Implementation Of An AWE System On Chinese Non-English Majors' Skill Of Lexical Application In Writing

The automated writing evaluation (AWE) systems, increasingly utilized in the context of English as a foreign language (EFL) teaching and learning, enable EFL learners to proceed with their writing practice at their own rate with automated feedback in computer-based language learning environments. In the process of learner-computer interaction,learners have opportunities to notice their language problems according to the automated feedback provided by AWE systems and make some interactive modification in their subsequent writing. Regarding written output, EFL learners' skill of lexical application exerts great influence on their writing quality. However, there is not much research reported on the learners' lexical application in the context of AWE-based writing.Therefore, the study is conducted to investigate if there is any positive effect of the AWE system on non-English majors' skill of lexical application in AWE-based writing.The current study attempts to answer the following three research questions:? Does the AWE-based writing have positive effects on improving students' skill of lexical application?? In terms of lexical application in AWE-based writing, are there differences among students of three English proficiency levels?? In terms of lexical application in AWE-based writing, are there differences between male and female students?The study is based on the data collected from writing tasks on the Pigai system (www.pigai.org), and students' responses to the questionnaire concerning the Pigai system, in an English Audio-Video Speaking Course (EAVSC) for one semester from September 2015 to January 2016. The author served as an observer in class with the permission of the instructor. In the study, the Coh-Metrix tool, the Range 32 and SPSS 22.0 were employed to measure and analyze the linguistic features in students' writing texts in terms of word count, word length,type-token ration, word frequency, hypemym and connectives.The findings from data analysis of students' writing and their responses to the questionnaire items revealed that there was enhancement of students' lexical application in linguistic features through a comparison between their first and final drafts. Students tended to make revisions with more concrete words and low-frequency words in the process of learner-computer interaction. High-achievers took initiatives to make modifications in sentence-level supplement though they were not required. Intermediate-achievers were most active in revision. For low-achievers, modifications often remained at the level of spelling and grammar in their reconstruction of texts. Male students used more concrete words and connectives than females. The majority of students perceived that the automated feedback provided by the Pigai system enables them to realize their language weakness in writing and helped promote their writing skill. The study hopes to shed light on the effective use of the Pigai system in classroom practice and improvement of the system development with respect to its automated feedback.