Supporting Students' Text-Based Evidence Use Via Formative Automated Writing and Revision Assessment

Richard Correnti, Elaine Lin Wang, Lindsay Clare Matsumura, Diane Litman, Zhexiong Liu, Tianwen Li

ResearchPosted on rand.org Nov 13, 2024Published in: The Routledge International Handbook of Automated Essay Evaluation, Chapter 12, pages 221-243 (2024). DOI: 10.4324/9781003397618-15

This chapter describes work on developing a classroom-based formative assessment system focused on improving elementary students’ text-based evidence-use in argument writing. The chapter discusses design choices for building such a system, the development of enabling artificial intelligence technologies, and the results of classroom evaluations. The initial version of this system (called eRevise) was organized around a writing construct central to argument writing, namely, evidence use. The current version of the system (called eRevise +RF ) adds formative assessment of a second construct central to writing, namely, revision aligned with feedback. Automated scoring and feedback provided to students for both constructs are based on features aligned with rubrics used in human scoring. System validity is demonstrated via improvement in students' writing across drafts aligned with the feedback students received. Interactive effects between teaching and students' use of such systems in classroom settings are also discussed.

Topics

Document Details

  • Publisher: Taylor & Francis Group
  • Availability: Non-RAND
  • Year: 2024
  • Pages: 23
  • Document Number: EP-70749

This publication is part of the RAND external publication series. Many RAND studies are published in peer-reviewed scholarly journals, as chapters in commercial books, or as documents published by other organizations.

RAND is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.