Constructing and Validating a Multidimensional Model of Students’ Statistical Disposition in Indonesian Higher Education

Abstract

Background: Statistical disposition significantly influences students’ attitudes, beliefs, and behaviours toward learning statistics. Given the increasing demand for statistical literacy in our data-driven society, there is a growing need for a valid and reliable framework to measure this construct. However, comprehensive empirical models, especially in the Indonesian higher education context, are still limited.

Objective: This study aims to theoretically construct and empirically validate a dimensional model along with its indicators to measure students’ statistical disposition in higher education settings.

Methodology: A quantitative approach was adopted using the descriptive method and cross-sectional survey design. Participants included undergraduate, master's, and doctoral students from a state university in Indonesia. Confirmatory Factor Analysis (CFA) was employed to examine model fit and validate the constructs.

Results: The initial model with seven dimensions and nine indicators (two adapted from prior studies) failed to meet the acceptable fit criteria. After six iterative modifications, the final model, comprising six dimensions and thirteen indicators, showed good model fit and statistical validity.

Conclusion: The validated model offers a theoretically sound and empirically supported framework for assessing statistical disposition among university students, contributing to enhanced evaluation and understanding in the field of statistics education.

Unique Contribution: This study presents one of the first empirically validated multidimensional models of statistical disposition tailored for Indonesian higher education, filling a notable gap in international statistics education research.

Key Recommendation: Educators and curriculum developers should integrate the validated model into teaching strategies and assessment tools to foster students’ positive dispositions toward learning statistics and support ongoing educational improvements.