Expanding Learning Progressions: Modelling Student Cognitive Processes In The Domain Of Rational Number

Year: 2015

Author: Haryanto, Bakir, Curtis, David, Vosniadou, Stella

Type of paper: Abstract refereed

Cognitive Diagnostic Assessment (CDA) has emerged as a result of the blending of cognitive theory and psychometric analysis into an assessment framework that provides tools for investigating the structure of student knowledge and thinking skills. CDA relies on a cognitive model, which is known as a construct map in Wilson’s four building blocks or student model in Mislevy’s evidence centered design. Learning progressions are often used in CDA to as a cognitive model to describe the acquisition of student knowledge and skills in a particular domain. However, it still gives a general description of student cognition. Therefore, the main objective of this study is to propose a new cognitive model in CDA by expanding learning progressions into three dimensions of student cognition, namely, conceptual, procedural, and problem solving skills in the context of the rational number domain for grades 7 to 9. We will also examine the relationship among these dimensional cognitions and look further on the cognitive challenges students face as they develop their knowledge of rational number. A Bayesian Inference Network (BIN) will serve as the psychometric model from which several plausible models will be produced as an approach for modeling the expanding learning progression and uncovering the student cognitive challenges. Those Bayesian models will be evaluated in terms of their parameter recovery and robustness of the inference