EXPANDING AWE TO INCORPORATE READING AND WRITING EVALUATION

The majority of research on automated writing evaluation (AWE) to date has focused on writing tasks that require students to process little to no outside information. More recently, there has been an increased focus on the evaluation of source-based writing, which requires readers to process multiple pieces of information and integrate them for an essay. Critically, these writing assignments require readers to use both comprehension and writing skills, which may be difficult to disentangle from essays alone. This chapter presents a proposal for comprehension-aware AWE - that is, automated evaluations of essays that are sensitive to students’ processing and comprehension of source materials. We review computational approaches to assessing both comprehension and source-based writing as well as recent work which aims to integrate the two. We then describe some of the key components that would be necessary for a comprehension-aware AWE system and point to future directions that could build on this foundational proposal.