Literature on second language (L2) writing indicates that, on their own, L2 writers are not able to notice problems with the linguistic and rhetorical features of their drafts and do successful self-revision; and that there is the need to facilitate self-revision in the L2 writing classroom. In view of this need, this dissertation explored the potential of computer-based multimodal composing activities (CBMCA) to help L2 writers' do self-revision in academic writing. It analyzed how 22 English as a second language (ESL) students used the CBMCA to facilitate self-revision as they composed academic papers.

Data collection and analysis were based on a descriptive case study with embedded quantitative data and an integrated theoretical framework of Multimodality, (Kress &Van Leeuwen, 2001), Noticing Hypothesis (Schmidt, 1990), and Multidimensional Model of Revision (Stevenson, Schoonen, & Glopper, 2006). Data include surveys, students' revision history, posters, listening activity, integration of visual and written texts, reflections, stimulated recall interviews, final written drafts, and scores on those drafts.

The findings indicate that the CBMCA helped students discover specific information, rhetorical and linguistic elements, and organizational structure that they used to revise their written draft. In addition, students perceived the CBMCA as useful for self-revision and reported that the activities helped them develop "language" and the "voice" to convey ideas that they were struggling to express using the written mode alone. Further, the findings show that, contrary to findings in most previous research, the students did more content-level revisions than surface level revisions. There were inconclusive findings about the relationship between students' revision history and text quality: there was no significant correlation between revisions and text quality in the third assignment; however, there was a significant correlation between the total frequency of revision and text quality in the fourth assignment. In general, the findings indicate that the CBMCA have the potential to facilitate self-revision in the L2 writing classroom; and that there is the need for L2 writing researchers to re-conceptualize "draft", to focus on revision history rather than between-draft revisions; and pay equal attention to pre-text and point-of-inscription revisions. This dissertation has practical and theoretical implications for L2 writing pedagogy and research.