The Effects of Task Variation on the Accuracy and Complexity of Iranian EFL Learners’ Oral Performance (Research Paper)

The Effects of Task Variation on the Accuracy and Complexity of Iranian EFL Learners’ Oral Performance

^[1]Shokoufeh Abbasi Dogolsara

^[2]Saeideh Ahangari*

IJEAP- 1905-1385

^[3]Zohreh Seifoori

Abstract         

Task variation is an integrative method aiming at the importance of learner-to-learner interactions in a wide range of learning contexts and fostering authentic use of language and meaningful communication. This study investigated the impact of task variation on the accuracy and complexity of Iranian EFL learners’ oral speech. In so doing, 80 intermediate EFL learners, majoring English at the Islamic Azad University, Tonekabon branch, were randomly assigned to four equal groups based on their scores on OPT:  Experimental Group 1 (EXG1), treated by a variation of three tasks, i.e., dialogue completion, group dialogue, and dialogue unscrambling; Experimental Group 2 (EXG2), treated by a variation of two tasks including dialogue completion and dialogue unscrambling; Experimental Group 3 (EXG3), instructed through a single task (dialogue unscrambling), and the Control Group (CONG), instructed through teacher routine or conventional method. The speaking part of the Preliminary English Test (PET) was administered to the groups as a pretest and posttest. The four groups received their required instructions for ten sessions. The results of the ANOVA and Post Hoc analyses indicated that the three task variations were effective on two features of accuracy and complexity. It was also revealed that the EXG1, treated through a variety of three tasks, outperformed the other three groups in terms of accuracy and complexity. Task variation is hence advised to be incorporated in EFL teaching and testing contexts as a useful tool to improve learners’ oral performance on accuracy and complexity.

Keywords: Accuracy, Complexity, EFL learners, Oral performance, Task variation

1. Introduction

One of the main concerns of teachers is to encourage learners to use English for the sake of communication and interaction with others. Speaking can be improved by a variety of individually-based practices and collectively-oriented techniques through the media of games, role play, representation, etc. Given some experiments, speaking can be enhanced by incorporating group work activities (Oradee, 2012) and  drawing learners’ attention to some types of speaking performances that can have conclusive impacts on improving their speaking skill (Brown, 2007). Richard and Rodgers (2001) pointed out that speaking and communicating with others through the spoken language using learner’s already existing linguistic and communicative resources is viewed as the basis for second language acquisition in Task-based Language Teaching (TBLT); therefore, a great deal of tasks that are proposed in TBLT involve conversation. Over the last two decades, the TBLT has played a vital role in teaching English and has continued to attract the attention of language teachers and researchers (Ellis, 2003; Long, 2015).

In spite of the importance of the speaking skill in Iranian English universities, most Iranian English students suffer weaknesses in their speaking performance. One of the problems that Iranian English learners are facing is their weakness in producing accurate and complex speech. This can be attributed, in light of the researchers’ experience (Shoarnaghavi, Seifoori, & Ghafoori, 2014), to the prevailing traditional methods of teaching speaking. They argued that Iranian classrooms are often focus-on-forms dominated, i.e., they just learn how to use the grammatical forms in isolated sentences, but they do not know how to use them to express different meanings. Students most of the time are rarely required to say more than a few words in response to questions raised by the teacher. Moreover, the methods adopted rarely offer students a clear explanation or instruction of what constitutes speaking and how native speakers actually speak. Therefore, the speaking skill remains a far-fetched goal or an ambiguous entity that is assumed to be hard to reach by students like other skills.  According to Dolati and Mikaili (2011), speaking, also, is not given enough attention while teaching English to students. They argued that this problem may be due to the syllabus, the methods of teaching and time limitation of the school programs.

Thus far, a lot of studies have been done to investigate two components of speaking including accuracy and complexity as dependent variables based on task-based instruction, and their variation have been assessed with respect to independent variables such as acquisitional levels or task features, the most notable of which have been shown by Hilton (2008), Larsen-Freeman (2006), and Tonkin (2007). To the knowledge of the researchers, however, few studies have addressed this issue in Iran so far. Worded differently, there exist few Iranian studies to have concentrated on the impact of task variation on the two features of speaking, including accuracy and complexity.

The three models of task variation (dialogue unscrambling, group dialogue, and dialogue completion tasks) are as practical tasks which are frequently used in EFL speaking classes at pre-intermediate and intermediate levels. These tasks and their potential interaction on oral speech accuracy and complexity have not been investigated thoroughly in the Iranian context yet.

The present study, therefore, aimed to examine the impact of task variation on the accuracy and complexity of L2 oral proficiency among undergraduate Iranian EFL learners. To this end, the EXG1, EXG2, EXG3 were considered as experimental groups of study, and there was one control group titled CONG. The EXG1 was treated by a variation among three tasks including dialogue completion, group dialogue, and dialogue unscrambling; the EXG2 underwent treatment through a variation of two tasks involving dialogue completion and dialogue unscrambling; the EXG3 was treated by a single task, i.e., dialogue unscrambling; and the CONG received placebo through teacher conventional method. Drawing on the variables based on which the study was designed and carried out, it was hypothesized that:

Hypothesis One: There are not any differences in the accuracy of oral speech scores of Iranian EFL learners instructed by various tasks.

Hypothesis Two: There are not any differences in the complexity of oral speech scores of Iranian EFL learners instructed by various tasks.

