.png){kind=icon}
.png){kind=icon}
Return
Applying Bayesian Networks to Profile L2 learners' Communicative Competence in a Scenario-Based Language Assessment
.png){kind=icon}
PRE
AI
H
DOI：10.1080/15434303.2025.2588117.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
As technology-based innovative assessments, such as scenario-based language assessments (SBLAs), continue to evolve in their ability to measure complex constructs that reflect learners' cognitive structures, it is essential to identify psychometric approaches that support the interpretation and understanding of assessment results. Bayesian networks offer a robust modeling framework for this purpose, as they enable logical reasoning about the interrelationships among variables in assessments with complex constructs. This study utilized a Bayesian network to model how L2 learners' ability to achieve a scenario goal depended on their topical knowledge and L2 knowledge, skills, and abilities. While exploratory in nature, the findings illustrate the potential of Bayesian networks to provide meaningful interpretations of the relationships among measured constructs within an SBLA, ultimately enhancing the interpretive value of assessment results.
Keywords:
Bayesian networks
scenario-based language assessments
L2 learners
communicative competence
psychometric modeling
Journal
L
IF:
2.8
Papers:
30
Citations:
1.1K
Organization
No organization information available
Cited Papers