An Application of Many-Facet Rasch Model Approach to Assessing Creativity in Science

• In Event: Rasch Modeling: Methodology and Application

Abstract

Creativity assessments mostly involve human raters for scoring student creativity products. The examination of rater performance in the scoring process is critically important for the reliability, validity, and fairness of the scores. This study evaluates a creativity assessment in science based on the conceptual framework of rater-mediated assessments. Twenty-four sixth graders consisting of gifted and non-gifted students completed two science tasks that are scored by three groups of raters with different levels of expertise. A many-Facet Rasch Model (MFRM) is used for the data analysis. Major findings include (a) non-gifted students showed higher creative ability than gifted students in these two science tasks and (b) raters with different levels of expertise had comparable scoring performance in creativity assessments.

Authors

• Haiying Long, The University of Kansas

• Jue Wang, University of Science and Technology of China