Improving Speaking Fluency in a Task-Based Language Teaching Approach: The Case of EFL Learners at PUNIV-Cazenga

Speaking Fluency

The term “speaking fluency” is linked to the meaning of “communication” (Harmer, 2007, p. 142). For example, in a conversation, a learner can make a grammatical error, such as Maria live in Cazenga [live vs. lives]), but the learner can still speak the sentence with some fluency (Crowther et al., 2015). The learner can speak without searching for words, so that his or her speech is quickly understood. In fact, speaking fluency has been defined as the “automaticity and speed of speech production” (Brand & Götz, 2011, p. 256). However, automaticity and speed of speech production may not always make a speech comprehensible, comprehensibility being “a measure of listeners’ perceived ease or difficulty of understanding L2 speech” (Crowther et al., 2015, p. 81).

Although a grammatical error may not impede communication, as was stated above (Maria live in Cazenga as opposed to Maria lives in Cazenga), some grammatical errors can be distracting, and therefore, they can detract from fluency. In the following example, taken from a learner’s speech before the teaching was implemented in the present study, the speech was rapidly produced: “I’m see here two friends, is ready to give order or the food” (see Table 2, Clauses 1 and 2). Here, the learner intends to say I can see two friends who are ready to order the food, but the way the learner speaks the sentence may not be very comprehensible to the listener. This probability is supported by the argument that “accuracy and fluency do not operate in complete independence from each other” (Housen & Kuiken, 2009, p. 469), but they work in a complementary manner to maximize EFL learners’ speaking proficiency.

However, most studies have treated fluency and accuracy as separate components. In their study, in which they assessed the effects of different tasks on speech comprehensibility, Crowther et al. (2015) mentioned solely “segmentals, word stress, rhythm, and speech rate” as examples of fluency categories (p. 80). Similarly, in assessing the correlation between speaking fluency and accuracy, Brand and Götz (2011) used only temporal variables of fluency such as “speech rate, length of speech runs or the number and length of filled and unfilled pauses” (p. 257).

Speaking fluency is determined by several components such as speech rate or number of filled and unfilled pauses, number of errors, and use of formulaic language (Bøhn, 2015; Gut, 2009; Housen & Kuiken, 2009). A broader definition of speaking fluency, therefore, is needed in exploratory studies. That is, speaking fluency should be more broadly defined as the learners’ ability to produce a speech that is rapid and comprehensible (Brand & Götz, 2011; Crowther et al., 2015). In such a broad definition, searching for words is not observable. Furthermore, the grammar allows the listener to get information without ambiguities, and performance aspects of speech—such as er, erm, and ah—are used to maintain the flow of discourse (Brand & Götz, 2011; Nakatani, 2010).

Coping With Speaking Problems

How learners cope with their speaking problems has been widely studied (Lam, 2007; Nakatani, 2010; Rohani, 2011), but few or none of the studies have been conducted in the Angolan context. In the study by Lam (2007), the strategies that learners used to cope with their speaking problems were investigated in a Hong Kong high school when learners were engaged in a speaking task. Through the use of a stimulated recall method, which consisted of requesting learners to elaborate on the strategies they had just utilized, learners were found to request clarifications, simplify their speeches, plan their ideas in advance, and monitor other learners’ contributions. The stimulated recall method, therefore, was considered as an essential research tool because it enabled the researcher to gain understanding of learners’ strategies that were still found in their “short-term memory” (Lam, 2007, p. 58).

According to Lam (2007), when a long time has elapsed, after the learners have used the strategies, the strategies become routinized and move into the long-term memory. Consequently, the learners cannot remember the strategies, and the strategies become inaccessible to the teacher, who loses an opportunity to develop learners’ oral language (Gass & Mackey, 2000; Lam, 2007).

It has been suggested that once the strategies are identified, their effectiveness should be investigated (Lam, 2007). This suggestion was taken by Nakatani (2010), who investigated whether the use of specific strategies enhanced 62 college students’ oral communicative abilities in communicative tasks in Japan. In a 12-week experiment, the students were taught the oral communicative strategies, such as “paraphrasing, using gestures, and asking questions for clarification” (Nakatani, 2010, p. 117), and the findings indicated that the strategies for maintaining the flow of discourse and those for negotiating meaning were more conducive to the development of oral proficiency. However, little progress was noticeable in terms of language development, and, according to Nakatani, the little development probably occurred because the students did not have enough opportunity to improve the form.

