English language learners and automated scoring of essays: Critical considerations

This article presents considerations for using automated scoring systems to evaluate second language writing. A distinction is made between English language learners in English-medium educational systems and those studying English in their own countries for a variety of purposes, and between learning-to-write and writing-to-learn in a second language (Manchón, 2011a), extending Manchón's framework from instruction to assessment and drawing implications for construct definition. Next, an approach to validity based on articulating an interpretive argument is presented and discussed with reference to a recent study of the use of e-rater on the TOEFL. Challenges and opportunities for the use of automated scoring system are presented.

► Discusses differences between assessing writing and assessing language through writing for second language writers of English. ► Presents a validity framework for evaluating automated scoring of second language writing. ► Discusses challenges and opportunities for automated scoring for second language writing assessment.