August 2019 – Volume 23, Number 2
Saraswati Dawadi
The Open University, England
<saraswoti_dawadi
yahoo.com>
Abstract
There is a substantial body of research into the effects of planning time on second language (L2) oral performance, but different effects were found – some of them showing increased accuracy, others increased complexity and/or fluency of language. In this study, 30 English as a foreign language learner individually performed a narrative and an argumentative task under planned and unplanned conditions. Then, ten of the participants took part in a retrospective debriefing. Both the oral narratives and retrospective reports were transcribed and analyzed manually. The results revealed that planning led to syntactically more varied but less accurate language whilst fluency remained unaffected. Significant effects of task type could be observed, with argumentation outperforming narration on most measures. In addition, an interaction between planning and task type was statistically proven for almost all measures of fluency and for a single measure of complexity. Furthermore, most participants had positive attitudes towards pre-task planning. The findings also suggest that choosing suitable task-based implementational conditions can assist L2 learners in improving oral performances.
Key words: Tasks, task planning, narration, argumentation, English as a foreign language
Introduction
A task refers to an activity which requires learners to use language in order to attain an objective. Since 1990s, tasks have been used in second language (L2) classroom as a major and influential tool to promote interaction (Bygate, 2001; Long, 2007; Ortega, 1995, 2005; Robinson, 2005; Salimi, Alavinia & Hosseini, 2012; Yi & Ni, 2015). Tasks with different communicative demands and discourse characteristics are expected to yield different types of language (Foster & Skehan, 1996; Tan, 2013). For example, an argumentative task requires learners to generate information with a good reason, which makes the task more demanding than simply describing a picture. Consequently, these two tasks might lead to different outcomes. However, there is divergence of views among researchers about the effects of task types on L2 performance. Therefore, research on task-based language teaching (TBLT) has witnessed a steady interest in investigating the effects of task type on L2 performance. Consequently, pedagogical practices have been changed though there are “real issues in the design and implementation of TBLT courses” (Ellis, 2017, p.508).
Task effects on L2 production seems to be affected by task planning. Pre-task planning gives learners an opportunity to plan what language to use to perform the given task. Ellis (2005) highlights that pre-task planning contributes to the selective attention to language form. The availability of pre-task planning time is also expected to reduce some of the cognitive burdens created by a task and to support learners to mentally organize the contents of a task and work on the formulation of the content (Bygate & Samuda, 2005). This kind of preparation may better support learners in assessing task demands and available linguistic resources. As a result, learners might focus attention on language form in their subsequent performances. The opportunity to plan might also lessen communicative stress and enable learners to focus their attention on form (Skehan, 1998). Thus, planning seems to act as an external trigger to encourage learners to focus both on form and meaning. However, there is scant evidence of planning affecting all three dimensions of language production- complexity (the use of a wide range of structures and vocabulary), accuracy (the correct use of a language), and fluency (the ability to produce L2 with native like rapidity without undue pausing, hesitation or reformulation) (CAF) (Skehan, 2009).
Literature Review
The literature review will first identify key features of existing research about planning effects on L2 performance. It will then go on to consider the findings of existing research about the effects of task type on L2 production. Then, it will present a gap in research, particularly in two key areas within this topic. These comprise studies which consider the interaction of planning and task type on L2 performance.
Research on Pre-task Planning Effects
Task planning seems to have evolved into an area of inquiry in its own right and it has recently become a burgeoning area of investigation in the field of TBLT (Ortega, 2005). Consequently, there is an extensive body of research on task planning. Some of the major pre-task planning studies are summarized in Table 1. The table indicates that previous studies in task-planning are not conclusive. For instance, a number of studies (Alazani, 2016; Ellis, 1987; Haghverdi, Biria & Khalaji, 2013; Li, Chen & Sun, 2015; Mochizuki & Ortega, 2008; Salimi, Alavinia & Hosseini, 2012) reported that planning triggered more accurate language. Ellis (1987) further argued that increased planning time leads to higher accuracy. However, as pointed out by Crookes (1989), the study lacks clarity in the conceptualization of the independent variable. It is not clear whether this effect is as a result of the planning opportunity or linked to task-related constraints.
Additionally, the proficiency differences of the participants, with different L1 background, are not controlled for, which might potentially seriously confound the accuracy of the analysis. Furthermore, Ellis’s claim that the longer the planning time, the higher the L2 accuracy is not well supported by Mehnert’s (1998) study, which found that just one minute planning triggered more accurate language but allowing more minutes (5 minutes or 10 minutes) did not result in any additional gains in accuracy. A recent study by Li, et al. (2015), which investigated the effects of different lengths of planning time (nil, 30 seconds, 1 minute, 2 minutes, 3 minutes, and 5 minutes) on L2 learners’ oral test performances, also made similar claim that “the positive effects of planning was not always observed in line with the increase of time” (p.38); one-minute planning time seemed to be the threshold to have positive effects on L2 performances. The authors also claimed that among the CAF measures, accuracy improved the most with pre-task planning.
Table 1: Summary of Some Previous Pre-Task Planning Studies
| Study | participants | tasks | Operationalization of planning | Major findings |
| Ellis (1987) | 17 English as a foreign language (EFL) learners | story telling | 1.write, 2. tell 3. plan and tell | Increased accuracy |
| Crookes (1989) | 40 Japanese EFL learners | description and explanation | 10-min unguided planning + note taking and no planning | Increased complexity but no effects on accuracy |
| Ortega (1995) | 14 Spanish learners | oral narrative | 8 min unguided planning and no planning | Increased complexity but no effects on accuracy |
| Wendel (1997) | 40 Japanese EFL learners | oral narratives | 10-minute unguided planning with note taking and no planning | Increased fluency and complexity, but no significant effects on accuracy |
| Ortega (999) | 64 Spanish learners | story telling | 10-min unguided planning+ note making and no planning | Increased fluency and complexity but no significant effects on accuracy |
| Mehnert (1998) | 31 German learners | instruction and exposition | 1 min, 5 min, 10 min and no planning | Increased fluency, increased accuracy with only 1 min planning, and greater lexical diversity with 10 min planning only. |
| Yuan and Ellis (2003) | 42 EFL learners | oral narrative | 10-min unguided planning + note making and no planning | Increased fluency and complexity but no significant effects on accuracy |
| Philp, Oliver and Mackey (2006) | 42 ESL learners | information gap activities | 2-min, and 5-min unguided planning and no planning | Limited benefits on fluency and accuracy, but increased complexity with 5 min planning. |
| Mochizuki and Ortega (2008) | 56 EFL learners | picture story-retelling | 5-min unguided, 5-min guided and no planning | more accurate relative clauses with guided planning, but no significant effects on complexity and fluency |
| Mehrang and Rahimpour (2010) | 64 EFL learners | Picture (cartoon scripts) stories | 30-Sec, and 5-min unguided planning+ note taking | Increased complexity but no effects on fluency and accuracy |
| Salimi, et al. (2012) | 50 ESL learners | decision-making | 10-min unguided planning and no planning | Increased accuracy with planning |
| Abdi, Eslami, and Zahedi (2012) | 40 EFL learners | decision-making | 10-minute unguided planning and no planning | Increased fluency but no effect on accuracy. |
| Haghverdi, et al. (2013) | 90 EFL learners | narrative | 10-min pre-task planning, online planning, and no planning | Increased accuracy with pre-task planning. Pre-task planners outperformed both the groups. |
| Asgarikia (2014) | 60 EFL learners | personal and narrative |
10-min unguided planning and no planning | Reduced accuracy but increased fluency |
| Bamanger and Gashan (2015) | 52 EFL learners | two-way information gap activity | 5-min unguided planning and no planning | Increase in all the three measures: accuracy, fluency and complexity. |
| Li, et al. (2015) | 95 EFL learners | opinion-giving | nil, 30 seconds, 1 min, 2 min, 3-min, and 5-min unguided planning | Increased accuracy and fluency (with 1-minute planning time being the threshold) but no marked effects on complexity. |
| Alanazi (2016) | 36 EFL learners | picture-cued story telling | 5min guided planning, 5min unguided planning and no-planning | Minor influence on accuracy with guided planning but no effects on fluency |
| Tabari (2016) | 78 EFL learners | descriptive | 10-min unguided planning and no planning | Increased fluency but no marked effects on complexity and accuracy |
| Rahimi and Zhang (2018) | 80 EFL learners | argumentative | 10min unguided planning and no planning | Increased complexity but reduced accuracy and fluency. |
However, there are a number of studies that have demonstrated the tendency for planning to trigger more complex, but not more accurate, L2 task performance. For instance, Mehrang and Rahimpour (2010), having asked English as a foreign language (EFL) learners to narrate stories based on some cartoon scripts under planned and unplanned conditions, reported that pre-task planning triggered more complex language but there was little or no effects of pre-task planning on accuracy of the learner performances. Similar effects of planning on L2 performances were reported in slightly different settings by Crookes (1989), Ortega (1995, 1999), Philp, et al. (2006) and Wendel (1997). Crookes (1989) asked L2 learners to perform two monologic production tasks under two conditions (no planning and 10-minute planning). Yuan and Ellis (2004) worked with EFL learners, who were provided with a series of wordless pictures and asked to create a story. Ortega (1995) worked with undergraduate students (the learners of Spanish as a foreign language), who were asked to produce oral narratives (a monologic task). Later, Ortega (1999) worked with advanced level learners of Spanish as a foreign language, who were asked to perform two familiar narrative tasks (dialogic performance task) with and without pre-task planning time. Philp, et al. (2006) asked ESL learners to perform three interactive tasks under planned and unplanned conditions. Similarly, Wendel (1997) worked with EFL learners and asked the learners to perform oral narrative tasks under planned and unplanned condition. Furthermore, Rahimi and Zhang (2018), having worked with EFL learners, reported that planning increased complexity but reduced accuracy. However, some studies (e.g. Li, et al., 2015; Mochizuki & Ortega, 2008; Tabari, 2016) reported no marked effects of planning on complexity.
