skip to main content skip to footer

Evaluating the Construct-Coverage of the e-rater Scoring Engine AES

Author(s):
Quinlan, Thomas; Higgins, Derrick; Wolff, Susanne
Publication Year:
2009
Report Number:
RR-09-01
Source:
ETS Research Report
Document Type:
Report
Page Count:
35
Subject/Key Words:
Automated Essay Scoring (AES), Electronic Essay Rater (E-rater), Writing Assessment

Abstract

Factor analyses conducted by Attali and Powers (2008) suggest that e-rater features capture low-level aspects of essay quality, such as sentence complexity, vocabulary, and conventions. Future e-rater development should focus on (a) deepening and expanding coverage of the construct, such as by developing measures of essay content and organization, as well as on (b) addressing accuracy issues in existing features.

Read More