and increasing learners’ motivation (Park 2012).

Notwithstanding the popularity of TBLT in second language acquisition and the benefits that it provides, there are issues that sometimes prevent TBLT from being adopted for language learning. One issue is that some learners believe in the importance of grammar learning and prefer being taught using explicit and systematic grammar instruction (Bao and Du 2015; Carless 2007; Lai et al. 2011). A task-based course built entirely on TBLT usually results in learner dissatisfaction due to a lack of primary focus on explicit grammar instruction (Lai et al. 2011; Lopes 2004; McDonough and Chaikitmongkol 2007). Another significant issue is the lack of time for developing materials on the part of the teacher and the lack of time built into the syllabus for integrating the TBLT approach in teaching (Bao and Kirkebæk 2013; Carless 2003; McDonough and Chaikitmongkol 2007). If TBLT was not designed as part of the syllabus during the planning stages of the course, teachers generally had little time to design tasks as they taught specific class periods or units. Therefore, when teachers tried to incorporate TBLT elements, as Bao and Du (2015: 294) mentioned, the approach “left little time for task completion or for sufficient explanation on questions generated from tasks”. Researchers have stressed the importance of “road-testing” TBLT; that is, there is need for more research in different social contexts and different classroom settings in order to shed light on the implementation of TBLT (Bygate 2015; Carless 2007; Klapper 2003; Ortega 2007; Seedhouse 1999; Van den Branden 2006).

As a further complication to this project, despite a growing interest in learning Chinese in America and the other parts of the world, teaching and learning Chinese poses different challenges than teaching and learning some other languages, including the problem of how to effectively teach with ample consideration of context (Du and Kirkebæk 2012). Planning for teaching is very similar to designing, in that people who participate in either of these processes need to take into consideration the contexts in which they work. As Ralph and Wand (2009: 109) stated, “design is a specification of an object, manifested by an agent, intended to accomplish goals, in a particular environment, using a set of primitive components, satisfying a set of requirements, subject to constraints”. Or, as American painter Kenneth Noland once said, “Context is the key from which comes the understanding of everything.”

Given the above issues, it is critical to think about how to implement TBLT at the syllabus level, taking into consideration specific teaching contexts, such as how to cater to students’ beliefs about the value of explicit grammar instruction. It is also important to consider how to use TBLT to enhance students’ communicative skills and use of language, especially in the social context where more and more Chinese students are coming to America to study. Chinese as a foreign language (CFL) students can take advantage of this context in their learning and the interaction between American students and Chinese students can enhance cultural understanding as well. Additionally, advances in technology and Web 2.0 tools, such as wikis, blogs, social media sites, and so on, have been applied widely in language learning, and the interactive features of these tools can help achieve the intended purpose of TBLT.

This study investigates the effectiveness of a wiki-enhanced design, development, and implementation of TBLT at the syllabus level to support low-intermediate CFL learners. Results of this study are expected to broaden the empirical basis for the use of TBLT and shed light on teaching CFL with TBLT in context.

2 Literature review and background

Substantial research on TBLT has been done in various contexts in the field of foreign language education (Lai and Li 2011). However, there are only a few empirical studies investigating the use of TBLT in the context of CFL. Bao and Du (2015) completed an exploratory study on the use of TBLT in teaching beginning Chinese learners. The results show that TBLT benefits learners in terms of increasing participation, easing learner anxiety, and creating speaking opportunities. The challenges arising from the use of TBLT, such as the lack of pronunciation exercises, the lack of differentiating learner differences in learning styles, and the lack of instructional support, show that the use of TBLT is not a one-size-fits-all method, and that those using TBLT should take various contextual factors into consideration when designing a course using TBLT.

Lai et al. (2011) examined how TBLT works in an online beginning Chinese course for high school students, and Lin et al. (2014) completed a qualitative study investigating whether TBLT works in a virtual language learning (Second Life) environment. Both studies explored the use of TBLT in the context of online/virtual learning environments and found positive perceptions from the learners when using TBLT to support Chinese language learners.