Similarly, previous studies on pre-task planning have produced mixed results regarding the effects of planning on fluency. While some studies found that pre-task planning produced more fluent language (e.g. Abdi, et al., 2012; Ortega 1999; Tabari, 2016; Wendel, 1997; Yuan & Ellis, 2003), others reported no effects of such planning on fluency (e.g. Alanazi, 2016; Asgarikia, 2014; Mehrang & Rahimpour, 2010; Mochizuki & Ortega, 2008).
To recapitulate, a large number of studies have manipulated pre-task planning. However, the results are not yet conclusive. One of the main reasons for producing mixed results by previous studies might be: pre-task planning was operationalized in different ways in those studies. In the current study, following Bamanger and Gashan (2015), pre-task planning was operationalized as five minutes unguided planning and no planning. Additionally, it should be noted that most of the studies reviewed above did not explore whether or not the effects of planning are the same across different tasks. Therefore, this study aimed to carry out further investigation in this area. The following section briefly reviews major studies on task type and L2 performance.
Research on Task Type Effects
Exploring the extent to which tasks and task conditions can affect the way learners process the target language has been a major focus of many task-based studies during the last few decades. Skehan (1998) believes that the extent to which learners expand the complexity of their language or attend to the accuracy or fluency of their performance is affected by task type and different conditions in which tasks are carried out. However, there are some divergences of views with regard to whether task type affects L2 performance.
There are some researchers who believe that task type does not appear to matter when comparing the linguistic gains that learners show. For instance, Rahimpour and Mehran (2010), having worked with upper-intermediate Iranian learners of English (who performed two different narrative tasks, i.e. structured vs. unstructured, based on cartoon scripts), claimed that there was no significant effect of task structure on CAF measures. Additionally, Baturay, Tokmak, Dogusoy and Daloglu’s (2011) study, which compared L2 learners’ oral performances on narrative, descriptive and prediction-personal reaction tasks in an oral exam, reported no significant differences between the three task performances in terms of oral performance scores. However, the study did not explore the task type effects on CAF measures.
However, some other researchers argue that tasks are capable of engendering predictable use of language features i.e. task variation yields variation in performance, which Ellis (1994) calls “task-induced variation” (p.135). Robinson (2011) views that task types may affect speech production in different ways; Tarone (1985) argued that when L2 learners perform different tasks, their utterances show systematic variability in some morphological and grammatical forms. Additionally, Tarone and Parrish (1988) provided evidence that narratives tend to lead L2 learners to more accurate use of articles than non-narrative tasks. The authors argued that different tasks place different degrees of communicative pressure upon the speaker which produces different outcomes. Furthermore, highlighting the point that different tasks trigger different outcomes, Bygate (1999a) argued:
Feed people with narrative tasks and they will crunch up some aspects of language in one way, sharpening certain linguistic teeth, i.e. cognitively mapping certain types of language against certain types of communicative demand. Feed them different tasks, and different linguistic teeth might develop. (p39).
Highlighting the possibility that task variation can trigger different outcomes, TBLT researchers have classified tasks in various ways. For instance, Long (1989) draws a distinction between open and closed tasks where open tasks lead to higher use of conversational gambits such as clarification requests and confirmation checks. Pica, Kanagy, and Falodun (1993) propose that language tasks can be analyzed based on the interactional activity: interactional requirements (i.e. required or optional participation) and interactional relationship (i.e. information held by different participants). These classifications indicate that tasks have several features that enable tasks to be ranked in complexity or difficulty. Thus, Skehan (1996) proposes that tasks complexity can be considered based on language factors (such as lexical complexity, syntactic complexity, redundancy and variety) and cognitive factors (such as familiarity of task type and discourse genre, familiarity of material in the task, reasoning operation required during the task performance and nature of input material used in the task). Brown, Anderson, Shilcock, and Yule (1984), propose two dimensions for task complexity level: one concerned with type of information (i.e. increasing complexity of tasks as they move from static to dynamic and abstract) and the other with scale (that concerns the number of participants or objects included in in the tasks).
Bygate (1999b), having investigated the oral performances of Hungarian secondary students in argumentative and narrative tasks with the major focus on grammatical complexity, reported that narrative tasks triggered more complex language in terms of the number of words per T-unit; but argumentation contained more ‘verbal’ output than the narration did. Rezazadeh, Tavakoli and Rasekh (2011) also found that argumentation produced more complex language, whereas instruction task produced more accurate and fluent language. Furthermore, the study by Skehan and Foster (1997) showed that personal tasks produced much more fluent language than decision-making and narrative tasks. Asgarikia (2014), having studied Iranian EFL learners’ performances, also reported that personal task produced more fluent language than narrative task, although, there was no significant difference between personal and narrative task with regard to accuracy.
To reiterate, the above literature indicates that previous research on task type effects on L2 performance is not conclusive. It should also be noted that the studies reviewed above explored the independent role of planning and task types on L2 performance, though some of them involved some sort of planning. However, as pointed out by Rahimpour and Mehrang (2010), it might be worth exploring the combined effects of task types and task planning on L2 oral performance, which is one of the aims of the current study.
Research Exploring Planning and Task Type Interaction
To the best of the author’s knowledge, only four studies (Asgarikia, 2014; Foster & Skehan, 1996; Mehnert, 1998; Tan, 2013) have explored the interaction between planning and task type effects. Foster and Skehan (1996), having asked EFL learners to perform three different tasks (narrative, decision-making, personal information exchange) under no planning, detailed planning (i.e. planning with guidance) and undetailed planning (planning without guidance) conditions, reported that planned performances were more fluent and complex than unplanned performances. Additionally, the study found interactions between task type and planning conditions, such that the effects of planning were greater with the narrative and decision-making tasks than with the personal information exchange task. However, Tan’s (2013) study on EFL learners’ narrative and argumentative task and Asgarikia’s (2014) study on EFL learners’ personal and narrative tasks under planned (10-minute planning) and unplanned conditions indicated no significant interaction between planning and task type. Mehnert (1998) also found no significant interaction between planning and task type. In the study, L2 learners were asked to perform instruction task and exposition task under one minute, five minutes and 10 minutes planning conditions. The results revealed that the participants with the longest planning time (10 minutes) produced most fluent and most complex language. However, Tan (2013) reported that planning had no effects on EFL performances, but argumentation produced slightly more fluent and complex language than narration.
