Investigating the Washback Effects of Task-based Instruction on the Iranian EFL Learners’ Vocabulary Learning

The current research was an attempt to explore the washback impact of task-based instruction (TBI) on EFL Iranian learners’ vocabulary development. To this end, conducting an Oxford Placement Test (OPT), 30 out of 72 EFL Iranian learners studying in an English language institute, were randomly selected. Then, they were assigned to experimental (N=15) group, and control group (N=15). The experimental group of the study was taught through TBLI and the control group of the study was instructed the vocabulary via conventional teaching instruction for 12 ninety-minute sessions. The class sessions were held twice a week. The data was examined through independent sample t-test. The findings revealed that there was significantly different between two groups’ performance in the posttest, thereby TBI had a significant effect on developing Iranian EFL learners’ vocabulary development. Furthermore, some pedagogical implications for EFL instructors and learners as well as syllabus designers have been presented.


Introduction
Vocabulary learning is one of the crucial components for improving effective communication skills, and it has been regarded as a central part of language by both researchers and practitioners.Many researchers working on vocabulary acquisition have claimed that acquiring a word requires myriads encounters with the word in different forms (Horst, Cobb, & Meara, 1998;Nation, 1990;Schmidt, 2001).Vocabulary is an important component of a language as well as the basis of communicating domain in L1 and L2 contexts.It is one component that amalgamates four skills of listening, speaking, reading and writing all together.Teaching and learning EFL/ESL vocabularies is also one of the main obstacles for foreign language students encounter while learning a new language.One way to simplify this time-consuming activity and to accelerate the development of vocabulary capacity is to assist students in recognizing the role and function of cognitive tasks during the process of L1 and L2 vocabulary learning (Hatch & Brown, 1995).
According to Shepard, Kagan and Wurtz, (1998), assessment and teaching have always been regarded to be more inseparable acquisition processes in the childhood than other period of life.As Bachman (1995) argues, assessment and testing are important dimensions of every teaching and learning practice.They provide teachers with the necessary information about their students' language ability and enable them to make sound professional judgments and decisions within the context of their classes.
Accordingly, in the fields of education and applied linguistics it is widely believed that testing affects learning and teaching process.This influence is called 'washback' (Alderson & Wall, 1993), 'backwash' (Hughes, 1993, cited in Bailey 1996), or 'test impact' (Bachman & Palmer, 1996).According to Eckstein and Noah (1993), a number of washback studies have highlighted the positive or negative outcomes of standardized tests, however, washback exists in any type of assessment including Task-based language assessment (TBLA) in which test results affect test-takers' future course of development and learning , and thus are regarded as high-stake tests.

Statement of the Problem
One branch of foreign language education, which seems to receive not much attention, is the innovation of differentiated instruction and creative assessment in the foreign language classroom.The best way of teaching and testing generally and in EFL/ESL settings particularly, have always obsessed the mind of many scholars and teachers.Finding a perfect methodology suitable for both teaching and testing of foreign languages, has been an unaccomplished wish for which there is a long way to be paved.
Consequently, in the recent decades, the attitudes towards assessing students have been changed dramatically.The utilization of assessment as a means of improving curriculum has become mainly common not only in general education (Chapman & Snyder, 2000), but also in language education (Alderson & Wall, 1993;Cheng, 1997) and its different aspects including the assessment paradigms.Considering this problems and difficulties, this study was set to use task-based assessment as an assessment and washback effect teaching tool for enhancing and developing vocabulary learning.

Review of the Literature
Luke Prodromou (1995) defines wash back as the direct and indirect effect of examination on teaching methods.Alderson and Wall (1993): maintain that "The term wash back is itself a neutral one, can be related to 'influence'.If the test is poor, then the wash back may be felt to be negative.But if the wash back hypothesis hold, the good tests should have good effects (as yet undefined) rather than negative effective effects" (p.117).In Iranian EFL context, Tavakoli and Skehan (2005) introduced a model of TBLI in relation to language testing.They tried to show that the rating assigned to the learners on the basis of their performance on a task is the result of a whole range of elements, only one of which can possibly be their underlying competence.
In a study, Rulon and McCreary (1986) compared the negotiation for meaning in a teacher-fronted group and a task-based teaching group.The results proved that because in TBLT and group work learners focused on meaning, interaction was stimulated that resulted in development of L2 learning.In another study conducted by Fotos and Ellis (1991) in Japan, TBLT was applied to communicate about grammar.The results showed effectiveness of this approach in both successful learning and communication.Moreover, communicative grammar-based tasks facilitated the acquisition of implicit knowledge in Japanese college-level EFL learners.In addition, Pica et al. (1996) showed that negotiation for meaning might be effective if accompanied with other pedagogical principles that promote language acquisition.Studies of comparison between interactive and monologic tasks by Skehan and Foster (1997) indicated that the interactive task produces much more precision while the monologic task results in more fluency.Foster andSkehan (1996, 1999) and Mehnert (1998) conducted other studies on investigation the impact of task planning and found that task planning results in positive impact on accuracy and complexity of learner's performance.Foster and Skehan (1996) investigated the impact of meaning/form-focused strategic planning, undetailed strategic planning, and minimal strategic planning on EFL learners' speech.The concluded that that the meaning/form-focused strategic planning condition made speech complex and fluent speech than the undetailed strategic planning condition.

Research Question and Hypothesis
The present study was set answer the following question: RQ.Is there any significant difference between the traditional assessment and TBLA concerning their washback effect on EFL learners' vocabulary development?HO.There is no significant difference between the traditional assessment and TBLA concerning their washback effect on EFL learners' vocabulary development.