Bao (2012) did a study in an after school program with lower secondary school (6th grade to 10th grade) learners in Denmark. The result shows that the use of TBLT increases learners’ participation and motivates them to learn Chinese in a face-to-face learning environment. Kirkebæk (2012) discovered in his study that the use of TBLT helps high-school learners learn Chinese characters in school. Du and Kirkebæk (2012) argue that in the implementation of TBLT, both teachers and learners should have agreed-upon learning goals and the same understanding of the method.

In addition to the above-mentioned empirical studies, several CFL teachers have conducted action research on the implementation of TBLT. For example, Lu (2011) examined her teaching experience with TBLT and drew conclusions about strategies for task design and task implementation. In designing a task, teachers should bear in mind that (1) topic choice should not overload the learner, (2) comprehensible input should be available via both auditory and visual channels, and (3) the level of difficulty should be low when adding unfamiliar linguistic forms. In implementing tasks, teachers should take into consideration both learners’ anxiety and learners’ individual differences.

The research on the use of TBLT in the field of CFL has shown benefits of TBLT, but more studies are needed to find out how TBLT could be designed and implemented at the syllabus level while taking into consideration different learning contexts. Furthermore, research needs to be done to explore the use of TBLT at the syllabus level in supporting CFL learners in a face-to-face learning environment.

3 Theoretical framework and the design of the TBLT wiki-fortified syllabus

This study is built upon the Educational Engineering Research Theory and its Distributed Design Approach (Colpaert 2010, Colpaert 2016a, Colpaert 2016b) and follows the guiding principles in selecting, grading, and sequencing tasks proposed by Nunan (1993). According to the Educational Engineering theory, neither technology nor pedagogy nor content should be the starting point for designing a learning environment. Analyzing real-world situations in a systematic and verifiable way should be the first step in design, and the design should take into account the specificity of the context. The ultimate aim of design is to build the best possible (optimal) educational artifacts including “documents, tools, content, concepts, models, and solutions such as textbooks, syllabi, lesson plans, curricula, graded readers, exercises, tests, applications or electronic learning platforms” (Colpaert 2016a: 5). Distributed Design refers to the idea that the design process should take into account many actors and factors in order to create the best educational artifacts. In considering the current situation in teaching Chinese, all the different contextual factors, such as the lack of time for using TBLT, students’ belief that they need solid coverage of grammar, the affordances that wikis could provide, and an understanding of who the learners are all significant factors that should play an important role in the process of implementing a TBLT approach to course design at the syllabus level.

The Distributed Design approach is based on the ADDIE instructional design model, and it has five phases: Analyze, Design, Develop, Implement, and Evaluate. It is an artifact-development concept applied in the environment of learning for the construction of performance-based learning (Branch 2009). According to this model, instruction is considered a potential intervention and should be used “only when a lack of knowledge and skills has been validated as the main cause for a performance gap” (Branch 2009: 20). The five-phase Distributed Design approach is illustrated in Figure 1. Following the figure, a detailed description of the five phases in the context of this research project is provided.

Figure 1:

The distributed design approach based on the ADDIE instructional design model (Colpaert 2016b).

3.2 Design

The purpose of the “design” phase is to verify the desired performances and the appropriate testing methods for those performances. This phase includes three sub-phases: conceptualization, specification, and prototyping (Colpaert 2016b). Conceptualization helps with deciding on the content, pedagogy, and technology. Based on the results of the needs analysis, the instructional goal was set as improving learners’ language proficiency, especially their speaking ability and speaking skills. To achieve this goal, two points needed to be considered in the design of the syllabus. One was that interactive and collaborative learning should happen not only in the classroom but should also happen beyond the classroom after the face-to-face meeting time is over. The other was that the content should include not only the content delineated in the textbook and mentioned in the classroom but should also include learning content built during the interactions between the learners and the teacher and between the learners and native speakers in the real world. These interactions would take place within the confines of completing certain language learning tasks, as guided by the instructor.