In another study, complementing the study by Lam (2007), Rohani (2011) aimed to find the strategies that 23 EFL learners utilized for coping with oral communication in Indonesia. In particular, focus was placed on the understanding of whether the learners improved the strategies over time. By using a pre- and postinstruction questionnaire, as well as interviews with the learners, Rohani found that between the strategies for coping with speaking difficulties and the ones for coping with listening difficulties, learners made more positive changes in their strategies for coping with speaking difficulties; at the beginning of the instruction, learners tended to use more strategies for coping with listening difficulties, and therefore, they were more passive participants in the oral interactions. At the end of the instruction, however, learners reported that they had maximized their vocabulary and had reduced the use of gestures, with only a few learners reporting that they had maximized the use of gestures to overcome problems with anxiety. Overall, Rohani (2011) noticed that, in the instruction, learners were helped to gain “a more balanced use of strategies for coping with speaking and listening problems” (p. 94).

Feasibility of the TBLT Approach for Speaking Fluency

All the studies discussed in the subsection above (Lam, 2007; Nakatani, 2010; Rohani, 2011) have used tasks to elicit research data on the aspects of speaking proficiency. However, except Rohani (2011), no study has underscored the relevance of task and the use of TBLT approach. Task is a teaching tool that can enhance EFL learners’ development of speaking fluency. According to Rohani (2011), tasks are worth using because they help learners in focusing on meaning, learning real language, and stimulating natural acquisition process. This statement is supported by Hismanoglu and Hismanoglu (2011), who have stated that “the type of discourse emerging from task is aimed to be similar to the one that emerges naturally in the real world” (p. 48). In a similar manner, Ellis (2009) has argued that “all tasks are designed to instigate the same kind of interactional processes [ . . . ] that arise in naturally occurring language use” (p. 227).

From the perspective of supporting learners in a real-like environment, Buriro and Hayat (2010) conducted a study in Pakistan, in which they aimed to help EFL learners maximize their use of the target language and gain speaking fluency. After being instructed with role-play tasks, learners made subjective evaluation of the instruction, and perceived that they had made significant progress with their speaking fluency. According to Buriro and Hayat (2010), the findings served as a springboard for English language teacher’s future development.

Although the learners’ subjective evaluations of their speaking fluency provided insights into how beneficial the TBLT approach was, lack of learners’ actual speeches to assess the development of speaking fluency, as was done in the study by Nakatani (2010), might have enhanced the findings. Thus, besides seeking learners’ opinions, future studies in different contexts should obtain learners’ actual speeches before and after the instruction to assess how the progress with speaking fluency is made in the TBLT approach.

In summary, TBLT approach, as conceived by Long and Crookes (1992) and Ellis (2009), builds on situations in which learners use language as if they were using it outside the classroom. During classroom interactions, learners practice speaking by focusing “primarily on meaning,” conveying “information,” relying on “their linguistic and non-linguistic resources,” and utilizing language to achieve communicative purposes, as opposed to primarily focusing on linguistic forms (Ellis, 2009, p. 223). The TBLT approach also grants learners an opportunity to improve their use of linguistic forms because, as stated by Ellis (2009), the approach “need not be seen as an alternative to more traditional, form-focused approach but can be used alongside them” (p. 221). This position is different from the one taken in the task version by Prabhu (1987), in which tasks are not focused on linguistic forms, but the linguistic forms are expected to be learned by osmosis (Knight, 2001).

Because learners in EFL contexts need to learn linguistic forms, particularly for examination purposes in schools (Zheng & Borg, 2014), they can have a full benefit from the use of a TBLT approach that focuses on meaning and form. Learners can be supported in expressing meanings and can be helped in improving linguistic forms by using “recasts” and “prompts” (Guchte, Braaksma, Rijlaarsdam, & Bimmel, 2015).

Recasts have been defined as ways of providing corrective feedback, in which the teacher reformulates the learner’s speech in the correct form (Guchte et al., 2015; Lyster & Ranta, 1997). In this corrective feedback, the learner is not required to pause to notice the correct form. According to Guchte et al. (2015), recasts are “implicit” strategies for supporting learners in improving linguistic forms (p. 248). In the TBLT approach, these corrective-feedback strategies are appropriate when learners are engaged in the meaning-making phase of a task. Prompts, however, require learners to pause to notice the correct form, and they are appropriate when learners are in the form-focus phase of a task in the TBLT approach. What remains to be known, however, is the empirical evidence of how learners in EFL contexts, particularly in the Angolan context, can improve their speaking fluency in a TBLT approach that focuses on meaning and form. Therefore, the following research questions are answered in the present study.

Research Questions

The first research question is the core of this study because it helps in assessing how learners progress in terms of their speaking fluency from different perspectives. First, the question provides support in assessing the progress in terms of word frequency before and after the instruction, and second, it aids to evaluate the extent to which learners’ utterances exhibit clarity in terms of grammatical accuracy. Third, the question helps in exploring more general aspects of face-to-face interactions, such as the strategies that learners utilize when learners are engaged in a task (Lam, 2007; Nakatani, 2010).

