Impact of Automated Essay Scorer with Feedback for Essay Writing

Ng, Sing Yii (2018) Impact of Automated Essay Scorer with Feedback for Essay Writing. PhD thesis, Universiti Malaysia Sarawak (UNIMAS).

PDF (Please get the password by email to repository@unimas.my, or call ext: 3914/ 3942/ 3933) Ng Sing ft.pdf Restricted to Registered users only Download (3MB)

Abstract

Immediacy in feedback is necessary to ensure the recursive cycles of improvement in essay writing in line with the process writing approach that enables the cognition of information. However, grading and providing immediate feedback to students’ essays are time consuming and laborious and often lead to teacher burnout. This study aimed to develop and customise a computational method of grading to provide immediate feedback to students called the Automated Essay Scoring Feedback (AESF) and examine its impact on students writing achievement, motivation to write and their acceptance toward AESF. Differences in students’ writing achievement, motivation to write and acceptance of AESF based on gender, English proficiency and computer use experience were also investigated using a quasi-experimental research approach. Teachers’ views on the usability of AESF were also collected through interviews. Participants consisted of 360 Upper 6 students and 24 teachers from five schools in Kuching, Sarawak. Data were collected using a pre-test and a post-test (for writing achievement), Instructional Material Motivation Survey (IMMS) (for motivation for writing), Technology Acceptance Model questionnaire (TAM) (for acceptance of AESF) and general demographics such as gender, English proficiency and computer use experience Data were analysed using the Statistical Packages for the Social Sciences (SPSS). As the data violated parametric tests assumptions, data were analysed using nonparametric tests of Wilcoxon Signed Rank Test, Mann Whitney-U Test, Kruskal Wallis and Spearman’s Rank Correlation Coefficient Test. Students who used the AESF showed better writing achievement, especially in males and students with lower English proficiency. Students using AESF were found to be moderately motivated, and those who preferred immediate paragraph level scoring by AESF have higher motivation to write. Students have moderate acceptance of AESF regardless of gender, language proficiency and computer use experience. Students’ motivation to write was strongly and positively correlated to their acceptance of AESF. Teachers in this study also agreed that AESF could be used as a teaching and assessment tool but opined that it should not be solely depended on to determine students’ writing achievement for summative purposes. Hence, AESF should be further developed and fine-tuned to cater for students need for ongoing writing practise and help teachers make essay marking and feedback provision more manageable.

Actions (For repository members only: login required)

View Item