To sum up, the literature on TBLT indicates that there have been several attempts to explore the effects of pre-task planning and task types on L2 performances. However, the results are not yet conclusive. Additionally, there is almost no research investigating learners’ attitudes towards pre-task planning. As Ortega (2005) rightly points out, the strategic process engaged during pre-task planning has not yet entered the stage of scholarly discussion. Consequently, there is potentially a great deal that the EFL profession does not know about EFL learners’ attitudes towards pre-task planning and the extent to which they are engaged in task planning. Thus, it is “important methodologically to document the fidelity of treatments” that is, whether participants engaged in the planning behaviors researchers expected them to engage (Ortega, 2005, p.78). Furthermore, there has been very little task-based research in the Nepalese context and no research, to the author’s knowledge, has explored the effects of pre-task planning and task type on English oral performances of post graduate level students in Nepal, a small developing country in south-Asia. Therefore, there is no evidence whether or not pre-task planning works in the Nepalese EFL context, and how the Nepalese EFL learners perceive task planning. This study’s research questions aim to assist in filling these research gaps.
Research Questions
- Are there any effects of pre-task planning on the CAF of EFL learners’ oral performances?
- Are there any effects of task types on the CAF of EFL learners’ oral performances?
- Are there any interactions between the effects of task type and pre-task planning on the CAF of EFL learners’ oral performances?
- What are the Nepalese EFL learners’ attitudes towards pre-task planning?
I hypothesize that the opportunity for pre-task planning will encourage both text and language planning in most cases. Thus, pre-task planning will trigger more fluent and more complex language. However, it is “less likely to contribute to translation, execution, and monitoring and thus may have limited effect on accuracy” (Ellis & Yuan, 2004, p.66). Nevertheless, the Nepalese students may have positive attitudes towards pre-task planning as it offers them an opportunity to be familiar with the task and also plan their performances, which may ultimately make them more confident in their own abilities needed for the tasks performance.
Regarding the task type effects, I hypothesize that the two types of task place different degrees of communicative pressure upon the students which produces different outcomes. In other words, the two tasks in this study will trigger different outcomes. Furthermore, I hypothesize that there will be no significant interaction between the effects of task type and pre-task planning on the CAF of EFL learners’ oral performances. In other words, pre-task planning may not intensify effects associated with characteristics of the two tasks used in this study. This follows from Asgarikia’s (2014) and Mehnert’s (1998) studies which did not find any kind of interaction between task types and task planning.
Methodology
Participants
The participants (N=30) in this study were post graduate students at Tribhuvan University, Nepal, earning master’s degree in English language teaching. They were randomly selected for this study and an informed consent was obtained from each of them before they participated in this study. They shared the same mother tongue (Nepali) and the average time spent studying English was 14.5 years (SD=.714) and their mean age was 23.5 (SD=2.56).
The Tasks
Two different tasks, one narrative and one opinion, were selected for this study to enable its results to be compared with previous studies (e.g. Bygate, 1999; Tan, 2013). Another reason for the tasks selection was that these tasks are commonly used in the Nepalese EFL classrooms. The narrative task, adopted from Heaton (1975), contained six pictures in the correct order (See appendix A). The story was about a boy who went swimming in the ocean but lost his clothes and how he resolved the problem by participating in a race. The task required the participants to narrate a story on the basis of the pictures. The opinion task, adapted from McDowell and Gakeman (2003), required participants to give their opinion about whether new technology has positively or negatively affected the types of relationships people form, and to support their opinion with reasons and examples (See appendix A). In each task, the participants were given 5 minutes for task completion; almost all the participants used all the time given to them. They performed both tasks on one-to-one basis in the presence of the researcher, who was just a silent listener during the planning (in the case of planned group) and the tasks performances.
+/- Pre-task Planning Time
Planning time, in this study, was interpreted as the presence or absence of time before the tasks performances. Thus, planning time was operationalized at two levels: no planning vs. pre-task planning. While the participants in the +planning condition were given five minutes for each task to plan their performance (individual planning with no guidance), the participants, under the -planning condition were allowed just 30 seconds to familiarize themselves with the tasks.
Research Design
This study employed a 2×2 between-subjects design manipulating the factors planning condition (+/- pre task planning) and narrative and argumentative tasks. The participants were randomly selected from the Department of English Education, Tribhuvan University and they were randomly divided into two groups to ensure that the two groups were equivalent in terms of their language competency. Each group performed two tasks: Narrative and Argumentative, under the same condition. For instance, a group performing an argumentative task under a +planning condition performed a narrative task under the same condition. All the participants performed the tasks individually and their voices were audio recorded.
Measures
All the data was transcribed verbatim and coded for CAF measures. Prior research indicates that numerous sub-measures have been developed for CAF (see Plonsky & Kim, 2016 for detail). In order to obtain a general as well as precise account of the linguistic properties of L2 production within each dimension of proficiency, different units of each dimension were observed following a number of previous studies (Ortega, 1995, 2005; Philp et al. 2006; Tan, 2013; Yuan & Ellis, 2003).
Complexity was measured by counting the number of sub-ordinate clauses per T-unit ( A T-unit is defined as each independent utterance providing referential or pragmatic meaning; it is made up of one simple independent finite or an independent finite clause plus one or more dependent finite or infinite clauses), words per T-unit (number of words per T-unit excluding repetitions, filled pauses, false starts, lexical replacements and reformulations), syntactic variety (number of different verb forms in terms of tense, modality and voice), and new lexical items (the type token ratio, which refers to the number of unique words (types) divided by the total number of words (tokens) in the text, was calculated using the web-based computational tool Coh Metrix).
Accuracy was measured by counting the number of error free T-units (percentage of error free T-units), error free clauses (percentage of clauses which contains no errors in terms of syntax, morphology or word order), errors per 100 words (number of any sort of errors per 100 words), and self-corrections (number of immediate grammatical correction made by the speaker). Fluency was measured in terms of the number of syllables per minute (number of syllables produced in each minute), filled pauses (number of fillers like um, uh, you know, you see, and well), repetitions (number of words, phrases or clauses that are repeated with no modification) and false starts (number of utterances that are abandoned before completion).
Procedures
All the participants were given clear instructions for the tasks completion by the researcher herself, after they consented for this study. In the planning group, students were asked to plan their performance in terms of content and organization. No further guidance was given. They were provided with a sheet of paper during planning time but they were told that these notes would be removed during task performance to ensure that the language elicited by the tasks was only oral. Furthermore, the notes provided some evidence about what the participants did during the planning time. However, in the no-planning group, students were not provided with any sheet of paper to write.
Participants in both groups performed the tasks individually in a quiet room just in the presence of the researcher. Both groups performed the narrative task first. Their speech was audio recorded. On completion of tasks, each participant was asked to complete a questionnaire which comprised several questions concerning their attitudes towards pre-task planning, the tasks used in the study, and their focus during pre-task planning and performing the tasks. Additionally, ten participants were randomly selected for a retrospective interview to assemble immediate retrospective accounts that tapped into participants’ memories regarding their pre-task planning and task performance. Their planning notes and responses to the questionnaire were used to stimulate accurate recall and help them interpret the cognitive operations they employed in task planning and performance.
