Asghar Salimi
Aghajani Jahanbakhsh

The present study investigated the effect of task-based language teaching on accuracy, fluency and complexity of low and high proficiency level Iranian EFL learners. This quasi-experimental study involved four intact classes, two at each level of proficiency. There were forty learners in each proficiency level and all of them were male. They were divided into two groups of advanced and intermediate based on their proficiency levels. Then, each proficiency group was divided into two groups of experimental and control. The experimental groups were instructed according to TBLT principles and procedures and the control groups were instructed based on the PPP method. All participants of the study received instruction for one semester of fifteen sessions. At the end of the treatment, a valid pictorial narrative task was employed to collect written data from the participants. The collected written data was quantified and analyzed in accordance with the measure of accuracy introduced by Ellis (2012). The findings of the study revealed that the participants of high proficiency level who received task-based instruction outperformed the other groups in terms of their written accuracy. This study can have some implications for language teachers, task designers, and SLA researchers

Accuracy, Proficiency Level, Task, Task-Based Language Teaching,