The present study attempts to further the understanding of the effect of output in SLA by investigating to what extent an output task (i.e., reconstruction task) is effective in EFL writing. Data in the form of pre-, post- and delayed post-tests were collected from 28 students (14 in input-oriented group while 14 in output group) at one Chinese University. The results revealed a positive and enduring effect of the output task on learners grammatical accuracy in writing as compared with input-oriented instruction that has long been prevalent in Chinese EFL classroom. However the results did not find an equal impact of the task on different rules: The rules carrying meaning (e.g., connective) or having semantic equivalents in target language (e.g. passive -ed or comparative -er or superlative -est ) are produced more accurately by the output group while less meaningful morphology (e.g. plural noun -s , third singular -s ) or local rule (e.g. prepositions) were most difficult for these EFL learners. We argue semantic and functional awareness is the key format for learners to acquire and internalise the grammatical forms.
- output task
- grammatical accuracy