INVESTIGATING THE REVISION PROCESS IN A WEB-BASED PEER REVIEW SYSTEM USING MACHINE LEARNING
University of Tartu, Centre for Academic Writing and Communication (ESTONIA)
In the teaching of writing, certain aspects of peer feedback on writing remain controversial; in particular, in the context of second language (L2) writing, there remain questions about how peer feedback supports the L2 writing process. (Liu & Sadler, 2003).
Peer feedback can have a positive effect on second language writing (Lundstrom & Baker, 2009). Additionally, technology related research fields increasingly testify that learning to write benefits from peers communicating in a web-based environment for language learning (Matsuda et al., 2003; Hyland & Hyland, 2006). These studies report how computer mediated communication itself provides a rich learning experience for second language learners, but also highlight that different web-based platforms provide different results.
This study investigates the effectiveness of one such web-based platform: SWoRD, a web-based peer review system. The basic principles behind the development of web-based peer review systems are as follows:
1) writing in higher education plays a significant role to develop content knowledge;
2) it assists in organising writing assignments and reduces the workload of instructors;
3) it provides students with an authentic audience to whom they give and from who they receive feedback; and
4) it enables students to practice writing through constant revision for the purpose of learning how to write (Cho & Schunn, 2004).
The aim of this study is to better understand how L2 learner-writers conduct peer feedback activities by looking at the types of feedback they provide and the factors that lead to the implementation of this feedback. A machine learning approach was used to predict how eight features, related to the peer feedback comments, performed on an L2 writing task amongst a group of first, second, and third year Bachelor students (N=43) from the department of Chemistry at the University of Tartu, Estonia. The results of this study indicate that peer feedback is effective amongst L2 writers using SWoRD web-based peer review system, if feedback instances propose specific alterations and if more than one peer makes a reference to a similar aspect in the text that needs changing.
 Cho, K., & Schunn, C. D. (2004). The SWoRD is mightier than the pen: Scaffolded writing and rewriting in the discipline. Advanced Learning Technologies, 2004. Proceedings. IEEE International Conference on, 545–549.
 Crossley, S. A. (2013). Advancing research in second language writing through computational tools and machine learning techniques: A research agenda. Language Teaching, 46(02), 256-271.
 Hyland, K., & Hyland, F. (2006). Feedback on second language students’ writing. Language Teaching, 39, 83–101.
 Liu, J., & Sadler, R. W. (2003). The effect and affect of peer review in electronic versus traditional modes on L2 writing. Journal of English for Academic Purposes, 2(3), 193–227.
 Matsuda, P. K., Canagarajah, A. S., Harklau, L., Hyland, K., & Warschauer, M. (2003). Changing currents in second language writing research: A colloquium. Journal of Second Language Writing, 12(2), 151–179.