Data Analysis
In order to explore the effects of pre-task planning and task type effects on each measure, a series of independent-sample t-tests were performed and post-hoc power analysis was made to observe the effect size, which is an indication of the degree to which the phenomenon under study is manifested (Cohen, 1988). Similarly, in order to examine the effects of the interaction between the two independent variables (planning and task type) on the dependent variables (CAF measures) two-way between-groups analysis of variance (ANOVA) were conducted. The normality of the data was checked before running t-tests and ANOVA; the significant value of the Shapiro-Wilk Test was greater than .05 in all the data set, suggesting that the data was normally distributed.
Results
Effects of Pre-task Planning Time
Complexity. The effects of pre-task planning on CAF measures show a mixed picture. In the case of complexity, planning led to syntactically more varied language (p<0.05), but lexical complexity remained unaffected in both tasks (p>0.05). None of the other measures reached significance (p>0.05).
Table 2. Descriptives for Complexity for the Factor +/_ Pre-task Planning
| Argumentation | |||||||
| No Planning | Planning | ||||||
| Dependent variable | M | SD | M | SD | P-value | Cohen’s d | Partial |
| Sub clauses per T-unit | 0.22 | 0.09 | 0.28 | 0.08 | 0.07 | -0.70 | 0.52 |
| Words per T-unit | 12.95 | 2.18 | 13.18 | 2.41 | 0.78 | -.10 | 0.50 |
| Syntactic variety | 39.20* | 7.38 | 47.40* | 7.04 | 0.04 | -1.13 | 0.82 |
| Type token ratio | 0.41 | 0.04 | 0.43 | 0.04 | 0.29 | -0.50 | 0.62 |
| Narration | |||||||
| No planning | Planning | ||||||
| Dependent variable | M | SD | M | SD | P-value | Cohen’s d | Partial |
| Sub clauses per T-unit | 0.18 | 0.04 | 0.21 | 0.07 | 0.09 | -0.52 | 0.38 |
| Words per T-unit | 10.21 | 1.31 | 9.70 | 0.94 | 0.23 | 0.44 | 0.50 |
| Syntactic variety | 43.00* | 6.01 | 50.80* | 10.20 | 0.04 | -0.93 | 0.65 |
| Type token ratio | 0.42 | 0.02 | 0.40 | 0.03 | 0.36 | 0.78 | 0.89 |
Note: M = mean; SD = standard deviation; * = significant at p<.05.
Accuracy. The mean percentage in all accuracy measures decreased under the planned condition in both tasks. However, the statistical analysis yielded significant differences only in terms of the percentage of error free T-units (p<0.05); and the number of errors per hundred words in both tasks (p<.05).
Table 3. Descriptives for Accuracy for the Factor +/- Pre task Planning Time
| Argumentation | |||||||
| No planning | Planning | ||||||
| Dependent variable | M | SD | M | SD | P-Value | Cohen’s d | Partial |
| Error-free T-units | 80.77* | 7.79 | 74.86* | 11.20 | 0.04 | 0.61 | 0.33 |
| Error-free clauses | 85.72 | 5.15 | 81.55 | 8.20 | 0.96 | 0.61 | 0.96 |
| Errors per 100 words | 1.97* | 0.61 | 3.92* | 1.10 | 0.01 | -2.19 | 0.99 |
| Self correction | 1.53 | 0.74 | 1.00 | 0.92 | 0.84 | 0.63 | 0.94 |
| Narration | |||||||
| No planning | Planning | ||||||
| Dependent variable | M | SD | M | SD | P-Value | Cohen’s d | Partial |
| Error-free T-units | 69.73* | 10.09 | 61.73* | 9.34 | 0.04 | 0.82 | 0.54 |
| Error-free clauses | 74.30 | 16.54 | 68.71 | 8.78 | 0.25 | 0.42 | 0.50 |
| Errors per 100 words | 2.10* | 1.66 | 4.96* | 1.17 | 0.03 | -1.99 | 0.99 |
| Self correction | 3.73 | 1.43 | 3.00 | 1.41 | 0.09 | 0.51 | 0.36 |
Note: M = mean; SD = standard deviation; * = significant at p<0.05.
Fluency. The relationship between planning and fluency is complicated. While the total number of syllables produced per minute indicated no planning effects, planning was associated with more repetitions in both tasks (p<0.05). The number of filled pauses increased with planning time in argumentation but decreased in narration (p<0.05), but the opposite picture emerged with false starts. The number of false starts sharply decreased in planned argumentation but sharply increased in planned narration; the groups in both tasks yielded significant differences (p<0.05).
Table 4. Descriptives for Fluency for the Factor +/- Pre-task Planning Time
| Argumentation | |||||||
| No planning | Planning | ||||||
| Dependent variables | M | SD | M | SD | P-value | Cohen’s d | Partial |
| syllables per min | 143.13 | 17.97 | 133.53 | 19.39 | 0.34 | 0.51 | 0.67 |
| Filled pauses | 40.53* | 18.17 | 54.93* | 14.46 | 0.02 | -0.88 | 0.48 |
| Repetitions | 53.13* | 16.92 | 64.86* | 16.54 | 0.01 | -0.70 | 0.22 |
| False starts | 7.73* | 3.03 | 4.20* | 1.54 | 0.02 | 1.47 | 0.93 |
| Narration | |||||||
| No planning | Planning | ||||||
| Dependent variables | M | SD | M | SD | P-value | Cohen’s d | Partial |
| syllables per min | 116.12 | 14.89 | 128.23 | 29.76 | 0.06 | -0.51 | 0.30 |
| Filled pauses | 60.13* | 6.27 | 43.46* | 10.37 | 0.73 | 1.95 | 0.99 |
| Repetitions | 24.06* | 10.29 | 52.13* | 10.69 | 0.04 | -2.68 | 0.99 |
| False starts | 3.80* | 1.85 | 7.00* | 3.22 | 0.01 | -1.22 | 0.71 |
Note: M = mean; SD = standard deviation; * = significant at p<0.05.
Task Type Effects
Complexity. The argumentation in general produced more complex language than the narration did. The argumentative texts were much more complex in terms of the number of sub-ordinate clauses per T-unit (t(58)=2.72, p<0.05) and the number of words per T-unit (t(58)=6.69, p<0.05). However, syntactic variety and type-token ratio remained unaffected.
Table 5. Descriptives for Task Type Effects on Complexity
| Argumentation | Narration | ||||||
| Dependent variable | M | SD | M | SD | P-value | Cohen’s d | Partial |
| Sub clause per T-unit | 0.25* | 0.09 | 0.19* | .06 | 0.01 | 0.78 | 0.28 |
| Words per T-unit | 13.06* | 2.26 | 9.95* | 1.15 | 0.01 | 1.73 | 0.97 |
| Syn. variety | 43.60 | 7.23 | 46.90 | 8.09 | 0.38 | -0.43 | 0.62 |
| Type-token ratio | 0.42 | 0.04 | 0.41 | 0.03 | 0.12 | 0.28 | 0.21 |
Note: M = mean; SD = standard deviation; * = significant at p<.05.
Accuracy. The argumentation promoted more accurate language than the narration; significant differences can be observed in all the first three measures (p<0.05). However, the participants self-corrected more in narration than in argumentation, t(58)=-6.80, p<0.05).
Table 6. Descriptives for Task Type Effects on Accuracy
| Argumentation | Narration | ||||||
| Dependent variable | M | SD | M | SD | P-value | Cohen’s d | Partial |
| Error-free T-units | 77.85* | 9.53 | 65.71* | 10.31 | 0.01 | 1.22 | 0.71 |
| Error-free clauses | 83.63* | 7.05 | 71.51* | 13.32 | 0.01 | 1.14 | 0.64 |
| Errors per hundred words | 2.99* | 1.00 | 3.51* | 1.52 | 0.02 | -0.40 | 0.10 |
| Self-corrections | 1.26* | 0.86 | 3.36* | 1.42 | 0.01 | -1.79 | 0.98 |
Note: M = mean; SD = standard deviation; * = significant at p<0.05.
