ABSTRACT
As many current K–12 English language proficiency (ELP) assessments in the U.S. are computer-based, there is increasing interest in and demand for automated scoring of spoken and written constructed responses. Within this context, this chapter first provides an overview of the main benefits and challenges associated with using automated scoring in K–12 ELP assessments. The chapter then introduces the main components of a typical automated scoring system (preprocessing, feature extraction, scoring model, and filtering model) and reviews several automated scoring systems that have been used for scoring extended written responses and monologic spoken responses. Finally, the chapter discusses important considerations that should be taken into account by stakeholders who are considering applying automated scoring to K–12 ELP assessments and makes recommendations to ensure a responsible use of the technology.