2. Review of the Related Literature

2.1. Speaking Proficiency

According to Tavakoli and Foster (2008), both accuracy and complexity are the essential features of speaking skill, but which one should receive priority is a matter of great concern for second language teachers’ and learners. Accuracy, according to Harmer (2001), refers to the correct and acceptable use of vocabulary, grammar, and pronunciation in L2 speakers’ language. Accounts on studies by Rahimpour (2008) and Larsen- Freeman (2006) reveal that accuracy is operationally measured by the number of error- free T-units i.e., the percentage of T-units that do not contain errors. Yuan and Ellis (2003), Guará-Tavares (2008), and Fatemi, Tafazoli, and Ghanbarizadeh (2015) measure accuracy by calculating the number of error-free clauses, taking all errors in syntax, morphology, and lexical choice into account. Michel, Kuiken, and Vedder (2007) embark on summing the total number of errors per the Analysis of Speech Unit (AS-unit). Complexity, however, refers to how many clauses the learner is able to connect or include in a sentence (Kim, Nam & Lee, 2016). It can be measured at two levels: grammatical complexity and lexical complexity. Grammatical complexity is usually measured by focusing on three criteria including global complexity, phrasal complexity, and complexity by subordination. Read (2000) suggests taking into account the relative frequency of the words, literally known as lexical sophistication, as a measure for the lexical complexity. The units that are most frequently used for measuring complexity are the T-unit (Larsen-Freeman, 2006; Sangarum, 2005)

Oradee (2012) maintains that appropriate and effective activities for oral speech improvement need to be introduced because they are integral and crucial components of speaking classes. He stresses that the adoption and implementation of properly communicative and suitable activities creates a supportive and effective environment, through which a proper relationship among the learners as well as between the learners and the teacher is fostered and hereby helps the effective improvement of speaking performance; Likewise, these activities decrease learners’ anxiety and stress, resulting in their better achievement and success.

Thornbury (2005) favors discussion as an oral exercise and integrative activity in speaking classes that specifically focus on dimensions of speaking skill. Likewise, Gentry, Gable, and Rizza (2002) and Thornbury (2005) show the significant role of discussion activities which make demand of teachers’ attention to their learners’ perception of the activities. Sotoudehnama and Hashamdar (2016) point to discussion task in EFL speaking practice contexts as one of the most effective and beneficial ways of practicing oral communications freely involving cooperation among the interlocutors and the ultimate result can be improvement in the learners speaking skill. Learners in classes may be organized quite purposefully in pairs or small groups to participate creatively in a discussion on various subjects. Fay, Garrod, and Carletta (2000), Al-Issa and Al-Qubtan (2010), and Miles (2009) highlight the importance of discussion exercise and claim that it integrates all the different language skills, activates the meaningful oral language, and facilitates the complex process of speaking mastery. What’s more, it improves the learners’ cooperation, responsibility, autonomy, and decision making which all encourage student-centered classrooms and enhance an independent and dynamic atmosphere in the classrooms. The results of the study by Oradee (2012) on the effect of group discussion on EFL learners’ oral proficiency yielded to the conclusion that discussion exercise was effective in oral proficiency improvement and the learners showed positive attitudes toward the group discussion practice fostered in their speaking classroom. Dialogic tasks, as Karami, Jafarigohar, Tajeddin, and Rouhi (2017) point out, can lead to greater accuracy and this effect on accuracy improvement is owing to students’ communication-driven tendency to the precision, focus on form, and recycling of partner’s language, reusing the correct language, and editing it.

Unscrambling task relies on learners’ ability to insert sentences to construct texts with the scaffoldings of lexical cohesion and sentence relations. It is regarded as an important task in learning since it examines students’ awareness in text structure and coherence. Students need to recognize the lexical cohesion ties, considered as a property of text, so as to connect sentences or intermingle textual information, which results in their comprehension of the text constructed. The unscrambling exercise which is an example of text construction involves some shuffled sentences from a text which are given to students to establish the sentence continuation for text coherence (Johns & Lixun, 1999; Ockey, 2009). This kind of text construction task involves students to benefit their meaning-making, connection-building, and relationships-constructing abilities to make logical connection among the scrambled items or information to construct a coherent text (Yeh & Yang, 2011). In the text construction process, students are supposed to organize related units and information in their memory which is useful for the comprehension of the text constructed (Ruiz-Primo, 2004). After the construction of a meaningful text and establishing a comprehensible paragraph through unscrambling task, the students go through monitoring and evaluating process that may enable them to possibly reformulate the cognitive and sentence structure (Yeh & Yang, 2011). 

About the effective use of dialogues in EFL and ESL contexts, Kuehn (2009) highlights five reasons why they play a role in listening and speaking classes: first, they are manifestation of real-life speech and represent the way speech is used in real-life contexts; second, they have learners practice culture in different social situations; third, students find chances to get involved in roleplaying; fourth, dialogue patterns supply rooms for practicing and learning new vocabulary and sentence structure; fifth, scaffolding learning is initiated and encouraged by dialogue representation which results in improved conversation ability.   

2.2. Task variation and their Orientation into SLA

The concept of task variation, which was once investigated into the field of sociolinguistics, found its way beyond the confines of sociolinguistic research and was explored within the language classrooms environments. The early indications of task- variability research findings were found in the works of Tarone (1983, 1985 &1988) and in task-based models represented by Skehan (1998), who reflected instances of variations relying upon the differences between task types and task conditions as two mainstreams of task revelations.

Tarone’s seminal works (1983; 1985; 1988) and Ellis’s (1985) focal attention to second language acquisition were the main sparks through which task variability gained an established position in the field of Second Language Acquisition (SLA). Tarone’s viewpoints originating from Labov’s stylistic continuum (1972), drew attention to the notion that learners differ in their language use, or shift styles, in light of their degree and capacity of paying attention to form, maintaining that at the highest level and degree of formal attention learners appear to adopt and show the careful style, and, in contrast, at the lowest degree of attention, they resort to the vernacular style.

Gregg (1990) pointed out that the variationist perspective does not provide any input to the dynamics of L2 acquisition. Mitchell and Myles (2004) attributed this lack of input provision to the absence of so-called psycholinguistic plausibility and the way learners continuously calculate probabilities and make a choice between alternative forms. Although variationist approach was not fully able to construct a comprehensive model to account for the theoretical and methodological directions of SLA research studies, the construct of variability has ever been revitalized by various scholars of different agendas, one of which was rooted in Van Patten’s (1990) investigation into the dichotomy existing between form and meaning whereby he made an attempt  to  revive the variability theory in order to explain the pressure underlying L2 learning process. Alongside, the emergence of task-based approach was a breakthrough into the pedagogy of L2 which attempted to introduce a principled view of task effects on various aspects of L2.