Fluency. The participants were more fluent in the argumentation than in the narration; the argumentation produced more syllables per minute than the narration (t(58)=3.75, p<0.05) with less filled pauses (p<0.05). However, the argumentation was associated with more repetitions than the narration.
Table 7. Descriptives for Task Type Effects on Fluency
| Argumentation | Narration | ||||||
| Dependent variable | M | SD | M | SD | P-value | Cohen’s d | Partial |
| Syllables per minute | 138.33* | 18.01 | 122.13* | 21.21 | 0.01 | 0.82 | 0.32 |
| Filled pauses | 47.63* | 28.06 | 51.79* | 9.96 | 0.01 | -0.20 | 0.02 |
| Repetitions | 58.90* | 20.14 | 38.09* | 10.41 | 0.01 | 1.30 | 0.78 |
| False starts | 5.96 | 2.60 | 5.40 | 3.05 | 0.08 | 0.20 | 0.11 |
Note: M = mean; SD = standard deviation; * = significant at p<0.05.
Interaction Effects Between Planning and Task Types
Complexity. The interaction effects can be seen on only one of the complexity measures, i.e. the number of subordinate clauses per T-unit (F (1, 28) = 6.07, p<0.05).
Table 8. Interaction Effects on Complexity Measures
| Variables | F-value | P-value | Partial | Power |
| Sub clauses per T-unit | 6.07 | *0.01 | 0.79 | 0.82 |
| Words per T-unit | 0.62 | 0.44 | 0.02 | 0.11 |
| Syn. variety | 1.58 | 0.22 | 0.85 | 0.71 |
| Type token ratio | 3.97 | 0.17 | 0.06 | 0.26 |
* = significant at p<0.05.
Accuracy. There was no significant interaction between planning condition and task type for accuracy measures (p>0.05).
Table 9. Interaction Effects on Accuracy Measures
| Variables | F-value | P-value | partial | power |
| Error-free T-units | 1.32 | 0.26 | 0.06 | 0.38 |
| Error-free clauses | 0.07 | 0.79 | 0.01 | 0.39 |
| Errors per 100 words | 0.13 | 0.73 | 0.02 | 0.06 |
| Self- corrections | 0.11 | 0.74 | 0.02 | 0.52 |
* = significant at p<0.05
Fluency. The three fluency measures (the number of syllables per minute, the number of repetitions and false starts) indicate a statistically significant interaction between the planning condition and task type (p<0.05). However, a statistically significant interaction was not attained for the number of filled pauses (p>0.05).
Table 10. Interaction Effects on Fluency Measures
| Variables | F-value | P-value | Partial | Power |
| Syllables per minute | 9.51 | *0.00 | 0.67 | 0.87 |
| Filled pauses | 0.03 | 0.99 | 0.02 | 0.15 |
| Repetitions | 5.47 | *0.02 | 0.53 | 0.69 |
| False starts | 16.71 | *0.00 | 0.50 | 0.99 |
* = significant at p<0.05.
EFL Learners’ Attitudes Towards Pre-task Planning
In order to understand participants’ attitudes towards pre-task planning and the cognitive operations involved in planning and task performance, each participant was asked to respond to a questionnaire and 10 of them also took part in a retrospective interview. The participants’ own attitudes towards pre-task planning are considered to be important as they might help the researcher to find out how and why planning worked or did not work (Kawachi, 2005).
It was found that the majority of participants (23 out of 30) had positive attitudes towards planning. Being able to jot down ideas and organize contents were the two distinct benefits afforded by pre-task planning:
I found pre-task planning very much helpful for me. It helped me to remember important points necessary for the task completion. Additionally, I could plan about how to present the points in a correct order (Participant C).
Pre-task planning offered me an opportunity to organize my thoughts (Participant F).
During planning, I could list all the important information related to the question (Participant B).
Half or more (55%) also thought that planning provided them with an opportunity to assess task demands and prioritize their attention accordingly (Ortega, 1999).
I was really lost when I first read the question. I could not understand what kind of information had to be included in my answer; planning time helped me to understand the task better (Participant A).
Pre-task planning helped me to understand the task demands and focus only on important points needed for the task performance (Participant D).
However, it was found that the participants’ appraisals of the putative benefits of pre-task planning fell into three patterns.
The majority of participants (24 out of 30) stated that planning definitely helped them organize their thoughts, retrieve lexis and grammar, which led to better performance. A smaller proportion (4 out of 30) was ambivalent in their appraisal. Finally, very few participants (2 out of 30) also reported that availability or lack of time did not make a difference to their task performance. They were skeptical about the advantages of planning; therefore, they reported that planning time made no difference because, even with planning, they still could not really understand what they had not understood in the first performance: “I am not sure whether or not planning helped me to perform the task better (Participant J); “Pre-task planning did not help me much as I could not understand the link between the third and fourth pictures even after the planning time was over (Participant E).” However, the majority of participants (26 out of 30) indicated that they would prefer to have pre-task planning when they have to perform language tasks.
All the participants were found to have maximally utilized their planning time. In the narrative tasks, almost all the participants (13 out of 15 ) spent five minutes understanding the content, outlining the main events, and retrieving lexis and grammar. However, two participants, who did not make any notes during planning, reported that they just tried to see the linkage between the pictures. Likewise, in the argumentative task, most of the participants (12 out of 15) engaged themselves in establishing the main ideas, and collecting some specific examples and reasons to support their views. However, three of the participants seemed just to retrieve lexis during planning and none seemed to try to rehearse speech in either task.
During tasks performance, very similar to Kawachi’s (2005) findings, some sort of divergence in task approach was found. The majority of participants (23 out of 30) were inclined to prioritize communication, while a few others (4 out of 30) expressed their natural predisposition towards accuracy in a very candid way and mentioned that they were worried about making mistakes. Furthermore, just three participants prioritized complexity. Interestingly, none tried to balance the CAF dimensions in their speech. It was also revealed that most of the participants noticed their errors, but none tried to correct them as they did not have time to edit their production: “I did not correct my errors as I did not have enough time to complete my story. I was just trying to speak as fast as I could” (Participant E); “I could notice some mistakes in my speech but I did not correct them as I thought that I would not finish the story within the given time” (Participant G).
Additionally, a lack of transfer of planning to performance was reported by some participants: “I could not remember many important points that I had noted down during my planning time. I just forgot them” (Participant B); “When I started to tell the answer, I was a bit nervous. So, I could not remember what I had planned to say” (Participant F).
Discussion
Task Planning
With respect to the first research question (i.e. Are there any effects of pre-task planning on the CAF of EFL learners’ oral performances?), the results revealed varied outcomes. Pre-task planning seems to affect all three CAF measures, but in different ways. With regard to complexity, it replicates the findings of most previous studies (Crookes, 1989; Foster & Skehan, 1996; Mehnert, 1998; Ortega, 1995; Philp, et al., 2006; Yuan & Ellis, 2004;) as planned performances were syntactically more varied than unplanned ones. However, unlike Yuan and Ellis’s (2004) claim that planning triggers lexically more varied language, lexical complexity in this study remained unaffected. Additionally, this finding does not support Crookes’ (1989) and Ortega’s (1995) conclusion that planning can lead L2 learners to produce more developed speech in the short term. It might indicate that planning condition did not focus on detailed lexical searches. The notes collected from planning groups also indicated that the students did not search for lexical items as most of the notes contained very few lexical items and most of them were verbs in past forms. Thus, it can be assumed that planning was presumably accompanied by a search for verb forms and necessary patterns to express meaning. Most of the interviewed participants also reported that their main focus was on grammar during planning time. The lack of pre-task planning effects on lexical variety might also reflect the fact that they prioritized lexical search during performance. Levelt’s (1989) speaking model, which describes speech production as an autonomous process and comprised three overlapping stages, conceptualization (which provides general knowledge and discourse knowledge as an input to the next stage), formulation (which translates the conceptual representation into linguistic structures) and articulation (which transforms linguistics structures into actual speech), also claims that speakers prioritize conceptualization over formulation and articulation.