Participants
Conducting an Oxford Placement Test (OPT), 30 out of 72 EFL Iranian learners studying in English a language institute, were randomly selected.Then, the the participants were assigned to experimental (N=15) group, and control group (N=15).They ranged between 19 to 24 years old and all of them are Persian native speakers.

Instrumentations
The instruments consisted of an Oxford Placement Test (OPT), a vocabulary test as pretest, and a posttest.First, in order to check the level of general language proficiency of the participants at the beginning of the study and find out a homogenous sample, an OPT was utilized.Second, a standard vocabulary pretest was used as the other required instrument in this study.The aim was to determine whether participants are homogeneous in their knowledge of English language vocabularies.This vocabulary test consisted of 80 multiple-choice test items.To be statistically acceptable, both the internal consistency and reliability co-efficient was examined through Cronbach Alpha bearing evidence to the test's reliability upon piloting.Finally, another vocabulary test as the posttest was administered.This test consisted of 80 vocabulary multiple choice questions.The only difference between this test and the pretest was that the sequence of the items had been changed in order to avoid "practice effect" (Bachman, 1990) on the part of the participants.

Research Design
The method adopted in this study was quantitative research in nature.Since the subjects taken from the intact language classes, their selection violates the criteria of randomization so quasi-experimental design was utilized in the way that there were the experimental and the control groups based on convenience-random sampling.There were also a pretest and a posttest for all the groups and a treatment period.In fact, the research design included pretest-treatment-posttest.

Data Collection Procedure
Prior to the study, all of the participants in the experimental and control groups were evaluated for their general English proficiency.Afterwards, the extent of their skill in English language vocabulary was questioned by using a standard pre-test administered to the students at the beginning of the semester one week before the beginning of instruction.The purpose of pretest was to establish both groups' homogeneity on their knowledge of vocabularies before the inception of the experiment, and later to use the pretesting information for later comparisons with the posttest.After administration of OPT general proficiency and vocabulary pretest, 15 students were assigned to the experimental group, and 15 students were included in the control group, while, the latter receiving traditional lecture-discussion instruction.Both experimental and control groups were independently taught the vocabulary for 12 ninety-minute sessions.The class sessions were held twice a week.To nullify the potential effect of methodology, both groups were taught by the same instructor (researcher).Every three session the researcher administered a quiz for both groups.The experimental group was given the task-based quiz and the control group was assessed through classic method (multiple choice, fill-in-the-blanks, true-false and matching test items).Thus, the researcher administered three separate quizzes during the ten sessions and tried to find out the differential effect (if any) of the two assessment types on the following vocabulary learning and final achievement of the participants in the two groups.Three task-based vocabulary quizzes (for experimental group) and three traditional vocabulary quizzes (for control group) were administered every 3 sessions.There were 30 multiple-choice test items in each quiz for control group and 30 vocabulary task items for the experimental group.Finally, the post-test was administered for the two groups after the treatment sessions.

Results
Due to the aforementioned research question, the null hypothesis and the design of the study, different descriptive statistics (like mean, standard deviation, and standard error of the mean) and independent sample t-test were used to analyze the data.Therefore, to ensure the normality of distribution, One-Sample Kolmogorov-Smirnov test was run.Table 1 demonstrates the results of this test.As it is shown in Table 1, p-value for both sets of scores was higher than 0.05.Hence, the scores were normally distributed and two parametric tests of independent sample t-test could be appropriate to be administered.
Therefore, after conducting one Sample Kolmogorov-Smirnov Test, the probable difference between the students' English vocabulary development in the experimental group (EG) and those in the control group (CG) was checked by independent sample t-test.First, the descriptive statistics of two groups are illustrated in Table 2.In addition, the results of independent sample t-test revealed a significant difference in how EG outperformed on the test compared to their counterparts in the control group.As it can be seen in Table 3, mean of the first experimental group whom received vocabulary instruction via TBLI is 15.23, and the control group is 13.244 with the level of significance of .000.Since the level of Sig. was less than 0.05 set for the study, F (2, 29) = 1.744, p= 004<.05).Thus, it was concluded that there was a significant difference between two groups' performance in the posttest, thereby answering the research question, which stated TBI had a significant effect on developing Iranian EFL learners' vocabulary development.
Therefore, the results of the independent sample t-test rejected the research null hypothesis, as well.

Discussion
This study was seeking to investigate the washback effect of TBLA on the EFL Iranian learners of Iran.Alderson and Hamp-Lyons (1996) argue that washback studies have been primarily concerned with the teachers' perspectives and have barely addresses this effect from students' points of views while for a better understanding of how washback occurs as a result of different assessment procedures within the classroom, researchers need to investigate changes in students' motivation, learning styles, learning strategies and educational outcomes and achievements.However, the result of this study urges teachers and task/curriculum designers to adapt task features and predict what task would best deserve academic and curriculum application in terms of fostering task-based vocabulary retention.In addition, Richards and Renandya (2002) argue that vocabulary is a central component of language proficiency and provides much of the basis for how well learners acquire four language skills.

Conclusions
The current study sought to make sure that if these students could then utilize these newly-acquired vocabularies in the learning and accomplishing the administered tasks.Since, vocabulary is regarded as a principal component of language proficiency as it allows learners to use the four language skills: listening, speaking, reading, and writing, this study attempted to exercise a TBLA to help Iranian student to develop their vocabulary reservoir.In terms of the findings of this study, it can be concluded that weaving TBLA into English learning classes in general and vocabulary learning in particular is helpful and washback may offer opportunities to notice the gaps in their developing foreign language vocabulary as a sub-skill.

Table 1 .
One sample Kolmogorov-Smirnov Test

Table 2 .
Descriptive statistics of EG and the CG

Table 3 .
Independent sample T-test Results