2.3. Empirical Studies on the Issue

Some studies have ever concentrated on the impact of task varieties on speaking accuracy and complexity. Teng (2007) investigated differences in the accuracy and complexity of EFL test takers’ oral speech in light of various task types involving answering questions, presentation, and picture description, through which he reported significant differences in the candidates’ complexity and fluency performance. Answering question variety of oral task was more productive than picture description type of task in the EFL candidates’ performance on complexity. Analytically rating score indicated significant differences in the subjects’ performance on complexity and fluency.

Birjandi and Alipour (2010) examined EFL intermediate learners’ performance on speaking accuracy and complexity employing two methods of individual and group pre-task planning in Iran context.  To this end, the individual and the group pre-task planning groups were tested through personal, decision-making, and narrative tasks. The findings of the study were indicative of two points: the learners who practiced individual planning outperformed the group planning learners in terms of speaking accuracy, and the group planning learners performed significantly better than the individual planners in terms of complexity. Seifoori and Goudarzi (2012) investigated the effectiveness of educational production-oriented program (oral production) on the grammatical accuracy of EFL learners’ oral performance by assigning the participants to an output experimental group and a non-output control group. The findings of the study indicated the positive and facilitative role of oral output in enhancing accuracy and fluency of the participants’ oral performance.

Shoarnaghavi, Seifoori, and Ghafoori (2014) investigated the probable impact of two varieties of tasks (divergent versus non-divergent tasks) on the enhancement of accuracy and complexity of Iranian intermediate EFL learners’ oral speech. The study was conducted to thirty learners studying Teaching English as a Foreign Language (TEFL). Analyzing the subjects’ recorded and transcribed oral representations showed the effect of the divergent tasks on the accuracy of oral production of the
learners. There were not, however, any significant differences in the complexity of speech between the groups because the complexity of speech remained almost unaffected.

Safari Vesal, Safari Vesal, and Tavakoli (2015) explored the impact of three task varieties involving introduction, long monologue, and discussion on the complexity and accuracy in EFL candidates’ oral production. The findings revealed that the three task varieties produced significant effects on the candidates’ complexity and accuracy. High complexity and accuracy were shown through the involvement of introduction and discussions tasks because they brought about further chances for participants’ interaction and exchange of information. Witton-Davies (2016) did a comparative study on the degree of complexity and fluency in light of two different varieties of oral performance tasks (monologue versus dialogue) which are considered as two common ways of eliciting learners’ speech features. Analysis of the data showed that the monologue task was more effective on complexity enhancement in that speech units and clauses were longer and there were more samples of clausal subordination. The dialogue tasks, however, resulted in significant fluency with respect to faster speech and less pausing than the monologue task design.

3. Methodology

All requirements of a quasi-experimental study, including pretest, posttest, randomization, treatment for the experimental groups and placebo for the control group were met in this study. The independent variable of the study was task variation, and the dependent variables, as examined in this study, involved accuracy and complexity in L2 oral performance. Task variation, as the independent variable in this study, involved using three different varieties of tasks including dialogue completion, group dialogue, and dialogue unscrambling tasks in such a way that one experimental group (EXG1) was treated by means of a combination of three tasks (dialogue completion, group dialogue, and dialogue unscrambling tasks, i.e., a variation among three types of tasks), the second experimental group (EXG2) through a combination of two tasks (dialogue completion plus dialogue unscrambling tasks, i.e., a variation between two types of tasks), the third experimental group (EXG3) by a single task (dialogue unscrambling), and the CONG was instructed through the normal routine procedure for teaching the materials for the course which was teacher-centered or teacher existing method including activities such as reading the conversation dialogues aloud fulfilled by either the teacher or the students and listening to the conversations texts. The participants in the experimental groups received a 10-session treatment, and the control group received a 10-session placebo. The participants in the four groups sat for a pretest before treatment and a posttest after treatment.

3.1. Participants

The participants in this study were 80 Iranian intermediate EFL learners majoring in English translation at the Islamic Azad University, Tonekabon branch. Of the participants, 45 were female and 35 were male. Their age ranged from 19 to 28 and shared the same language. All the participants’ first language was Persian. In order to obtain a homogenized group, the participants were given the Oxford Placement Test (OPT). The test was administered to 101 Iranian intermediate EFL learners, out of whom eighty were selected based on the results of the test. That is, the students whose scores fell between one standard deviation below and above the mean on OPT scores were considered as participants of the study. Then they were randomly assigned to four equal groups, each of whom comprised 20 ones, including EXG1, EXG2, and EXG3, all as the experimental groups of the study, and CONG as the control group.

3.2. Instruments

The Oxford Placement Test (OPT): To keep and ensure the homogeneity of the participants, the OPT was administered to 101 sophomore undergraduate students majoring in English translation. The OPT was composed of two main parts, including 60 multiple-choice items. The time allotted for the test was 30 minutes.

Pretest of Oral Proficiency (Preliminary English Test): The purpose of this test administered prior to the treatment was to find the possible initial differences in the English speaking ability of the four groups. The speaking section of Objective PET (Hashemi & Thomas, 2010) was used as the pretest of this study. It included four parts of oral questions. It took about 12 minutes for each candidate to answer the oral questions of the speaking section of the test. There were two raters (i.e. the researcher herself and an expert, a Ph.D. in TEFL as a university lecturer), but each participant interacted with one of them. The rater asked the participant questions in turn, using standardized questions. The participants responded to the oral questions of the speaking section of the test about present circumstances, past experiences, and future plans. Each examinee’s oral performance was recorded, transcribed and then measured based on two components of accuracy and complexity according to their measuring scale. The measures were operationalized as follows: accuracy was measured by dividing participants’ total number of errors by the total number of words produced and multiplying the result by 100. All errors in syntax, morphology or lexical choice were counted, including repetitions. Errors which were immediately self-corrected were counted and errors in pronunciation were not included in the analysis (Birjandi & Alipour, 2010; Fortkamp, 2000; Mehnert, 1998; Sangarum, 2005).  Complexity, in this study, was measured by counting the number of content words divided by total number of words and multiplying by 100 (Birjandi & Alipour, 2010; Ortega, 1999). In order to estimate the reliability of the PET (as the instrument for the pretest and posttest in this study) in an EFL context like Iran, it was administered to fifteen Iranian EFL students other than the target group participants, having the same characteristics as the target group participants in this study. Its reliability was measured through Cronbach alpha which showed a high index (0.81).