In the case of accuracy, the planning group produced less accurate language. The findings are consistent with a number of previous planning studies (Crookes, 1989; Ortega, 1999; Yuan & Ellis, 2004). However, the findings do not support Mochizuki and Ortega’s (2008) finding that planning leads to more accurate language. Nevertheless, it should be noted that, unlike in Mochizuki and Ortega’s study, the participants in the current study were not guided during planning time. Thus, discrepancies in the results of the two studies might have resulted from the presence or absence of guidance to the students during their planning time. Additionally, three other possible explanations for this reduced accuracy can be made. First, it could be associated with the utilization of pre-task planning time. The majority of participants in this study reported that they were mainly concerned with information organization during planning. Additionally, the notes collected from the planning group contained some lexical items such as first, second, and finally, and those words were mostly followed by a few phrases or sentences. This indicates that when learners are allowed to plan, they choose to focus on what they want to say rather than planning grammatical forms. Second, the results in the study are inconsistent with Ortega’s (1999) finding that most L2 learners used rehearsal strategies during planning; none of the interviewed participants in this study reported that they rehearsed their speech. Third, the participants in both groups were under pressure to complete the tasks rapidly. This might have restricted their on-line planning causing them to have prioritized meaning over accuracy (Yuan & Ellis, 2004). Some participants also reported that they overlooked their errors because of the time constraint:
When I was telling the story, I tried my best to speak as fast as possible. I did not even pay attention to my grammar mistakes. I just kept on speaking (Participant A).
I knew that I had to speak fast to complete the story on time. So, I did not try to correct my errors, though I noticed some in my speech (Participant D).
With regard to fluency, consistent with Mehrang and Rahimpour’s (2010) finding, participants under the planned condition were no more fluent than the participants under no planning condition. However, the results did not corroborate previous research findings on task planning (e.g. Foster & Skehan, 1996; Mehnert, 1998; Ortega, 1995, 1999). Two explanations can be offered to account for this discrepancy. One reason could be the time factor, as both pre-task planners and no-planners were required to perform the tasks under time pressure. However, the participants in most previous studies were allowed to take as long or as little time as they wanted. As claimed by Yuan and Ellis (2004), the need to complete the tasks within a limited time and to produce a minimum number of sentences for each picture in the narration and a minimum number of examples and reasons to support their opinion in the argumentation might have caused the no-planners to speak more rapidly than they might have done, had they been free to perform the tasks in their own time. Another possible explanation could be the nature of unguided planning which might mitigate the desired effects of planning on the participants.
Furthermore, the effects of planning seem to be somehow similar to all EFL learners, irrespective of their linguistic level. The participants in the study were earning master’s degree in ELT (so they were supposed to be advanced level learners) ,but the results in the study were largely similar to those studies that worked with participants at a lower level (e.g. Ortega, 1995; Philp, et al., 2006).
In general, planning seems to affect complexity positively, accuracy negatively but it seems to have limited effects on fluency. In other words, this study finds that the primary competition is between complexity and accuracy; complexity is promoted to the detriment of accuracy.
Task Types
In order to address the second research question (i.e. ‘Are there any effects of task types on the CAF of L2 learners’ oral performances?’), the participants were asked to perform two different tasks: narrative and argumentative tasks. The two types of task, like in most previous studies (Bygate, 1999; Foster & Skehan, 1996; Mehnert, 1998; Rezazadeh, et al., 2011; Tan 2013), pushed learners in different directions; their effects could be observed in all three dimensions (CAF), with argumentation outperforming narration. In the case of complexity, argumentation produced syntactically more complex language. The numbers of sub-ordinate clauses and words per T-unit were significantly higher in argumentative texts. However, lexical complexity was not affected. With regard to accuracy, striking differences could be observed between the two tasks. The argumentative task produced far more accurate language than the narrative one, although the participants self-corrected more in the narration. A somewhat similar picture emerged for fluency measures. The argumentative task produced more fluent speech than the narrative one in terms of the speed fluency measure (the number of syllables produced per minute) although it involved more repetitions.
Thus, the findings provide support to the claim made by Bygate (1999a) and Skehan (1998) that the extent to which learners expand the complexity of their language or attend to the accuracy or fluency of their performance is affected by the nature of the task. In other words, the language variation indicated in this study might have resulted from the demand and the purposes of the two tasks along with the cognitive load and clarity of tasks goals. It has been widely argued that different tasks generate different patterns of language use (Foster & Skehan, 1996; Long, 1989; Pica, et al., 1993; Skehan, 1996).
The findings of the study are consistent with most previous studies as the results indicated tasks type effects on EFL oral performance. However, the results in terms of CAF measures, in particular, are different. For example, contrary to Tarone and Parish’s (1988) claim that narratives produce more accurate language than non-narratives, the narrative task in the current study produced less accurate language. Similarly, in the study by Bygate (1999b), narration produced more words per T-unit than argumentation ,but in this study, argumentation produced more words per T-units than narration, even though the tasks used in the two studies were similar, and also task planning was operationalized in a similar way. However, the findings are in line with Tan’s (2013) claim that argumentation produce more complex and fluent language than narration.
Argumentation seems significantly more effective than narration in promoting L2 performances. This finding may be accounted for by a task-demand-based explanation, namely, that the demands of the argumentative task for elaborated content may push learners to produce more words and more complex structures. However, the finding does not support Skehan’s (1996) claim that narratives have lower cognitive and linguistic demands than argumentative task as the majority of participants through questionnaire and post task interviews reported that the argumentative task was less complex to them than the narrative task. They gave two main reasons for their difficulty with the narration: it required fixed grammatical patterns (past tense), and they needed some sort of imagination to create a linkage between the pictures as the pictures did not seem to have clear connections. Additionally, the finding does not support Foster and Skehan’s (1996) another claim that most difficult task produces most complex language as the narratives (more difficult task) produced less complex language in the study.
There are two tentative explanations for the task type effects observed in this study (i.e. argumentation outperformed narration in all the three measures). First, students’ performances in this study might have been affected by the order of presenting the students with the tasks. Specifically, giving students the narrative task (i.e. more complex task) first and then the argumentative task (i.e. less complex task) would negatively affect students’ performances. In other words, had the tasks been presented in a simple to complex order, this might have led to different outcomes. Second, their performances might have been affected by their task familiarity. Dawadi (2019) argues that students perform better in a familiar task than in an unfamiliar task. The post-task interviews indicated that students were more familiar with argumentative tasks than with picture narratives. Nevertheless, the findings of this study suggest that when L2 tasks do not require learners to express complex ideas, learners attempt to achieve all the three aspects of speech performance (CAF).
Planning and Task Type Interaction
The third research question probes the interaction between planning and task type, i.e. it investigates “whether the two conditions taken together generate results which could not be accounted for by taking them singly” (Skehan & Foster, 1997, p. 200). It is worth pointing out that the two independent variables (task type and planning condition) interact if the effect of one of the variables differs depending on the level of the other. In other words, the effect of planning condition differs as a function of task type, or the effects of task type differ as a function of the planning condition.
Consistent with Foster and Skehan’s (1996) findings, this study does provide statistical evidence indicating an interaction between planning and task types. Among the four measures of fluency, significant interaction could be observed on three measures (i.e. the number of syllables per minute, repetitions, and false starts). In the case of complexity, significant interaction could be observed on a single measure, i.e. the number of sub-ordinate clauses per T-unit. However, none of the accuracy measures indicated significant interaction between planning and task type effects.
One explanation for these results would be the effects of the participants’ attitude towards accuracy (i.e. allotting less value to it than fluency and complexity) during their task performance. The majority of the participants reported that they had the greatest focus on fluency and least on accuracy when performing both the tasks. However, in comparison to the argumentative task, they had more focus on grammatical forms, particularly the verb forms, when performing the narrative task. Therefore, it seems important to consider learners’ attitudes as moderating factor when making predictions with regard to the benefits of providing planning time for L2 production.
