Automated feedback in a large-scale implementation of a formative writing system: Implications for improving student

Practice can help improve students writing skills, most particularly when the students are supported with frequent feedback and taught strategies for planning, revising and editing their compositions. Formative systems incorporating automated writing scoring provide the opportunities for students to write, receive feedback, and then revise essays in a timely iterative cycle. This paper investigates the use of the formative writing tool WriteToLearn through mining student logs and essays in order to understand the use of the system and explore the extent to which students improve their writing based on the feedback from the system. In the implementation, students were given writing assignments and were able to write and revise essays. With each submission, the students received feedback on aspects of their writing including scores and instruction about different writing traits, redundancy, as well as grammar and spelling issues. The data collected included over a million student essays written in response to approximately 200 pre-defined prompts as well as a log record of all student actions, revisions and feedback given by the computer. Analyses examined the change in student performance over revisions of essays as well as the effects of different actions occurring within the system and the amount of time spent working on assignments. Implications are discussed for large-scale data analytics on writing assessment in order to understand the role of feedback in writing, to drive improvements in formative technology as well as to aid in designing better kinds of feedback and scaffolding for students to support the writing process. Analyzing Formative Writing 2 Automated feedback in a large-scale implementation of a formative writing system: Implications for improving student writing