Posttest of Oral Proficiency (PET): It was equal in all respects to the pretest except for the rearrangement of the questions in speaking parts. This rearrangement was done to control the probable testing effect. The participants were again required to participate in four parts of the speaking section of the PET. The questions were responded by the participants and rated by the same raters as the pretest phase. Inter-rater reliability was calculated, and the average of the scores given by the two raters was considered as the final score of each participant. The inter-rater reliability between the scores of two raters showed an index of 0.81 which is a good sign of reliability.

3.3. Teaching Materials

The materials which were employed as treatment for all groups in this study consisted of 20 conversation samples. They were selected and represented in the forms of dialogue unscrambling, group dialogue and, dialogue completion exercises in equal quantity to serve the required treatment for each group. These various exercises as the material for treatment for the comparison groups were selected from a native modern English conversation and speaking source, Interchange 2 (Richards, 2012). According to Richards (2012), the Interchange series (numbers 1, 2 and 3) have been written for speaking and listening skills. The materials were screened on the basis of linguistic difficulty and intended purpose of treatment, and those which were compatible with their proficiency level were chosen for instruction.

3.4. Procedure

To begin with, the OPT was administered to 101 Iranian EFL sophomore students with an age range of 19 to 28, all majoring in translation. Then 80 candidates whose scores fell between one standard deviation below and above the mean on OPT scores were selected and randomly assigned to four equal 20-person groups: EXG1, EXG2, EXG3, and CONG. The participants in the four groups were given the speaking section of PET, including four parts, as the pretest. The test was administered orally through which two components of accuracy and complexity were taken into consideration.

The participants in each group were given their required treatment in their speaking classes: The participants in the EXG1 were treated through an integration of three tasks (dialogue completion, group dialogue, and dialogue unscrambling, i.e., a variation among three types of tasks). Each treatment session was equally divided into three thirty-minute time span, within which each task by itself was presented and practiced in thirty minutes. The first thirty minutes was spent on practicing group dialogue task in such a way that the participants in this group were presented the same dialogue texts as other comparison groups, through which they were required to practice conversation orally in groups or interactively with their classmates. The second thirty minutes of the treatment session was assigned to have the participants work on dialogue unscrambling task. Meanwhile, the last thirty minutes of the class was allotted to administering dialogue completion task to the same group. The researcher’s main concern throughout the treatment sessions was to foster discussions to the extent possible in order to maintain maximum chances for communication. This was true to be taken into consideration in the case of the other groups.

The participants in EXG2 received treatment through a combination of two tasks (dialogue completion plus dialogue unscrambling, i.e., a variation between two types of tasks). Each session was divided into two forty-five-minute halves in such a way that half of the treatment session time was spent on dialogue completion task practice and the second half (an equal amount of time) on the other task practice (dialogue unscrambling task practice). To present and practice the lesson through dialogue task practice, the participants were exposed to incomplete pieces or sections of meaningful dialogue texts, the same teaching material as the other groups, by which they were asked to fill and complete the blanks. It was referred to as a productive skill in which some parts of a conversation were left blank out of which the students were required to fill in the blanks to complete the conversation. This sort of completion exercise was done either individually or collectively through interaction with other peers.

TheEXG3, as the third experimental group, was treated through one task known as dialogue unscrambling task (i.e., variation through one type of task) whereby the twenty-member participants encountered conversation texts and exercises as teaching material via which they were supposed to unscramble the scrambled dialogue patterns during each session of treatment which prolonged one hour and thirty minutes.

Meanwhile, the participants in the CONG received a 10-session instruction based on the same material but through a different method which was the normal routine procedure for teaching the materials (teacher existing or conventional method, e. g., reading each sample of conversation dialogue aloud by the teacher or the students and listening to the dialogues on the tape recorder). The time allotted for each session of instruction presentation in the group was one hour and thirty minutes as it was in the case of the other three groups. Two samples of conversation were considered for each session.

After the completion of the treatment sessions, the same oral proficiency test (PET, Speaking section) was administered as posttest. Each participant’ interview both through the pretest and posttest took 10 to 12 minutes, and interviews were recorded to avoid the probable loss of data. The transcriptions were then coded for scoring and statistical analysis of complexity and accuracy to find out the impact of treatment for each group.

4. Results of the Study

This section reported and illustrated the analytical procedure for the data collected throughout doing the study in order to examine the research hypotheses according to which the research work was completed.

Table 1: Descriptive Statistics of the Participants' Scores on the OPT

The proficiency test (OPT) was administered as a homogeneity test to 101 participants at intermediate level, out of whom eighty were selected based on the results of the test. The mean score of the participants was 98.68 and those students whose score on the OPT fell within one standard deviation above and below the mean were selected as the participants of the study. As given in Table 2, according to the Levene's Test for Equality of Variances, it was inferred that the variance of participants was normal (F = 0.07, p= .78).

Table 2: Result of the Normality Test

4.1. Testing the First Null Hypothesis

A one-way ANOVA was conducted on the groups’ pretest accuracy scores in order to examine whether there was any difference in the participants’ performance on the accuracy of oral speech among the four groups of the study (EXG1, EXG2, EXG3, and CONG), the results of which were given in Tables 3 and 4.

Table 3: Descriptive Statistics on Pretest Accuracy for the EXG1, EXG2, EXG3, and CONG

As displayed in Table 3, there was no difference among the groups performance on oral speech accuracy. In order to show this, the inferential results were illustrated in Table 4.

Table 4: ANOVA Test of Pretest Accuracy

As Given in table 4, the results of conducting a one-way analysis of variance (ANOVA) showed that there was not any significant difference among the groups’ performance on oral speech accuracy (F =0.99, P = .40 > 0.05). Another ANOVA was conducted on the groups’ posttest scores on accuracy in order to compare the possible improvement of the participants from pretest to posttest as a result of being treated by the relevant tasks. The results of running ANOVA were presented in Tables 5 and 6.