Based on these considerations and the instructional goals, a set of textbooks that are appropriate for the skill level of the learners should be chosen, a pedagogical method that could enhance CFL learners’ overall language proficiency level, especially their speaking ability and skills, should be identified, and, if possible, technologies that could help the learners interact with each other outside the classroom should be utilized. Figure 2 shows the conceptualization.

Figure 2:

The conceptualization architecture (built upon Colpaert 2016b).

Next, the specification sub-phase offers a detailed description of what needs to be changed or created. First, the textbook was chosen: Integrated Chinese (Level 2 Part 1, by Liu, Yao, Bi, Shi and Ge 2010), which was determined to have the right level of difficulty and follow the right sequence of grammar to provide a smooth transition from the Integrated Chinese (Level 1 Part 2, by Liu et al. 2010) text that most of the students used in the previous semester. TBLT was chosen as the pedagogical approach because, when used appropriately, it has been shown to be beneficial in improving foreign language learners’ language proficiency in general and their speaking ability in particular. Due to the contextual factors previously mentioned, the TBLT approach needed to be altered somewhat before being applied in this situation. In this case, it was important that learning not be limited to the classroom. The most important learning needed to happen both in the classroom and after the class meeting was over. In order to encourage this learning outside the classroom, it was important to make the learning tasks engaging and to provide opportunities for students to get feedback from their peers and their instructors. Wikis, introduced in the mid-1990s (Wei et al. 2005), are dynamic, web-based environments where readers can be also authors and editors, and which allow for “distributed participation and collaboration (Knobel and Lankshear 2006: 81). Wikis have been shown to enhance collaborative learning in the field of language education (Díez-Bedmar and Pérez-Paredes 2012; Kessler 2009; Liu and Lee 2011; Wang et al. 2013). Through collaboration, users can create knowledge and engage learners to become part of an active learning community (Alm 2006; Farabaugh 2007). A wiki format was selected to facilitate interaction between learners outside the classroom while providing an engaging atmosphere that facilitated feedback.

At the prototyping sub-phase, the syllabus was designed taking into account the above-mentioned considerations. There were eight units in total during the semester. For each unit, the class meeting time focused on two primary goals using the communicative approach: students learning the language forms and students interacting with each other under the guidance of the teacher. Outside of the classroom, the learners had opportunities to work on their own, to receive feedback from the teacher and their co-learners, and to expand their learning by interacting with native speakers. Tasks within each unit were designed to help learning happen more effectively, with one major task assigned within each unit. Task selection and sequencing are discussed below.

3.2.1 Task-selection

The major tasks were designed and selected based on Ellis’s definition (2003, 2009) of tasks and the criteria for tasks as defined by Skehan (1998). According to Ellis (2009), a task qualifies as a task if (a) the primary focus is on meaning; that is to say, learners are mainly concerned with processing the semantic and pragmatic meaning of what they hear and say, (b) there is some kind of gap – In other words, there should be “a need to convey information, to express an opinion or to infer meaning” (Ellis 2009: 223), (c) learners rely on their own resources (linguistic and non-linguistic) in completing the task, and (d) there is a clearly defined outcome. The outcome should not be the use of language; that is to say, the language should only serve as the means of achieving the outcome, but not as an end. According to Skehan (1998), in addition to the four features of tasks, there should be a real-world relationship in a task. As Widdowson (1998) argues, the distinction between a task and an exercise is not in the interior, but in their meaning, goal, and outcome.

In selecting tasks, focused tasks and unfocused tasks were identified. Nunan (2004: 94) defined a focused task as “one in which a particular structure is required in order for a task to be completed” and an unfocused task as one in which no predigesting of grammatical resources is required. Willis and Willis (2007) made a distinction between a focus on language and a focus on form, which has become the basic guiding principle for the integration of linguistic forms into communicative tasks. A focus on language refers to a situation in which learners pause for a short moment in a meaning-focused activity in order to think and work out how best to express what they want to say. A focus on form refers to the situation in which “one or more lexical or grammatical forms are isolated and specified for study, or in which the teacher comments on student language by drawing attention to problems” (Willis and Willis 2007: 5). Some tasks that focused on form were designed for the pre-task phase in the classroom setting. While giving feedback on the wiki-essays, the instructor/researcher also focused on both form and language. However, by selecting unfocused tasks to be used as the major/core tasks when constructing the syllabus, the instructor/research put a focus on language.

