Comparative Analysis of SPSS, Winsteps, and Python in Psychometric Analysis: Applications in the Development and Validation of Educational and Psychological Tests

Abstract

This study aimed to provide a comprehensive comparative review of three widely used software packages in psychometric analysis: SPSS, Winsteps, and Python. The review examined their theoretical foundations, statistical capabilities, and applications in the development and validation of psychological and educational tests. A descriptive-analytical review methodology was adopted through the analysis of scholarly literature, reference books, and recent empirical studies, The comparison focused on usability, flexibility, psychometric functionality, support for modern measurement models, automation capabilities, and integration with artificial intelligence technologies.

The findings indicate that SPSS remains the preferred software for conventional statistical analyses and applications based on Classical Test Theory. Winsteps demonstrated superior performance in Rasch measurement, providing advanced tools for item calibration, person ability estimation, fit statistics, and differential item functioning analysis. In contrast, Python offers a comprehensive open-source environment that integrates psychometric analysis with data science, machine learning, and artificial intelligence, enabling advanced statistical modeling and automated analytical workflows.

The study concludes that these software packages should be viewed as complementary rather than competing tools, with software selection depending on research objectives, data characteristics, and the underlying psychometric model. It further recommends expanding researchers' training in modern psychometric software, integrating Python and Winsteps into graduate curricula, and conducting empirical comparative studies using real datasets to enhance the quality of psychological and educational measurement research.