Table 5: Descriptive Statistics on posttest Accuracy for the EXG1, EXG2, EXG3, and CONG

As seen in Table 5, as a comparison for the mean scores of the four groups, it was shown that the mean scores of the EXG1, EXG2, EXG3, and CONG in the posttest were all lower than those of the pretest. Accordingly, it is worth mentioning that, as in the calculation the errors have been considered so the less the obtained results, the better the accuracy would be.

Table 6: ANOVA Test of Posttest Accuracy

The results, obtained from conducting a one-way ANOVA in Table 6, indicated that there was a significant difference among the groups’ performance considering their scores on accuracy instructed by task variation, F = 14.24 > 1.00, P = .00 < 0.05. Given the two criteria, the null hypothesis, which was constructed based upon the supposition that there existed no significant differences in the accuracy scores of EFL learners treated by task variation, was rejected. It was, then, concluded that task variation was effective on learners’ achievement in oral speech accuracy. In order to show the exact place of difference among the groups’ performance on accuracy, a Post Hoc test was run, the statistical results of which are represented in Table 7. 

Table 7: Post Hoc Test Result on Accuracy for the EXG1, EXG2, EXG3, and CONG

As represented in Table 7, the level of significance between EXG1 and EXG2 was shown as 0.01 which was lower than 0.05. This indicated that the two groups were significantly different after the treatment. Also, the level of significance between EXG1 and EXG3 was shown as 0.03 which was again lower than 0.05. This indicated that the two groups were significantly different after the treatment. As for a comparison between EXG1 and CONG, the level of significance was 0.00 which was again lower than 0.05 and indicated that the two groups were significantly different after the treatment. There was shown a level of significance of 0.03 between EXG2 and CONG (p = 0.03 < 0.05) and 0.01 between EXG3 and CONG (p = 0.01 < 0.05), both of which were indicative of the significant difference between the groups after the treatment. It was, However, revealed that there was not any significant level of difference between EXG2 and EXG3, showing no significant difference after the treatment. Considering the post-test mean scores of the groups, the first experimental group which has been treated by the combination of three different tasks has received the lowest score and outperformed the other three groups. 

4. 2. Testing the Second Null Hypothesis

In order to examine whether there was any difference in the participants’ performance on the oral speech complexity pretest scores among the four groups of the study (EXG1, EXG2, EXG3, and CONG), a one-way ANOVA was conducted, the results of which were given in Tables 8 and 9.

Table 8: Descriptive Statistics on Pretest Complexity for the EXG1, EXG2, EXG3, and CONG

As seen in table 8, there was no difference among the groups considering their performance on oral speech complexity.  To find out this, the results were reflected in table 9.

Table 9: ANOVA Test of Pretest Complexity

According to Table 9, the one-way analysis of variance revealed that there was not any significant difference among the groups concerning their scores on complexity considering F = 0.19 , P = 0.90 > 0.05. To find out the significant difference in the participants’ performance on posttest oral speech complexity among the four groups, another ANOVA was run, the outcome of which is shown in Tables 10 and 11.     

Table 10: Descriptive Statistics on Posttest Complexity for the EXG1, EXG2, EXG3, and CONG

Table 10 indicated that mean scores among the four groups under study were different, showing that the groups had different performance on complexity on the posttest stage.

Table 11: ANOVA Test of Posttest Complexity

Based on the information in Table 11, the ANOVA analysis indicated significant differences in the complexity scores of the participants among the four groups (F = 6.05 > 1.00; P = 0.00 < 0.05).  Thus, it can be concluded that the task variation was effective on the improvement of oral speech complexity in the participants of different groups examined in this study. Accordingly, the null hypothesis that suggested there was not any significant difference in the oral speech complexity scores of EFL learners treated by task variation was rejected. In addition, to indicate the place of differences among the groups statistically, a Post Hoc test was conducted. Its results were illustrated in Table 12.

Table 12: Post Hoc Test on Complexity for the EXG1, EXG2, EXG3, and CONG

According to Table 12, the level of significance between EXG1 and EXG2 were shown as 0.04 which was lower than 0.05. This indicated that the two groups were significantly different after the treatment. Also, the level of significance between EXG1 and EXG3 was shown as 0.02 which was again lower than 0.05. This indicated that the two groups were significantly different after the treatment. As for a comparison between EXG1 and CONG, the level of significance was 0.00 which was again lower than 0.05 and indicated that the two groups were significantly different after the treatment. However, it was revealed that levels of significance between EXG2 and EXG3 and also between EXG2 and CONG were higher than 0.05, showing no significant difference after the treatment. In addition, as for EXG3 and CONG, the level of significance was 0.09. So it can be concluded that the two groups did not show any significant difference after the treatment. Considering the mean scores of the four groups, it can be concluded that the first experimental group which has been treated by way of three different tasks, by having the highest mean score has outperformed the other three groups.

5. Discussion

The main point that this study explored was to investigate the effects of task variation on two components of accuracy and complexity of Iranian intermediate EFL learners’ oral speech. For this purpose, eighty EFL learners were randomly assigned into four comparison groups, each comprising twenty participants. The research questions of this study separately asked whether the employment of task variation would have any effects on Iranian intermediate EFL learners’ accuracy and complexity.  In accordance with these research questions, two null hypotheses were also proposed stating that using task variation would have no impact on EFL learners’ accuracy and complexity in oral speech.

The statistical analysis related to accuracy results show significant differences in the groups’ achievement in accuracy after they received treatment (Table 6). It was inferred that there were significant differences among the groups’ oral accuracy scores instructed by various tasks. It was then concluded that task variation was effective on the EFL learners’ accuracy in oral speech.  Accordingly, the null hypothesis of accuracy was rejected. To represent the significant levels of differences among the posttest scores of EXG1, EXG2, EXG3, and CONG, a Post Hoc test was run to calculate the differences (Table 7). Based on the statistical analyses obtained from the Post Hoc results, it was shown that there were significant levels of differences among the groups after the treatment fulfillment. It was inferred that the EXG1, treated through a variety of three tasks involving dialogue completion, group dialogue, and dialogue unscrambling, had the highest performance of the three groups (EXG2, EXG3, and CONG) on the component of accuracy in oral speech. A variety of three tasks, through which the EXG1 was treated, resulted in higher achievement in the participants’ speech accuracy. There were also significant differences between EXG2 and CONG as well as between EXG3 and CONG, indicating that the groups were significantly different after being treated by their required tasks.

