The Effects Of Input-Based Practice And Output-Based Practice:a Quasi-Experimental Study With English L2Present Perfect As Target Grammar

Grammar instruction has long been one of the most controversial topics in SLA and language pedagogy. Scholars represented by Krashen hold that language learning is primarily the acquisition of implicit knowledge through natural exposure and language teaching can only lead to the growth in declarative knowledge, yielding little effect on the increase of implicit knowledge. However, with the advent of Teachability Hypothesis and growing recognition of the role of formal instruction, grammar teaching has been once again attached with great significance. Scholars at home or abroad have been seeking new and efficient ways of carrying out grammar instruction. This study, grounded in the theory of Processing Instruction and Output Hypothesis, explores the efficacy of input-based instruction and output-based instruction on English present perfect learning by Chinese EFL learners.The participants of the study are from two intact classes of a middle school in Nanjing. The study follows a research design of pretest, immediate posttest and delayed posttest, all of which include three tasks attempting to measure explicit and implicit knowledge. The treatment lasts40minutes and is immediately followed by a posttest and then a delayed posttest a week later. The tests results are submitted to statistic analysis through SPSS to compare the efficacy of the grammar instruction with two distinct practice modes. The study is guided by the following three questions:1) Which type of instruction is more effective in terms of the students’grammar performance, input-based instruction or output-based one?2) How does the effect of the instruction endure? Are there any differences between the two instruction types in this regard?3) Are there any difference between the two instruction types in terms of students’explicit knowledge or implicit knowledge measured by different instruments?The major findings are:1) regarding the results of the pretest and immediate posttest, the output-based group outperforms the input-based group and the difference reaches a significant level, which indicates that the output-based instruction is more effective;2) regarding the results of the immediate posttest and delayed posttest scores, the output group again outperforms the input group, which seems to support that the output-based instruction endures better than the input-based one;3) both types of instruction lead to a growth in the explicit knowledge and implicit knowledge of the present perfect and the instruction has a greater role in the explicit knowledge growth compared with the implicit one. Besides, the output-based group surpasses the input-based group in this regard with the difference reaching a significant level.The author attempts to explain and discuss the research results from both theory and research design point of view.