3.2.2 Sequencing tasks

Willis and Willis (2007)’s order of designing a task was adopted in this study: pre-task, core-task, and post-task. Nunan (2004: 31) specified the order of pre-task activities: “schema building, controlled practice, authentic listening practice, focusing on linguistic elements, providing freer practice, and introducing the pedagogical task”. Nunan pointed out that the sequence should start with communicative ends but not linguistic means; however, a focus on form should also exist in the pre-task activities. During the pre-task stage, the classroom activities focused on form by giving comprehensible input (Krashen 1987, Krashen 1988) and by following a process of guiding students through recognition practice and then production practice with meaning. Learners familiarized themselves with the use of the target language, and established connections between linguistic forms and communicative functions.

In the pre-task stage, Willis and Willis (2007) suggested that teachers warm up learners with a brief introduction to the topic and then provide learners with a vocabulary exercise in preparation for the next stage of the task. Furthermore, Nunan (2004) pointed out that the key items in the task need to be prepared during the pre-task phase. He suggested that other opportunities for language-focused work be created, such as giving learners planning time to prepare for the presentation of their work and asking learners to present their ideas through writing. Based on the guiding principles proposed by Nunan (2004) and Willis and Willis (2007), the same task cycle was used within each unit.

The task cycle involved practicing grammar, vocabulary, and four different language skills. This practice took place in different formats, primarily in the format of tasks; for example, pair or group tasks to perform different scenarios in class and individual tasks to read authentic texts outside class. These tasks not only satisfied the learners’ desire to have solid grammar coverage but also to helped with the core task presentation, which utilized the learner’s speaking skills and their other skills (such as their understanding of grammar use and vocabulary) in a meaningful context.

In this study, the core task – the in-class presentation – was designed as a summarization/summative task. To reach that point, the learners needed to carefully work on two other core tasks. One was to interact with native speakers and with peers, and the other was to write an essay, incorporating the co-learners’ and the teacher’s feedback in revising the essay and in making their in-class presentation.

Wikis in Blackboard were chosen to help with the latter task in the TBLT approach. The wikis allowed students in the class to create a page, then write and edit their essays quickly. They could modify their peers’ essays using track changes and add their comments. Wikis provided a means of effective sharing and collaboration between multiple writers, providing affordances not available in a blog format. Using wikis in Blackboard helped create a secure and comfortable environment where the sense of community, the collaboration, and the interaction between learners, co-learners, and the teacher could be enhanced outside of the classroom. Wikis also helped build partnerships where students could benefit from each other’s strengths and provide support to their peers. Finally, wikis provided a single location where all the essays could be searched and accessed quickly and easily. See Appendix A for a sample core task description. See Figure 3 for the flowchart of a one-unit task cycle.

3.2.3 Verifying testing method and rationale for testing method selection and design

Another important procedure that needs to be accomplished in the design phase is verifying the appropriate testing method. WebCAPE (Computer Adaptive Placement Exam) and two speaking assessments were administered both at the beginning of the semester and the end of the semester. WebCAPE, a computerized proficiency exam, evaluates the learners to see if their reading ability, their mastery of grammar, and their understanding of vocabulary has improved. The two speaking assessments were used in order to find out if the implementation of the TBLT syllabus has any impact on students’ speaking proficiency. The first speaking assessment was a narration assessment. The students told a story based on a cartoon. The second speaking assessment was a problem solving assessment. A scenario was given in which a guest lost his/her clothes while taking a shower and had to call the manager to solve the problem, and two students were paired together to finish the problem-solving task.

