Statistical Methods for Assessments in Simulations and Serious Games

• In Event: Uses and Issues in Cognitive Diagnostic Modeling

Abstract

Simulation or game-based assessments produce outcome data and process data. In this paper some statistical models which can potentially be used to analyze data from simulation or game-based assessments are introduced. Specifically, cognitive diagnostic models which can be used to estimate latent skills from outcome data so as to scale these assessments are presented under the framework of Bayesian networks; five prospective data mining methods which can be employed to discover problem solving strategies from process data are described. Some studies in the literature that apply some of these methods to analyze simulation or game-based assessments are presented as application examples. Recommendations are provided for selecting appropriate scaling and data mining methods for these assessments; future directions of research are proposed.

Authors

• Jianbin Fu, ETS

• Diego Zapata-Rivera, Educational Testing Service

• Elia Mavronikolas, Educational Testing Service