The Effect Of Output Tasks On The Chinese College English Majors' Acquisition Of Lexical Chunks

In this thesis, three concepts are involved, namely, lexical chunks, language input and language output. Lexical chunks as an important and ubiquitous language phenomenon have aroused growing interest of linguists. In the field of SLA (Second Language Acquisition) and Corpus linguistics, a large number of theoretical and empirical studies have been devoted to them. Language input has been a hot topic and received much attention from many scholars. Comparatively speaking, the important effects of language output on SLA have been neglected until Swain proposed the Output Hypothesis in the 1980s. She argues that the language output of L2 (second language) learners can help them use language fluently and accurately. A successful L2 learner not only needs to receive enough comprehensible input but also needs to produce comprehensible output. Although lexical chunks, language input and output have all been discussed heatedly, the relationship among them has been scarcely touched upon. Thus, this thesis aims to test the effects of different output and input tasks on Chinese College English majors'noticing and acquisition of English lexical chunks.Altogether 80 third-year English majors from the three parallel classes of the PLA University of Foreign Languages participated in the present study. They were classified into two EGs (experimental groups) consisting of 28 students and 26 students respectively, and one CG (control group) consisting of 26 students. After the experiment, all the data collected from the experiment were analyzed to answer the following research questions:(1) Do the two EGs notice more target lexical chunks in subsequent input than the CG who does not produce any target language after being exposed to different tasks?(2) Does EG 1 (experimental group 1) notice more target chunks in subsequent input and produce more correct target lexical chunks than EG 2 (experimental group 2) after being exposed to different output tasks?(3) Do the two EGs grasp more target lexical chunks than the CG in the posttest? Does EG 1 grasp more target lexical chunks than EG 2 in the posttest? The whole experiment includes the following steps:(1) The participants took a pretest in the first week to test whether they were on the same English proficiency level before the treatment.(2) In order to avoid the test effect, the treatment was carried out one week after the pretest. All the participants were asked to read an English passage and underline the chunks. Then EG 1 and EG 2 were asked to reconstruct the essay in the oral form and written form respectively. The CG was only required to answer some comprehension questions about the passage. After that, the same input material was given to all the participants again for another reading and underlining, followed by another new reconstruction and comprehension. Finally, five participants in each group were asked to complete a questionnaire.(3) The posttest was carried out one week after the treatment to see whether there was any increase after the experiment compared with the pretest.After the experiment, the oral data were transcribed into the written form first. Then, all the valid data were processed with the SPSS (11.5) software. After the data analysis and discussion, the following results and findings were obtained:(1) In the present study, the two EGs noticed more target lexical chunks in subsequent input than the CG who does not produce any target language after being exposed to different tasks. Thus, a more positive effect of output tasks than the input task on the noticing of linguistic features is confirmed.(2) A significant difference was only found between the two times of underlining for EG 1 (p=0.047); as for the production issue, both the two EGs showed significant differences (p=0.000 for both the two EGs). On the basis of the descriptive statistics, we found that EG 1 did better than EG 2. Therefore, EG 1 noticed more target chunks in subsequent input and produced more correct target lexical chunks than EG 2 after being exposed to different output tasks. With these results, we can conclude that the oral output task may be more effective to help learners notice lexical chunks than the written one.(3) The difference of the posttest scores among the three groups was not significant statistically (p=0.58) although their acquisition scores all increased from the pretest to the posttest. That is to say, the two EGs did not grasp more target lexical chunks in the posttest than the CG after the treatment. And no difference was found between the two EGs.