The second research question is complementary to the first research question. It is used to assess how the learners perceive their learning English in general, and how they have improved their speaking fluency in particular, after being taught with the TBLT approach. The second research question replicates the previous studies that were conducted in other contexts such as in Japan and Pakistan, by Nakatani (2010) and Buriro and Hayat (2010), respectively.

Participants

The participants were a group of 40 learners who had been randomly selected from the 360 learners attending ninth grade at PUNIV-Cazenga. Twenty-two of them were male and 18 were female, and their ages ranged from 16 to 18 years. After being selected, learners were requested for voluntary participation, and, because they were teenagers, their parents were also contacted to gain informed consent, as required by the American Psychological Association (2010).

Method

A case study design was used, in which audio-recorded picture descriptions and audio-recorded interviews served as research instruments (Gerring, 2007; Samaranayake, 2016). These instruments were used to gain insights into how learners improved their speaking fluency and what their perspectives were on being taught with the TBLT approach. Regarding audio-recorded picture descriptions, three learners (the case studies) were randomly selected from the 40 learners, and their speeches were analyzed to evaluate how the learners improved their speaking fluency in the TBLT approach.

The picture-description instrument was similar to the “photo elicitation,” which was defined by Harper (2002) as “based on the simple idea of inserting a photograph into a research interview” (p. 13). According to Harper, “Photo elicitation evokes information, feelings, and memories that are due to the photo-graph’s particular form of representation” (p. 13). On the basis of this statement, in this study, pictures were thought to provide information to which learners referred in producing their speeches. The pictures were also deemed to help learners express their feelings of what they saw and felt when looking at the pictures (Eggins, 2004).

A reservation concerning the use of pictures for eliciting speech was expressed by Crowther et al. (2015), who stated that tasks based on such tools were “not reflective of real world contexts [ . . . ]” (p. 82). In particular, Crowther et al. (2015) referred to examples of real-world language tasks, such as the ones that are part of IELTS and TOEFL iBT tests, which learners may complete for entering university.

From the perspective of TBLT approach, the reservation by Crowther et al. (2015) is legitimate. That is, as was stated in the “Introduction” section of this article, learners should be exposed to real-world language in the TBLT approach (Ellis, 2009; Long & Crookes, 1992; Skehan, 1998). However, pictures may also focus on learners’ real-world language (Harper, 2002), as was the case in the present study (see Soars & Soars, 2011). Therefore, for the present study, which was exploratory in nature, the use of pictures to elicit learners’ speeches was appropriate. The instruments used to collect data are described in more detail next.

Instruments

Treatment

Methods of Data Analysis

The two data sets were qualitatively analyzed. In the first data set, learners’ speeches were transcribed, and the words in each transcript (before and after the teaching) were counted. This word count helped in assessing how learners improved their speaking fluency in terms of speed production. Because speed production may not always guarantee comprehensibility, as exemplified in the “Literature Review” section, learners’ transcripts were assessed in terms of grammatical accuracy. The comparison of word counts and the assessment of grammatical accuracy provided some insights into how learners improved in terms of their speaking fluency.

In the second data set (the interviews), the learners’ utterances were also manually transcribed and shared with the learners, to check that the texts conformed to what had been stated. The transcripts were then translated into English for the purpose of analysis. After being translated, the transcripts were repeatedly read to identify themes (Bøhn, 2015; Evans, 2012; Yayli, 2011), and some utterances were chosen to illustrate the learners’ perspectives about being taught with the TBLT approach.

Discussion

This study assessed how the EFL learners at PUNIV-Cazenga improved their speaking fluency in the TBLT approach. Through picture descriptions, learners’ speeches were audio recorded before and after the teaching that utilized the TBLT approach. During the teaching, the learners were engaged in tasks that promoted meaning making and focus on linguistic forms through the use of “recasts” and “prompting” (Guchte et al., 2015, p. 248). The findings indicated that learners improved their speaking fluency by maximizing the speed of speech production, improving grammatical accuracy, elaborating on their utterances, and developing interactional language.

An inspection of the data (Tables 1-3) also indicates that the learners’ utterances were overall better in terms of comprehensibility than they were before the teaching was implemented. This improvement appeared to result from the use of the tasks that focused on meaning and linguistics forms, and served as a support for the argument made by Housen and Kuiken (2009) that fluency and accuracy are not independent of each other, but they are complementary.

