Citation

Tan, Choo Peng (2025) Students performance evaluations in learning probability through conversational-based agent. PhD thesis, Multimedia University.

Full text not available from this repository.

Abstract

This study investigated the effectiveness of a conversational-based agent (CA) in supporting undergraduate students’ learning of probability. Guided by the Design and Development Research (DDR) framework, the study involved designing, developing, and evaluating a conversational-based instructional system grounded in Vygotsky’s sociocultural theory, particularly the concepts of scaffolding and the Zone of Proximal Development. A quasi-experimental mixed-methods design was employed, involving 62 undergraduate students divided into an experimental group and a control group. The experimental group interacted with the CA, while the control group received conventional instruction. Data were collected using the Probability Achievement Test (PAT), Probability Attitude Instrument (PAI), Questionnaire on Students’ Perception of the Conversational-based Agent (SPCA), and semi-structured interviews. Analysis of covariance (ANCOVA) was used to compare post-test outcomes while controlling for pre-test scores. The findings indicated that students in the experimental group achieved higher probability achievement scores and held more favourable attitudes towards probability than those in the control group. The effect sizes for both achievement and attitude were large, suggesting substantial practical significance. However, the assumption of homogeneity of regression slopes was not fully met, indicating that the effects may vary with students’ prior knowledge and initial attitudes; therefore, the results should be interpreted with caution. Descriptive results further showed that students reported positive perceptions of the conversational-based agent, particularly regarding usability, support for understanding, and confidence in problem-solving. Qualitative findings corroborated the quantitative results, revealing that the conversational-based agent promoted active engagement, reduced learning anxiety, and provided flexible and personalised learning experiences. Overall, the findings suggest that the CA is a promising instructional tool for probability learning, supporting both the cognitive and affective aspects of learning. The study contributes to theory by demonstrating how CAs can operationalise scaffolding in a digital learning environment, to methodology through the integration of DDR with a quasi-experimental mixed-methods design, and to practice by providing design guidelines for implementing conversational-based learning systems in higher education.

Item Type:	Thesis (PhD)
Additional Information:	Call No.: LB1028.73 .T36 2025
Uncontrolled Keywords:	Intelligent tutoring systems
Subjects:	L Education > LB Theory and practice of education
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	05 Nov 2026 12:57
Last Modified:	05 Nov 2026 12:57
URII:	http://shdl.mmu.edu.my/id/eprint/17073

Downloads

Edit (login required)