The first assessment method, narration based on pictorial cue, and the second assessment method, problem-solving conversation, were chosen based on the general task framework summarized by Ellis (2003). In the framework, Ellis pointed out that tasks could be designed based on two features: “input, i.e. the nature of the input provided in the task, and processes, i.e. the nature of the cognitive operations and the discourse the task requires” out of others (Ellis 2003: 218). If it was designed by input, the key dimensions for consideration are medium – i.e. pictorial, oral, or written — and organization – i.e. tight structure or loose structure. If the task is to be designed by processes, the key dimensions for consideration are cognitive – i.e. exchanging information, exchanging opinion, or explaining/reasoning – and discourse mode – i.e. monologic or dialogic. Using this task framework assured that the design of the assessment took into account all the considerations so the assessments were comprehensive. See Table 1.

Table 1:

The features of the two assessments.

	Input					Processes
	Medium			Organization		Cognitive		Discourse Mode
	Pictorial	Oral	Written	Tight structure	Loose structure	Exchanging info	Explaining/Reasoning	Monologic	Dialogic
Narration	x			x		x		x
Problem-solving conversation			x		x		x		x

4 Research questions

The purpose of this study is to answer the following research questions:

1. Does using a wiki-enhanced, TBLT approach contextually designed at the syllabus level have any impact on low-intermediate CFL learners’ Chinese language proficiency?

2. Does using this approach have any impact on low-intermediate CFL learners’ speaking skills, especially their accuracy, fluency, and speed?

5 Methodology

This study adopted a quasi-experimental design to assess the impact of using the wiki-enhanced TBLT syllabus in teaching low-intermediate CFL learners. Two classes were randomly assigned as the control and experimental groups. The students in the experimental group were taught using the wiki-enhanced TBLT syllabus throughout the semester.

6 Results

The results show that the implementation of this approach significantly improved learners’ Chinese language proficiency and their fluency in speaking in narration and problem solving conversations. Their speaking speed and accuracy also significantly improved in narration but not in problem solving conversation.

A two-factor analysis of variance with replication yielded a main effect for the use of the wiki-enhanced TBLT approach, F (1, 32) = 4.16, p < 0.05, such that the experimental group improved significantly in the WebCAPE proficiency assessment over the control group. See Figure 4.

Figure 3:

The flowchart of one-unit task cycle.

Figure 4:

A comparison line of pre- and post-proficiency score between the two groups.

A two-factor analysis of variance with replication was done on the narration fluency, accuracy, and speed data. The results suggest that the implementation of the wiki-enhanced TBLT approach at the syllabus level significantly improved the students’ speaking accuracy in the experimental group, F(1,36) = 12.35, p < 0.01. The results also suggest that the implementation of the approach also significantly improved the students’ speaking fluency in the narration speaking assessment, F(1,36) = 4.02, p < 0.10. The students’ speaking speed also became significantly better after they received instruction using the TBLT approach, F(1,36) = 3.43, p < 0.10. See Figure 5 for the comparisons of means.

Figure 5:

Comparison of means: Narration accuracy, fluency, and speed at the beginning of the semester and at the end of the semester in two groups.

The two-factor analysis of variance with replication done on the problem-solving conversation speaking data shows results that were different from the narration speaking performance. The results show that there was no significant effect of instruction with the TBLT syllabus on the learners’ speaking accuracy and speed during their conversational problem-solving speaking performance, with F (1,24) = 0.70, p = 0.41 and F(1,28) = 0.95, p = 0.33.

These two sets of scores on problem-solving speaking accuracy and speed were further examined. It was found that the standard deviation of the gain scores in the experimental group was noticeably larger when compared with that of the control group. See Figures 6 and 7. The larger the variances are, the more difficult it is to detect any significance differences, despite the mean differences that exist between the two groups.

Figure 6:

A comparison of the gain score variances between the two groups: Problem-solving accuracy.

Figure 7:

A comparison of the gain score variances between the two groups: Problem-solving speed.

However, the results show that there was a significant effect of instruction with the TBLT syllabus implemented on students’ fluency, F(1,20) = 14.94, p < 0.01. See Figure 6 for a comparison between the two groups. See Figure 8 for a comparison of pre- and post-fluency between the two groups.