Another possible factor could be the participants’ knowledge of the tasks. Most of the participants reported that they were more familiar with argumentation than with narration. With pre-knowledge of the task, participants would direct more of their attentional resources to linguistic form and content while planning and performing the tasks. Additionally, the majority of participants reported that they utilized planning time differently for the two different tasks, i.e. they tried to see a linkage between pictures to outline the main events and retrieve lexis and grammar for the narrative task, but they engaged themselves in establishing the main ideas, and collecting some specific examples and reasons to support their views for the argumentative task. However, most of them also reported that they could not transfer what they had planned into their performance despite the fact that they had positive attitudes towards planning. Low level of transfer from planning to performance might indicate the need for further exploration.
EFL learners’ attitudes towards pre-task planning
Post-task interviews and questionnaire data revealed that most of the learners had positive attitudes towards pre-task planning as they seemed to believe that pre-task planning helped them collect important points and also organize their ideas. Many participants reported that pre-task planning is necessary to make them feel confident when performing a task in English. However, such benefits of task planning did not seem to be experienced by every participant. A few participants did not think that pre-task planning helped them to improve their language. One of the main reasons behind this might be that they were not guided or helped during the planning time. This finding might suggest that learners should be provided with some guidelines to maximize their learning from pre-task planning.
Implications of the study
This study has some pedagogical implications, despite the fact that it is restricted to an exploration of language produced during task performance. The findings suggest that analyzing task content is important because tasks can have different effects on L2 performance. The study also reinforces the claim that teachers’ choice of task should be accountable to theory and research findings (Skehan & Foster, 1997). There seems to be a clear role for teachers to ensure that good tasks are selected for classroom use ,and pre-task planning is carried out properly to promote L2 performance. It is hoped that the findings will provide classroom teachers with some guidelines on how to design a task and implement it in such a way that task implementation supports L2 development. In other words, language teachers might be encouraged to explore task implementation strategies, which can push learners to more sophisticated, accurate and fluent uses of their L2.
Limitations of the study
The study has revealed varied outcomes with regard to planning and task type effects on EFL oral performance. However, they are certainly not conclusive and comprehensive. Some limitations should be acknowledged in this research.
The first limitation considers students’ proficiency in the English language. In the study, students were randomly divided into two groups. However, a standardized measure of students’ English proficiency should be available, not assumed, to be sure that there is a reasonable degree of skill similarity among subjects. The second limitation concerns its sample size as this study was limited to 30 Nepali EFL learners, earning master’s degree in ELT. This may limit the generalization of the results to other circumstances. Future research in this area with a larger sample obtained from the same or similar population or learning conditions is recommended to validate the results of the current study. In addition, more research is needed to obtain an all-inclusive picture of the relationships between pre-task planning, task types and language outcomes in various cultural and geographical settings, thereby allowing future cross-cultural comparisons. Moreover, this study is limited to EFL oral performance and hence generalization of the findings to other language skills is limited. Thus, future research is needed to examine the effects of both pre-task planning and task types on other language skills using various tasks.
Thus, the extent to which the results obtained can be generalized to other learners with different proficiency levels and different learning contexts remains to be investigated. It should also be noted that this study has identified only some immediate effects of planning on L2 performance. As Mochizuki and Ortega (2008) suggest, longitudinal studies are needed to examine whether the benefits obtained from strategic planning at one time could be extended to a later time. Furthermore, this study has just measured linguistic and oral performance complexity, not ideational (conceptual or thought) complexity when performing those L2 tasks. Therefore, future research in this area has been recommended. There is a need to collect data on CAF from students in their L1 to determine what CAF relationships exist, and if they are similar in their L2. Perhaps, the variation in this study and other studies are due to basic participant differences in planning behaviors and oral expression. Finally, there is a need to investigate if the order of tasks presentation affects students’ performances.
Conclusion
The study has probed the nature of task-based EFL performance. It investigated four major issues associated with task based EFL performance: a) the effects of planning on EFL performance, b) the effects of task type on EFL performance, c) the interaction effects of planning and task type on EFL oral performance and d) EFL learners’ attitudes towards task performance. The findings indicate that task planning and task type can have effects on the nature of EFL performance and there are some interaction effects of planning and task types on EFL performance. However, it would be wrong to draw the conclusion that EFL learning and performance can be fully predicted on the basis of task characteristics and/or task implementation. What seems nearer the truth is that task-based “language use and development is a continual balance between the emergence, elaboration and exploitation of routines on the one hand, and ad hoc variation and creativity on the other” (Bygate, 1999b, p. 209). The patterning in the data discussed in this article suggests that this position is approximately correct. Furthermore, a trade-off relationship between CAF dimensions has been observed in this study. There is a disagreement regarding the dimensions involved in the trade-off. However, this study claims that gains in complexity are offset by losses in accuracy (Bygate, 2001) when the results for planning effects are considered. It is also worth pointing out that EFL learners prefer to have pre-task planning as they feel more confident when they have planning time prior to the task performance.
About the Author
Saraswati Dawadi is currently a PhD scholar at the Open University, England. She has earned an MA: TESOL from Lancaster University, England, as a Hornby Trust Scholar 2013/2014. Prior to starting her study in England, she was a lecturer at Tribhuvan University for eight years. Her interest sits broadly within language assessment, second language acquisition and English language teaching.
References
Abdi, M., Eslami, H., & Zahedi, Y. (2012). The impact of pre-task planning on the fluency and accuracy of Iranian EFL learners’ oral performance. Procedia – Social and Behavioral Sciences, 69, 2281 – 2288.
Alanazi, S. M. (2016). The impact of pre-task planning on the oral production of Saudi EFL learners. Journal of Humanities and Social Science, 21(2), 69-73.
Asgarikia, P. (2014). The effects of task type, strategic planning and no planning on written performance of Iranian intermediate EFL learners. Procedia -Social and Behavioral Sciences, 98, 276–285.
Bamanger, E. M. & Gashan, A. K. (2015). The effect of planning time on the fluency, accuracy, and complexity of EFL learners’ oral production. Journal of Educational Sciences, 27(1),1-15.
Baturay, M. H., Tokmak, H. S., Dogusy, B., & Daloglu, A. (2011). The impact of task type on oral performance of English language preparatory school students. Journal of Education, 49(1), 60-69.
Brown, G., Anderson, A., Shilcock, R., & Yule, G. (1984). Teaching talk: Strategies for production and assessment. Cambridge: Cambridge University Press.
Bygate, M. (2001). Effects of task repetition on the structure and control of language. In M. Bygate, P. Skehan, & M. Swain (Eds.), Researching pedagogic tasks: Second language learning, teaching and testing (pp. 23 – 48). London: Longman.
Bygate, M. (1999a). Task as context for the framing, reframing and unframing of language. System, 27 (1), 33-48.
Bygate, M. (1999b). Quality of language and purpose of task: Patterns of learners’ language on two oral communication tasks. Language Teaching Research, 3(3), 185-214.
Bygate, M., & Samuda, V. (2005). Integrative planning through the use of task-repetition. In R. Ellis (Ed.), Planning and task performance in a second language (pp. 37-74). Amsterdam: John Benjamins.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd Ed.). Hillsdale: Lawrence Erlbaum.
Crookes, G. (1989). Planning and inter language variation. Studies in Second Language Acquisition, 11(4), 367-383.
Dawadi, S. (2019). The effects of task repetition on EFL oral performances. ITL-International Journal of Applied Linguistics, 170(1), 3-23.
Ellis, R. (2017). Position paper: Moving task-based language teaching forward. Language Teaching, 50(4), 507-526.
