Feedback-Revision Project

Title

Towards Automatic Tracking of Student Responses to Teacher Feedback in Draft Revision

Period

01 January 2017 - 31 December 2018

Abstract

Student revision of their drafts is recognized as an important strategy to support the development of their writing skills. It involves a complex process of evaluating text, diagnosing problems and making changes to improve the text. Early research found that novice and experienced writers make different kinds of changes in the course of revision, and indicated that success in revision could be influenced by the effectiveness of the feedback received. More recent studies in English as a Second Language (ESL) writing have been overwhelmingly concerned with written corrective feedback (also known as grammar correction) and its effectiveness in improving the accuracy of student writing. However, there have been surprisingly few studies exploring what features of teacher written feedback influence student revisions. In most studies that investigated the impact of teacher feedback on student revisions, only a very limited number of ESL students at a single university in a English-speaking country were involved. Arguably, the small size of samples and the specific settings limit not only the generalization of research findings to other student populations, but also their applicability to practice in real educational settings.

To address the limitations of previous research, this project will adapt the existing frameworks to support fine-grained categorization of student revisions and teacher feedback. Based on the adapted frameworks and Natural Language Processing (NLP) methods, the project aims to design and implement an automatic classification system for identifying the types of student revisions and teacher feedback as well as detecting the connections between them. Additionally, the impact of the system on the teaching and learning process of writing will be investigated. This project is a collaboration between information technology (IT) and ESL researchers from four higher education institutions in Hong Kong and Taiwan, which builds a strong research partnership in the Chinese communities working towards the project's objectives. The project opens up opportunities for students to be promptly informed about their uses of different revision types and the links of their revisions with teacher feedback, adding a new layer of feedback to allow students to review their writing strategies. It also provides an efficient way for teachers to identify those students who have problems in revising their drafts and are in need of early assistance. Equally importantly, the project proposes a promising approach for researchers on revision and feedback to systematically analyze data collected from a large sample of students and teachers.

Objectives

  1. To develop an automatic classification system for identifying the categories of student revisions and teacher feedback as well as detecting the connections between them;
  2. To evaluate the agreement of the classification results between the system and human assessors;
  3. To examine the impact of the system on the quality of students’ final drafts; and
  4. To explore the views of students and teachers on the role of the system in the teaching and learning process of writing.

Screenshots









Publications

Journal Papers
Conference Papers

Prizes and Awards

Acknowledgement

This project was financially supported by General Research Fund (No. 18608816) of the Research Grants Council (RGC) of Hong Kong, China.