Furthermore, in the present study, learners increased their use of the strategies to maintain the flow of discourse. For example, as was stated earlier, before teaching, the learners used fillers as long pauses, whereas after teaching, they used fillers as short pauses, indicating the flow of speech. Another example was the use of mental processes such as think and guess, which helped learners to express their thoughts, feelings, and reactions. Those strategies were deemed to maximize speaking fluency. In fact, the findings were similar to those identified by Nakatani (2010), who observed that, among many strategies, the strategies for maintaining the flow of discourse were most conducive to the development of oral proficiency.

However, whereas in the study by Nakatani (2010) learners exhibited little progress in learning linguistic forms, in the present study, the learners made progress in terms of both the speed of speech production and linguistic forms. According to Nakatani (2010), as was stated in the “Literature Review” section, learners exhibited little progress with language development due to few opportunities to develop the form. In the present study, therefore, the different result may be attributed to the use of the TBLT approach that placed equal emphasis on meaning and linguistic forms.

Learners’ opinions on being taught with the TBLT were sought, and the results indicated that the approach might be used with success at PUNIV-Cazenga in particular, and in the Angolan context in general. After being taught with the TBLT approach, learners felt encouraged to speak, and that encouragement could be thought to make learners speak in any situation. In fact, learners believed that they were ready to participate in any communicative event outside the classroom. The focus on meaning and form, in the TBLT approach, was seen by the learners as an advantage from the standpoint of their examinations at school. However, they suggested a balance between the learning of linguistic forms and the practice of speaking fluency. One notable comment was made by a learner who suggested that all the teachers at PUNIV-Cazenga should adopt the TBLT approach.

The learners’ opinions on the usefulness of the TBLT approach are not new. In their study, in which they aimed to help EFL learners in maximizing their use of the target language and gaining speaking fluency, Buriro and Hayat (2010) found that learners were satisfied with the approach. The findings on the learner’s opinions in the present study, therefore, positively replicate the findings in the previous one. Thus, the findings on the learners’ opinions and the transcripts of the picture descriptions have considerable implications for teaching practice and future research.

Implications for Teaching Practice

The results of the present study imply that, at PUNIV-Cazenga, teachers should begin implementing the TBLT approach in their classes. This implication does not mean that the teachers should abandon the current syllabus (see the appendix), but rather they should incorporate task components (Ellis, 2009; Long & Crookes, 1992), so that the students learn English the way they are expected to use it outside the classroom. The use of the TBLT approach can provide teachers with an advantage from the standpoint of helping learners with their school examinations (Zheng & Borg, 2014) and with their use of the target language outside the classroom.

The results of the present study may also apply to other secondary schools in the Angolan context and to other countries in which English is learned as a foreign language. The use of the TBLT approach in such contexts may also be helpful in supporting learners with their examinations (Zheng & Borg, 2014) as well as with the language that is needed for daily communication outside the classroom.

Implications for Future Research

The present study has been an attempt to assess how EFL learners improve their speaking fluency at PUNIV-Cazenga and what opinions learners express on being taught with the approach. This study has complemented many studies that have thus far been conducted to investigate the way EFL learners can be helped in improving their speaking proficiency (Crowther et al., 2015; Lam, 2007; Nakatani, 2010; Rohani, 2011). These studies have focused on the strategies that learners utilize in coping with their speaking problems (Lam, 2007), how learners improve the strategies over time (Nakatani, 2010), and what types of tasks can probably help learners in coping with their speaking problems. From the perspective of finding ways of helping EFL learners with their speaking problems, the present study is similar to those undertaken in previous research.

However, the present study has been conducted in the Angolan context, in which findings on how learners can improve their speaking proficiency, in a TBLT approach, are underrepresented. Furthermore, in the present study, the term speaking fluency has been used in broad sense to explore several components that might contribute to the EFL learners’ improvement of speaking fluency. A notable contribution was to gather data not only from the learners’ utterances in an interview but also from the learners’ transcripts of their picture descriptions.

Many features were identified in the learners’ discourses when transcribing the pictures such as the development of vocabulary, improvement of grammatical accuracy, the use of interactional language, and elaboration of speech. According to Lam (2007), when features such as these are identified, it is necessary to investigate how they change when learners are helped with the speaking proficiency. Therefore, future studies in the Angolan context, and probably elsewhere, should investigate how the features change over time in the TBLT approach that is used to help learners in improving their speaking fluency.

Although this study has not aimed to assess the effects of the TBLT approach, the findings indicate that learners have exhibited better speaking fluency than they did before the teaching was implemented. This study used the TBLT approach as conceived by Long and Crookes (1992) and Ellis (2009), which focuses on meaning and linguistic form. Because there are other TBLT approaches that do not focus on linguistic forms, such as that of Prabhu (1987), future studies in the Angolan context, and probably elsewhere, might consider comparing the effects of different task types in developing speaking fluency.