Characterising students' writing processes using temporal keystroke analysis

This work aims to characterize students' writing processes using keystroke logs and understand how the extracted characteristics influence the text quality at specific moments of writing. Earlier works have proposed predictive models characterizing students' writing processes and mainly rely on distribution-based measures of pauses obtained from the overall keystroke logs. However, the effect of isolated phases of writing has not been evaluated in these models. Moreover, current theories on writing suggest that the quality of writing depends on when specific writing behaviours are performed. This view is not examined in the keystroke logging analysis literature. Addressing the mentioned challenges, the two contributions of this work are: a) characterizing students' writing processes connected to isolated writing phases and examining their influence on writing quality; and b) temporal analysis of keystrokes and investigating whether the significance of writing characteristics varies as students progress in their writing task. Our results suggest that characterizing students' writing based on isolated writing phases is slightly more predictive of writing quality. Additionally, the effect of several writing characteristics on writing quality changes when considering the time dimension.