The data obtained from the relevant statistical procedures on oral speech complexity, concerning the second hypothesis, revealed that task variation was effective on the improvement of the speech complexity in EXG1, EXG2, EXG3, and CONG, treated by various tasks. Worded differently, the results of the statistical analyses on complexity (Table 11) conducted on the posttest scores of the groups (EXG1, EXG2, EXG3, and CONG) showed significant differences in their performance on complexity. These findings indicated the efficacy of task variation on the intermediate EFL learners’ achievement in oral speech complexity.  The results of the Post Hoc test, conducted on the posttest complexity scores of EXG1, EXG2, EXG3, and CONG (Table 12), showed that there existed significant difference between EXG1 and EXG2, EXG1 and EXG3, as well as, EXG1 and CONG. indicating that the groups were significantly different after the treatment. It was shown that the EXG1, treated by a variety of three tasks including dialogue completion, group dialogue, and dialogue unscrambling, outperformed the other groups in terms of complexity achievement.

The finding of the study show that exposing the learners to various tasks supplies them with the situation to get more deeply engaged in the process of meaningful learning. This study showed that the nature of the tasks that L2 learners were engaged in prepares them to prioritize different aspects of language. It is discussed that L2 learner's performance differs from task to task. In other words, different task types will yield to different kinds of production by L2 learners. Therefore, a particular type of task that a learner is asked to perform will also result in variation. This kind of variation, which is due to task differences, is called "task-induced variation" by Ellis. He argues that this is best considered as a blanket term to cover the variability evident when learners perform different tasks (Ellis, 1994:135). This idea is also confirmed by Tarone (1982, 1983, 1985, 1988, 1990), who argues that as second language learners perform different tasks, their production of some grammatical, morphological, and phonological forms will vary in a particular manner. Accordingly, this study recommends employing task variation in EFL contexts where learners have limited opportunities to use the target language outside the classrooms. Likewise, the study proposes that using task variation in conversation classes is effective for intermediate-level learners, helping them enhance their complexity and accuracy in oral speech.

The findings of this study seem to be in line with the results achieved by Birjandi and Alipour, (2010), Munirah and Muhsin (2015), Safari Vesal, Safari Vesal, and Tavakoli (2015) Seifoori and Goudarzi (2012), Teng (2007), and Witton-Davies (2016), through which they examined EFL learners’ speaking accuracy and complexity through a variety of task-based instruction and reported effective results in learners’ improvement in these two components. The findings of the current study, however, as far as the complexity results are concerned, are not consistent with what Shoarnaghavi, Seifoori, and Ghafoori (2014) reported in that the two variety of tasks examined did not result in the participants’ enhancement in speech complexity, saying that there were not any significant differences in the performance of the groups on oral speech complexity. Moreover, the results of this study are not found to be consistent with Rahimpour and Mehrang’s (2010) findings on the impact of two varieties of task structure on the performance of Iranian EFL learners with regard to speech accuracy and complexity. The results of their study showed that the two varieties of task structure did not have any effect on the accuracy and complexity of oral performance. 

Therefore, the findings of this study, being in line with those of the studies mentioned above, can be a good justification for putting more emphasis on teaching two components of accuracy and complexity of oral speech through task variation in EFL classes. As it can be inferred from the results of this study, a variation of three tasks like dialogue completion, group dialogue, and dialogue unscrambling can be more effective than a single type of task such as dialogue completion, dialogue unscrambling, or group dialogue in developing accuracy and complexity in EFL classes.

The greater accuracy in the performance of the first group treated through different task types can be interpreted in terms of Long's view (1985) that different tasks will lead learners to stretch their interlanguage resources. The accuracy can also be interpreted in terms of Givon's (1985) pragmatic and syntactic modes demanding learners to use greater syntactic resources and abilities which will lead to an increase in grammatical accuracy. The accuracy can also be attributed to the load of attention paid by the learners to the different tasks than paid to just one task. The better performance in various task types can also be attributed to Tarone (1979), claiming that language varies with a subtle shift of situation, just as a chameleon changes as its surroundings change. She proposes that the interlanguage of L2 should be viewed as a variable system, a system that changes when a linguistic environment changes.

6. Conclusion

This study attempted to investigate the effectiveness of task variation of TBLT approach in enhancing accuracy and complexity of Iranian intermediate EFL learners’ oral speech. The findings of the study indicated that employing various tasks in the form of task variation produced positive effects on the learners’ two features of accuracy and complexity based upon their performance on their oral posttest. Their effects, however, were shown to be different from each other, depending upon the type of task variation.  Relying on the results of this study, it was revealed that the first experimental group who was treated through a variation of three tasks, i.e., group dialogue, dialogue completion, and dialogue unscrambling, outperformed other groups in terms of accuracy and complexity.

