Evaluating Ratings in a Small-Sample Speaking Assessment: A Hierarchical Rater Model Approach

• In Event: Rater Models

Abstract

We examined rater effects in a small-sample speaking performance assessment using a Bayesian Hierarchical Rater Model (HRM) approach. Results suggested that the HRM is effective for evaluating ratings with small sample sizes and nested structures and provides insight into rater effects related to rubric criteria.

Authors

• Mirai Nagasawa, The University of Alabama

• Stefanie A. Wind, UNIVERSITY OF ALABAMA