ABSTRACT
Automated essay scoring (AES) is a well-established technology in educational settings. The technology is now supported by a number of commercial vendors, and is used to evaluate millions of essay responses in instructional and assessment settings in Kindergarten through 12th grade (K–12), undergraduate, and graduate-level contexts. Earlier motivations for this technology stemmed from issues related to the time and costs associated with human scoring. Continued development on e-rater® (Attali & Burstein, 2006; Burstein et al., 1998) has deliberately focused on the development of a greater variety of features to more comprehensively address the writing construct (specifications for characteristics of writing being measured, or evaluated), using tangible markers in scoring rubric criteria. 1 The focus of this chapter is two-fold. It includes a description of the e-rater automated essay scoring system and its natural language processing (NLP)-centered approach, and a discussion of the system’s applications and development efforts for current and future educational settings.