Relying upon the results of the present study, the following pedagogical implications can be drawn. The findings of the current study can be of interest to learners, language teaching decision makers, and syllabus designers. The findings of this study are useful for EFL learners who have limited chance for improving accuracy and complexity of the target language outside classrooms. Being involved in task-based instruction involving task variation prepares learners to practice oral speech in the context of the various speaking tasks, encouraging them to focus on the accuracy in the initial stage of the task (pre-task stage), next on fluency and spontaneous speaking during performing the task, and then reflect and acquire more skills at the post-task stage. Learners, in this way, are given wider chance to improve conversation management with their usual focus on grammar and vocabulary. Supplying learners with various tasks increases their chance and creates them the situation to get more deeply engaged in the process of meaningful learning.  Likewise, the findings of the current study are beneficial to language teaching decision makers as stakeholders in the field of language teaching. In fact, utilizing a variety of speaking tasks of (i.e. dialogue completion, group dialogue, and dialogue unscrambling) can provide tremendous opportunities for learners to enhance their accuracy and complexity of speech compared to traditional language teaching methods. EFL teachers are needed to focus on the implementation of various tasks through which they can provide further chance to their learners to pay more attention to the simultaneous enhancement of the two components of oral speech, i.e., accuracy and complexity in students. Exposing students to various tasks in oral speech features allows teachers to create more opportunities for eliciting further characteristics from their candidates’ speaking abilities so that teachers can identify their learners’ weaknesses and strengths in speaking and inform them of ways of overcoming shortcomings in their speaking ability. Task variation is more likely to allow candidates to manifest their best levels of speaking abilities than the contexts where the students are taught through a single task.  This idea is supported by Skehan (1998b) claiming that as learners pay attention to one aspect of a demanding task, they find it difficult to attend to another aspect. Moreover, the findings of this study may be beneficial to syllabus designers to incorporate task variation of TBLT approach into the language methodology and to encourage language teachers to draw the learners’ attention to the two essential features of speaking skill i. e., accuracy and complexity while completing tasks. As a result, massive and considerable improvement on the learners’ oral speech can be observed.

The present study was conducted to eighty intermediate EFL learners incorporating a variation of three tasks. The future studies of similar nature are proposed to incorporate other levels of proficiency including pre-intermediate and advanced candidates. Given the design of the current study, three various tasks were chosen as independent variables to serve the purpose of the research work. It is suggested that the future researchers incorporate some more tasks of similar nature which are applicable to spoken language. Certain limitations were involved in the present study. One limitation was that the study was conducted with a small size of population including Iranian intermediate-level EFL learners studying English at the Islamic Azad University, Tonekabon branch. The study should involve more participants at different proficiency levels in order to generalize the results for larger groups. Another limitation of the present study was that it incorporated an unequal number of both genders. Thus, for other researchers, equal number of both genders may be considered in order to make broader generalization about the results.

References 

Al-Issa, A. S., & Al-Qubtan, R. (2010). Taking the floor: Oral presentations in EFL classrooms. TESOL Journal, 1, 227-246.

Batstone, R. (2005). Planning as discourse activity. In R. Ellis (Ed.), Planning and Task Performance in a Second Language (pp. 262-279). University of Auckland: John Benjamins.

Birjandi, P., & Alipour, S.  (2010). Comparing the effect of individual and group pre-task planning on EFL learners’ accuracy and complexity in speaking. Journal of English Language Studies, 1(1), 1-22.

Brown, H. D. (2007). Principle of language learning and teaching. White Plains: Pearson Longman.

Dolati, R., & Mikaili, P. (2011). Opinion related to the main reasons on Iranian students’ difficulties in spoken English proficiency. The Australian Journal of Basic and Applied Sciences, 5(11), 1142-1148.

Ellis, R. (1985). A variable competence model of second language acquisition. International Review of Applied Linguistics in Language Teaching, 23, 47-59.

Ellis, R. (1994). The study of second language acquisition. Oxford: Oxford University Press.

Ellis, R. (2003). Task-based language learning and teaching. Oxford: Oxford University Press.

Ellis, R. (2006).  The methodology of task-based teaching. Asian EFL Journal, 8(3), 19-45.

Fatemi, M. A., Tafazoli, N., & Ghanbarizadeh, M. (2015). The effect of pre-task planning and online planning on fluency and accuracy in L2 oral production between intermediate learners in Iran.  International Journal of Language Learning and Applied Linguistics World, 8(3), 80-89.

Fay, N., Garrod, S., & Carletta, J. (2000). Group discussion as interactive dialogue or as serial monologue: The influence of group size. Psychological Science, 11(6), 48-486.

Feneey, A. (2006). Task-based language teaching. ELT Journal, 60(2), 199-201.

Fortkamp, M. B. M. (2000). Working memory capacity and L2 speech production: An exploratory study. (Unpublished doctoral dissertation), Universidade Federal de Santa Catarina, Brazil.

Foster, P., & Skehan, P. (1996). The influence of planning and task type on second language performance. Studies in Second Language Acquisition, 18(3), 299-323.

Gardner, R. C. (2010). Motivation and second language acquisition: The socio-educational model. New York: Peter Lang.

Gentry, M., Gable, R. K., & Rizza, M. G. (2002). Students’ perceptions of classroom activities: Are there grade-level and gender differences? Journal of Educational Psychology, 94(3), 539–544.

Givon, T. (1985). Function, structure and language acquisition. In D. I. Slobin (Ed.), The Cross-linguistic Study of Language Acquisition (pp. 1008-1025). Hillsdale, NJ: Lawrence Erlbaum.

Gregg, K. (1990). The variable competence model of second language acquisition, and why it isn’t. Applied Linguistics, 11, 364-383.

Guará-Tavares, M. (2008). Pre-task planning, working memory capacity and L2 speech performance. (Unpublished doctoral dissertation). Universidade Federal de Santa Catarina, Florianópolis, Brazil.

Harmer, J. (2001). The practice of English language teaching. New York: Pearson Longman.

Hashemi, L., & Thomas, B. (2010). Objective PET. Cambridge: Cambridge University Press.

Hilton, H. (2008). The link between vocabulary knowledge and spoken L2 fluency. Language Learning Journal, 36, 153–166.

Jamshidnejad, A. (2011). Developing accuracy by using oral communication strategies in EFL interactions. Journal of Language Teaching and Research, 2(3), 530-536.

Johns, T. F., & Lixun, W. (1999). Four versions of a sentence-shuffling program. System, 27, 329-338.

Karami, M., Jafarigohar, M., Tajeddin, Z., & Rouhi, A. (2017). Input-induced variation in EFL learners’ oral production in terms of complexity, accuracy, and fluency. Iranian Journal of English for Academic Purposes, 6(2), 70-85.

Kim, Y., Nam, J., & Lee, S. Y.  (2016). Correlation of proficiency with complexity, accuracy, and fluency in spoken and written production: Evidence from L2 Korean. Journal of the National Council of Less Commonly Taught Languages, 19, 147-181.