Figure 8:

A comparison line of pre- and post-fluency between two groups: Problem-solving speaking assessment.

7 Discussion

The findings of this study suggest that the implementation of the wiki-enhanced TBLT approach contextually designed at the syllabus level significantly improved low-intermediate CFL learners’ language proficiency, fluency, accuracy, and speed in narrating a story. This finding generally confirms the results of previous studies (Lai et al. 2011; Robinson 2001, Wang et al. 2013). However, in this case the implementation of the TBLT-designed syllabus only improved the learners’ fluency, not their accuracy and speed in their spoken Chinese in conversational problem-solving situations. Due to the large standard deviation of the gain score difference in the experimental group, results are not significant despite the existence of a mean difference in the gain scores between the two groups. There are two possible reasons to explain this result. First, the implementation of the tech-enhanced TBLT at the syllabus level may have only been effective for some students in improving their problem-solving speaking speed and accuracy, but not for others in the experimental group.

Second, the large individual differences in students’ gain scores might be due to the way that the students were paired up when completing the conversational problem-solving assessment. They were paired up at the beginning of the semester based on their speaking skills and their Chinese language proficiency, that is, two students of the same ability and skills were paired up together to finish the speaking assessment at the beginning of the semester. However, at the end of the semester, some students were absent; therefore, some pairs were not matched in terms of their ability and skills. Either the change of partners in the pairs or the mismatched ability and skills in the pair or both might have been confounding variables. Accuracy and speed are two traits that could be easily influenced by a conversational partner, while fluency might be a trait that is less likely to be influenced by a conversational partner. Another possible reason for the large individual differences in the gain scores could be the influences from other factors, such as participants’ perception of the problem and their reactions to their partner’s opinions during the conversation. These influences could have compromised the improvement of their speaking speed in a dialogic situation.

Regardless of the reasons, the large individual differences in their gain scores combined with a small sample size prevented us from obtaining a conclusive result on whether the implementation of this approach has improved students’ problem-solving speaking speed and accuracy.

In responding to the call that more research should be done to “road-test” TBLT in different social contexts and different learning environments (Carless 2007; Klapper 2003; Du and Kirkebaek 2012; Ortega 2007; Seedhouse 1999; Van den Branden 2006), this study takes into consideration the possible problems that exist in the real world, such as students’ persistent belief in the usefulness of explicit grammar instruction (Bao and Du 2015; Carless 2007; Lai et al. 2011) and the lack of time allowed by a standard syllabus for integrating a TBLT approach in teaching (Lai et al. 2011; Lopes 2004; McDonough and Chaikitmongkol 2007). Therefore, this approach integrated TBLT at the syllabus level, utilized a Web 2.0 tool, a wiki, to extend students’ interactions and learning outside classroom, and made explicit grammar learning a component of the pre-task for each unit. This design was shown to be effective in improving students’ language proficiency and speaking skills (at least in the narrative speaking).

The pre-task activities, designed based on Nunan (2004)’s suggestions for building up pre-task activities, seem to work well in preparing learners to tackle core tasks. According to Nunan (2004), learners should familiarize themselves with the use of the target language at the beginning stage of learning. Therefore, the pre-task activities were designed to build learners’ familiarity with the unit topic by exposing them to it through different input channels and by using different exercises and activities. The teacher speaking Chinese in class, students being required to watch the authentic videos and to listen the supplemental listening exercises, and students completing the reading exercises after class all provided learners with many opportunities to be exposed to the language relevant to the unit topic through different channels, both visual and auditory. The warming up before a unit started helped the learners build schema about the topic and provided the necessary structure and vocabulary that were needed. The controlled spoken practice – pair and group communicative activities – familiarized the learners with the vocabulary and the structures within communicative contexts. Having these communicative activities happen in class took advantage of the face-to-face meeting time to help the learners build discrete skills that were crucial to successfully completing the core tasks. This conforms to Lu (2011)’s suggestion that student learning styles should be considered in course design and comprehensible input should be available via both auditory and visual channels.

