text reconstruction

In an attempt to contribute to the ongoing debate about how output tasks affect noticing of linguistic forms, the present study set out to investigate the effect of pushed output tasks on grammatical accuracy in sentence writing of Iranian EFL learners. Fifty homogenous Iranian EFL learners were randomly assigned to two experimental and control groups. Then, every group underwent ten different treatment sessions. The control group received writing instruction through conventional methods, while the experimental group received instruction through two pushed output tasks. In the case of the experimental group, in the first five treatment sessions, four grammatical structures were presented through picture cued tasks. The next five treatment sessions directed at other structures took place via reconstruction tasks. Two different versions of the writing section of the Preliminary English Test (PET) were used as pre/post-test. The results indicated that the experimental group significantly outperformed the control group. Therefore, it might be argued that pushed output-based tasks had a positive effect on the Iranian EFL learner’s grammatical accuracy in sentence writing. These findings provide empirical support for the output hypothesis and have pedagogical implications for the choice of output-oriented grammar tasks.