Ellis, R. (2005). Planning and task-based performance: Theory and research. In R. Ellis (Ed.), Planning and task performance in a second language (pp. 3-34). Amsterdanm: John Benjamins.
Ellis, R. (1994). The study of second language acquisition. Oxford: Oxford University Press.
Ellis, R. (1987). Inter language variability in narrative discourse: Style shifting in the use of the past tense. Studies in Second Language Acquisition, 9(1), 1-20.
Ellis, R. & Yuan, F. (2004). The effects of planning on fluency, complexity, and accuracy in second language narrative writing. SSLA, 26 (1), 59–84.
Foster, P., & Skehan, P. (1996). The influence of planning and task type on second language performance. Studies in Second Language Acquisition, 18 (03), 299-323.
Haghverdi, H. R., Biria, R., & Khalaji, H. R. (2013). The Impact of Task-planning and Gender on the Accuracy of Narrations Composed by Iranian EFL Learners. Journal of Language Teaching and Research, 4(1), 74–83.
Heaton, J. (1975). Beginning composition through pictures. London: Longman.
Kawauchi, C. (2005). The effects of strategic planning on the oral narratives of learners with low and high intermediate L2 proficiency. In R. Ellis (Ed.), Planning and task performance in a second language. (pp. 143–164). Amsterdam: John Benjamins.
Levelt, W. (1989). Speaking: From intention to articulation. Cambridge, MA: Massachusetts Institute of Technology Press.
Li, L., Chen, J. & Sun, L. (2015). The effects of different lengths of pre-task planning time on L2 learners’ oral test performance. TESOL Quarterly, 49(1), 38-66.
Long, M. (1989). Task, group, and task-group interaction. University of Hawaii Working Papers in English as a Second Language, 8(20), 1-26.
McDowell, C. & Gakeman, V. (2003). Cambridge practice test for IELTS one. Cambridge University Press: Cambridge.
Mehnert, U. (1998). The effects of different length of time for planning on second language performance. Studies on Second Language Acquisition, 20(1), 83-108.
Mehrang, F., & Rahimpour, M. (2010). The impact of task structure and planning conditions on oral performance of EFL learners. Procedia Social and Behavioral Sciences, 2(1), 3678–3686.
Mochizuki, N. & Ortega, L. (2008). Balancing communication and grammar in beginning-level foreign language classrooms: a study of guided planning and relativization. Language Teaching Research, 12(1), 11-37.
Ortega, L. (2005).What do learners plan? Learner driven attention to form during pre-task planning. In R. Ellis (Ed.). Planning and task performance in a second language (pp.77-10). Amsterdam; Philadelphia: John Benjamins.
Ortega, L. (1999). Planning and focus on form in L2 oral performance. Studies in Second Language Acquisition, 21(1), 109-148.
Ortega, L. (1995). The effect of planning in L2 Spanish oral narratives (Research Note No. 15). Honolulu: University of Hawai.
Philp, J., Oliver, R., & Mackey, A. (2006). The impact of planning time on children’s task-based interactions. System, 34(4), 547-565.
Pica, T., Kanagy, R., & Falodun, J. (1993). Choosing and using communicative tasks for second language instruction. In S. Gass & G. Crookes (Eds.), Tasks and language learning: Integrating theory and practice (pp. 9-34). Clevedon, UK: Multilingual Matters.
Plonsky, L., & Kim, Y. (2016). Task-based learner production: A substantive and methodological review. Annual Review of Applied Linguistics, 36, 73-97.
Rahimi, M., & Zhang, L. J. (2018). Effects of Task Complexity and Planning Conditions on L2 Argumentative Writing Production. Discourse, 55 (8), 1-17.
Rezazadeh, M., Tavakoli, M. & Rasekh, A. E. (2011). The role of task type in foreign language written production: Focusing on fluency, complexity, and accuracy. International Education Studies, 4(2), 169-176.
Robinson, P. (2011). Task-based language learning: A review of issues. Language Learning, 61(1), pp. 1–36.
Robinson, P. (2005). Cognitive complexity and task sequencing: Studies in componential framework for second language task design. International Review of Applied Linguistics in Language Teaching, 43(1), 1-32.
Salimi, A. Alavinia, P. & Hosseini, P. (2012). The Effect of strategic planning time and task complexity on L2 written accuracy. Theory and Practice in Language Studies, 2(11), 2398-2406.
Skehan, P. (2009). Modeling second language performance: Integrating complexity, accuracy, fluency, and lexis. Applied Linguistics, 30(4), 510-532.
Skehan, P. (1998). A cognitive approach to language learning. Oxford: Oxford University Press.
Skehan, P. (1996). A framework for the implementation of task-based instruction. Applied Linguistics, 17(1), 38-62.
Skehan, P., & Foster, P. (1997). Task type and task processing conditions as influences on foreign language performance. Language Teaching Research, 1(3), 1-27.
Tabari, M. A. (2016). The effects of planning time on complexity, accuracy, fluency, and lexical variety in L2 descriptive writing. Asian-Pacific Journal of Second and Foreign Language Education, 1(10), 1-15.
Tan, Y. (2013). The effects of planning condition and task type on the complexity, accuracy and fluency in EFL learners’ writing performance. An unpublished MA thesis, England: Lancaster University.
Tarone, E. E. (1985). Variability in interlanguage use: A study of style‐shifting in morphology and syntax. Language Learning, 35(3), 373-403.
Tarone, E. & Parrish, B. (1988). Task-related variation in interlanguage: The case of articles. Language Learning, 38(1), 21–44.
Yi, B. & Ni, C. (2015). Planning and working memory effects on Le performance in Chinse learners’ argumentative writing. Indonesian Journal of Applied Linguistics, 5(1), 44 –53.
Yuan, F. & R. Ellis (2003). The effects of pre-task and on-line planning on fluency, complexity and accuracy in L2 monologic oral production. Applied Linguistics, 24(1), 1–27.
Wendel, J. (1997). Planning and second language narrative production. Unpublished Doctoral Dissertation, Japan: Temple University.
Appendix A: Tasks Used in the Research
a. Argumentative task
Nowadays, the way many people interact with each other has changed because of technology. In what ways has technology affected the types of relationship people make? Has this become a positive or negative development? Give minimum three reasons and examples to support your answer.
b. Narrative task
Appendix B: Transcription of a Participant’s Story (First One Minute) and Calculation of CAF Measures.
Well, once upon a time, there was a small boy, who lived in a small and beautiful village. One day, um the boy made a secret plan to swim for a long time. Um there was a big river in his village. In the afternoon, he went to to the river alone for swimming. Uh he took off his clothes there. His clothes, then, he put his clothes in a rock there. After this, he he started to swim in the river, which was very big. Um um he enjoyed swimming a lot. He swimmed in the river in the river until he was too tired. But, when when he came out of the river, he did not found uh find his clothes there.
| Measures | numbers | Remarks | |
| complexity | No of sub-ordinate clauses per T-unit | 0.40 | Completion of each T-unit is indicated by a star. |
| Words per T-unit | 10.9 | There are 109 words excluding repetition, filled pauses, false starts and self-correction | |
| Syntactic variety | 14 | Verb forms | |
| Type-token ratio | – | It was calculated using Coh Metrix. | |
| accuracy | Percentage of error free T-units | 80% | Out of 10 T-units, eight are correct. |
| Percentage of error free clauses | 85.72% | Out of 14 clauses, 12 are correct. | |
| No of self-correction | 1 | Self-correction is in purple font. | |
| Errors per 100 words | 2 | Incorrect words are in green font. | |
| fluency | Number of syllables per minute | 133 | Excluding filled pauses and repetitions |
| Number of filled pauses | 7 | Filled pauses are in blue font | |
| Number of false start | 1 | False start is in yellow font | |
| Number of repetitions | 4 | Repetitions are in red font |
| Copyright rests with authors. Please cite TESL-EJ appropriately. Editor’s Note: The HTML version contains no page numbers. Please use the PDF version of this article for citations. |