The two core tasks, wiki-essay writing/commenting and in-class presentations, were the most influential and powerful activities in helping the learners learn. The in-class presentations were the culmination of all the preparation in the pre-task phase, giving the learners the chance to express their personalities, to convey their real thoughts, and to summarize what they learned through interacting with their co-learners in class and outside class. The presentations also gave them a chance to be creative, through which they found the fun of learning the language and got to know what strategies worked best for their own learning. Presenting in front of a large audience developed the students’ general presentation skills, gave them a real purpose for developing their speaking skills, and provided a real audience for the presentation. Because of these circumstances, students put more energy and effort into getting the presentation ready (Chen and Brown 2012). In addition, the in-class presentation required learners to interact with native speakers outside of class with a purpose and with meaning, which in turn motivated them to actively seek out native speakers with whom they could converse.

Creating two co-existing core tasks (wiki-essay writing/commenting and in-class presentations) that were on the same topic also worked well in helping students focus their time and energy on producing quality work for both tasks. Because students produced similar content twice – once in a written format and once in a spoken format – and were provided with substantive feedback from the instructor and the co-learners, they were able to better retain vocabulary and develop a deeper understanding of how language should be used.

The unit topics for the core tasks were purposefully chosen to be closely related to students’ real life. Additionally, the topics gave the students an opportunity to interact with native speakers. Being familiar with the topics made the tasks less challenging and made it more likely that the students would invest extra energy into being creative. This confirms Lu (2011)’s suggestion that the task topic for a TBLT approach should not overload the learner, especially when unfamiliar linguistic forms are used in the tasks. Unfamiliar topics might not engage the students, and most importantly, students would be too busy familiarizing themselves with the topic to have energy left for creation.

The use of a wiki in this study, as Nunan (2004) stated, expanded the learners’ learning beyond the classroom. It allowed the learners to get continuous feedback from the instructor and from their peers, or co-learners, when they were not in the classroom. Furthermore, it gave the learners an opportunity to read their peers’ work, giving them another opportunity to practice their reading and an opportunity to get to know their peers better. As a result, learners were exposed to the language in a learning environment even after class was over.

8 Conclusion

The findings show that the use of a syllabus that incorporated wikis and the TBLT approach based on the learning context proved to be effective in improving low-intermediate CFL learners’ language proficiency and the speed, accuracy and fluency of their narrative speaking, as well as the fluency of the learners’ conversational problem-solving speaking. The outcome of the implementation was facilitated not just by one or two phases of the TBLT approach but by the implementation of the approach as a whole. Particularly important was the implementation of the approach at the syllabus level, which gave students multiple opportunities to prepare for the core-tasks, to complete the core tasks, and to reflect upon their learning. No single phase of the TBLT approach would work independently and application to just one class or one unit may not result the outcomes that were achieved in this study.

TBLT has been investigated intensely, but few studies have been done to investigate how to integrate technology in task-based learning at the syllabus level for CFL. This study shows that it is beneficial to take into consideration the context under the Educational Engineering framework proposed by the Colpaert (2016a, 2016b) and to create opportunities for students to interact with their co-learners inside and outside of classroom. One of the ways to have this interaction is to use Web 2.0 technology. That said, this study also has limitations.

First, this study did not examine whether the pronunciation and tones were significantly improved in the experimental group after implementation of the syllabus. While Bao and Du (2015) found in their exploratory study that TBLT was perceived as not being able to help with improvement of pronunciation, further research needs to be done to find out if the TBLT approach integrated at the syllabus level could help with the tone and pronunciation.

Second, this study is limited to a small sample of students. The sample was also relatively homogeneous, with mostly European-American college students. There were only two heritage learners in the experimental group and two in the control group. Therefore, the results might not generalize to the student populations that have greater ethnic diversity, such as many heritage learners and African-American learners or with many students from different countries. Despite the limitations, this study adds to the field knowledge about designing a TBLT approach at the syllabus level using wikis along with other strategies in helping CFL learners improve their language skills and proficiency, hone their learning strategies, and motivate their learning.