C-SAW (Computer-Supported Argumentative Writer) is an online authoring software embodying design principles derived from theories on written argumentation, self-regulation and conceptual change as well as feedback from practitioners and users, in line with a design-based research approach. Designed to scaffold writing processes, C-SAW is intended as additional support in instructional designs using argumentative writing for learning. This article presents the results of a mixed-method study comparing undergraduate students writing with C-SAW or a text editor. Outcome measures included the number of arguments and the degree of their completion, knowledge of argument components, topic knowledge and changes in epistemic beliefs. Participants writing with C-SAW elaborated arguments to a greater degree, but there were otherwise no significant differences between conditions for other measured outcomes. Furthermore, results were influenced by informal reasoning skills that outweighed the effects of condition. These results are discussed with respect to the difficulties of studying the effects of digital tools on writing and learning in controlled first-use contexts and the importance of developing instructional designs with explicit learning outcomes that are aligned to the instructional principles embedded in digital tools.

Spontaneous writing observed in chats, instant messengers, and social media has become established as productive modes of communication and discourse genres. However, they remain understudied from the perspective of writing process research. In this paper, we present an empirical study wherein keystrokes made by chat users in a game were recorded. The distributions of the inter-key intervals were analyzed and fitted with ex-Gaussian distribution equation, and an argument for psycholinguistic interpretation of the distribution parameters is presented. This analysis leads to establishing a threshold of 500 ms for the identification of pauses in spontaneous writing. Furthermore, we demonstrate that pauses longer than 1.2 s may correspond to higher-level linguistic processing beyond a single propositional expression (functional element of the discourse).