ACTNext researchers contributed three chapters and the foreword of the new Handbook of Automated Scoring: Theory into Practice (2020, Taylor and Francis).
Scott Wood, Associate Director, Machine Scoring and Senior Research Scientist in Artificial Intelligence and Machine Learning for ACTNext, has a chapter titled, “Public Perception and Communication around Automated Essay Scoring.” Wood gave a preview of his chapter during Episode 6 of the ACTNext Navigator podcast.
ACT Principal Assessment Specialists Rick Meisner and Brad Bolender wrote “Quality Control for Automated Scoring in Large-Scale Assessment.”
Saad Khan, ACTNext’s Director of Artificial Intelligence and Machine Learning, and Yuchi Huang, Senior Research Scientist in AI/ML, contributed a chapter on “Deep Learning Networks for Automated Scoring Applications.”
The authoritative handbook provides “a scientifically grounded overview of the key research efforts required to move automated scoring systems into operational practice… organized into three parts that cover (1) theoretical foundations, (2) operational methodologies, and (3) practical illustrations, each with a commentary.”
Automated scoring engines
From the foreword by ACTNext Senior Vice President Alina von Davier:
“Automated scoring engines […] require a careful balancing of the contributions of technology, NLP, psychometrics, artificial intelligence, and the learning sciences. The present handbook is evidence that the theories, methodologies, and underlying technology that surround automated scoring have reached maturity, and that there is a growing acceptance of these technologies among experts and the public.”