An Integrated UTAUT, Machine Learning, and Association Rule Mining Analysis of Password Security Behavior Among Educational Personnel

Abstract

This study examined password security behavior among educational personnel at Krabi Technical College using an integrated Unified Theory of Acceptanceand Use of Technology (UTAUT), machine learning, and association rule mining framework. Based on 210 survey responses and 50 variables, behavioral intention (BI) and use behavior (UB) were operationalized as composite scores derived from UTAUT-aligned Likert-scale items. In contrast, password behaviors captured practical routines such as password reuse, recovery capability, composition practices, and perceived attackability. Explanatory results showed that BI was mainly associated with hedonic motivation (β = 0.414), performance expectancy (β = 0.285), and facilitating conditions (β = 0.162), while UB was associated with BI (β = 0.444), habit (β = 0.206), and social influence (β = 0.142). Although the structural equation model indicated an excellent global fit, severe multicollinearity among constructs necessitated interpreting path coefficients as conditional associations rather than independent causal effects. Predictive analysis using five-fold cross-validation showed that construct-only ridge regression achieved the best performance for BI (RMSE = 0.491, R 2 = 0.650) and UB (RMSE = 0.553, R 2 = 0.640), with no statistically significant improvement over standard linear regression. Association rule mining revealed co-occurrence patterns centered on facilitating conditions, recovery capability, habit, and social influence. The findings support a parsimonious, interpretable approach to cybersecurity behavior analysis and provide actionable guidance for institutional password training, recovery support, and policy design.

Received: 25 April 2026 | Revised: 16 June 2026 | Accepted:  8 July 2026

Conflicts of Interest

The authors declare that they have no conflicts of interest to this work.

Data Availability Statement

The data that support the findings of this study are openly available in GitHub at https://github.com/pjarupunphol/Datasets/blob/main/Krabi_Technical.zip.

Author Contribution Statement

Pita Jarupunphol: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration. Sujira Jeennoo: Conceptualization, Validation, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Project administration.