Rianne Conijn is a PhD student in Writing Analytics at the Department of Cognitive Science and Artificial Intelligence, Tilburg University, The Netherlands.
Students’ writing is commonly assessed by the final writing product, while ignoring the writing process. However, insight into the writing process could help us to better understand students’ writing, and moreover, improve learning and teaching of writing. Keystroke analysis has been introduced as a promising method to automatically extract information on the writing process. Yet, it is still considered difficult to align keystroke metrics with cognitive writing processes. Therefore, in Conijn’s PhD project she aims to identify 1) what kind of information would be useful to extract from the writing process; 2) how this information can (and cannot) be extracted from keystroke logs; and 3) how this information in turn can be used to feed back to students and teachers. In this talk, Conijn will discuss three studies in particular. First, she will detail what kind of information undergraduate students, PhD students, teachers, researchers, and teaching support would like to have on their (students’) writing process, based on focus groups conducted. Second, she will discuss how typographic error revisions – although they are only small actions – can have a large impact on keystroke analysis, and how we can distinguish typographic error revisions from other revisions. Lastly, Conijn will discuss my current work on designing a tagset of revisions, to be able to automatically extract a variety of properties from any revision made during the writing process.