Labov, W. (1972). Language in the inner city. Oxford: Basil Blackwell.

Larsen-Freeman, D. (2006). The emergence of complexity, fluency and accuracy in the oral and written production of five Chinese learners of English. Applied Linguistics, 27(4), 590-619.

Laufer, B., & Hulstijn, J. H. (2001). Incidental vocabulary acquisition in a second language: The construct of task induced involvement. Applied Linguistics, 22(1), 1-26.

Long, M. H. (1985). A role for instruction in second language acquisition: Task-based language teaching. Modelling and Assessing Second Language Acquisition, 18, 77-99.

Long, M. (2015). Second language acquisition and task-based language teaching. Oxford: Wiley Blackwell.

Mehnert, U. (1998). The effects of different lengths of time for planning on second language performance. Studies in Second Language Acquisition, 20, 83-108.

Michel, M. C., Kuiken, F., & Vedder, I. (2007). The influence of complexity in monologic versus dialogic tasks in Dutch L2. International Review of Applied Linguistics in Language Teaching, 45(3), 241-259.

Miles, R. (2009). Oral presentations for English proficiency purposes. Reflections on English Language Teaching, 8(2), 103–110.

Mitchell, R., & Myles, F. (2004). Second language learning theories. London: Arnold.

Munirah, M. & Muhsin, A. (2015). Using task-based approach in improving the students’ speaking accuracy and fluency. Journal of Education and Human Development, 4(3), 181-190.

Ockey, G. (2009). Developments and challenges in the use of computer-based testing for assessing second language ability. Modern Language Journal, 93, 836-847.

Oradee, T. (2012). Developing speaking skills using three communicative activities (discussion, problem-solving, and role- play). International Journal of Social Science and Humanity, 2(6), 532-533.

Ortega, L. (1999). Planning and focus on form in L2 oral performance. Studies in Second Language Acquisition, 21(1), 109-148.

Rahimpour, M. (2008). Implementation of task-based approaches to language teaching. Pajohesh-e-Zabanhay-e- Khareji, 41, 45-61.

Rahimpour, M., & Mehrang, F. (2010).Investigating effects of task structure on EFL learner’s oral performance. English Language Teaching, 3(4), 10-17.

Read, J. (2000). Assessing vocabulary. Cambridge: Cambridge University Press.

Ruiz-Primo, M. A. (2004). Examining concept maps as an assessment tool. In A.J. Cañas, J. D. Novak, & F.M. González (Eds.), Concept Maps: Theory, Methodology, Technology (pp. 555–562). Pamplona, Spain: Universidad Pública de Navarra.

Safari Vesal N., Safari Vesal, M, & Tavakoli, M. (2015). The effect of task type on complexity, accuracy, and fluency of Iranian EFL candidates' oral production: IELTS interview test in focus. Science Journal, 36, 1154-1168.

Sangarun, J. (2005). The effects of focusing on meaning and form in strategic planning. In R. Ellis (Ed.), Planning and Task Performance in a Second Language (pp.111-141). Philadelphia, PA: John Benjamins.

Seifoori, Z. & Goudarzi, S. (2012). The effect of oral production on grammatical accuracy and task-based fluency in speech of Iranian EFL learners. Educational Quarterly, 5(18), 54-66.

Shoarnaghavia, R.,  Seifoorib, Z., & Ghafoori, N. (2014). The Impact of divergent tasks on the accuracy and complexity of intermediate Iranian EFL learners' task-based oral speech. Procedia: Social and Behavioral Sciences, 98, 1762-1770.

Skehan, P. (1998). A cognitive approach to language learning. Oxford: Oxford University Press.

Sotoudehnama, E. & Hashamdar, M. (2016). Oral presentation vs. free discussion: Iranian intermediate EFL learners’ speaking proficiency and perception. Applied Research on English Language, 5(2), 211-236.

Tarone, E. (1982). Systematicity and attention in interlanguage1. Language learning, 32(1), 69-84.

Tarone, E. (1983). On the variability of interlanguage systems. Applied Linguistics, 4, 142-163.

Tarone, E. (1985). Variability in interlanguage use: A study of style-shifting in morphology and syntax. Language Learning, 35, 373-404.

Tarone, E. (1988). Variation in interlanguage. London: Edward Arnold.

Tarone, E. (1990). On variation in interlanguage: A response to Gregg. Applied Linguistics, 11(4), 392-400.

Tavakoli, P., & Foster, P. (2008). Task design and second language performance: The effect of narrative type on learner output. Language Learning, 58, 439–73.

Teng, H. C. (2007). A study of task type for L2 speaking assessment. Paper presented to the Annual Meeting of the International Society for Language Studies (ISLS). National Yunlin University of Science & Technology, Honolulu, 1-11.

Thornbury, S. (2005). How to teach speaking. Essex: Pearson Longman.

Tonkyn, A. (2007). Short-term changes in complexity, accuracy and fluency: Developing progress- sensitive proficiency measures. In S. Van Daele, A. Housen, F. Kuiken, M. Pierrard, & I. Vedder (Eds.), Complexity, Accuracy and Fluency in Second Language Use, Learning and Teaching (pp. 263-284). Department of Applied Linguistics: University of Reading.

Van Patten, B. (1990). Attending to form and content in the input: An experiment in consciousness. Studies in Second Language Acquisition, 12, 287-301.

Witton-Davies, G. (2016). A comparison of fluency and complexity in two different kinds of oral test. Paper presented to the Twenty-fifth International Symposium on English Teaching. National Taiwan University, 353-365.

Yahay, M., & Kheirzadeh, S. (2015). The impact of oral presentation on fluency and accuracy of Iranian EFL learners’ speaking. Journal of Applied Linguistics and Language Research, 2(5), 114-123.

Yeh, H. C., & Yang, Y-F. (2011). Metacognitive process in online text construction.
Educational Technology and Society, 14(3), 82-101.

Yuan, F., & Ellis, R. (2003). The effects of pre-task planning and on-line planning on fluency, complexity, and accuracy in L2 monologic oral production. Applied Linguistics, 24, 1-27.