Assessment of the Quality of Education in Universities with Different Institutional Cultures: Statistical Analysis of Relationships

Introduction. The quality of higher education remains a critical concern for policymakers and institutions globally, yet empirical cross-national research, particularly in post-Soviet contexts, remains scarce. Existing studies often rely on single-institution data, retrospective methodologies, and traditional input-based indicators, while neglecting emergent determinants such as institutional culture, leadership, and student resilience. This study addresses these gaps by investigating two research questions: (1) Do quality indicators differ significantly depending on the specific characteristics of the educational institution? and (2) Which quality constructs have predictive power in the relationship between institutional culture and student satisfaction? Research methods. A cross-sectional survey design was employed, collecting data from 719 students across four diverse universities in Russia and Uzbekistan: a large Russian federal university (KFU), its branch in Uzbekistan (KFUJ), a private Uzbek university (REU), and a national classical Uzbek university (NUU). Eight latent constructs (institutional culture, teaching effectiveness, academic staff quality, digital competency, educational resources, leadership and policy, academic resilience, and student satisfaction) were measured using 35 items on a 5-point Likert scale. Data were analyzed using confirmatory factor analysis (CFA), one-way ANOVA, and partial least squares structural equation modeling (PLS-SEM) with 500 bootstrap resamples. KEYWORDS Results. The measurement model demonstrated excellent fit (CFI=0.999, RMSEA=0.022). Statistically significant inter-institutional differences were found across all eight constructs (p<0.05), with NUU and KFU students reporting the highest overall quality scores and KFUJ the lowest. Leadership and policy emerged as the strongest predictor of student satisfaction (β=0.442, f2=0.648, p<0.001), followed by teaching effectiveness (β=0.202), academic staff quality (β=0.141), and academic resilience (β=0.138). Educational resources and digital competency showed only bivariate correlations with satisfaction but lost significance in the multivariate model due to multicollinearity with leadership. Conclusion. The study provides the first validated cross-national empirical comparison of an eight-variable quality assessment model in post-Soviet higher education. Leadership and policy-not traditional inputs like resources-constitute the most powerful driver of student satisfaction. These findings challenge prevailing quality models that emphasize teaching inputs and physical infrastructure while downplaying governance and student adaptive capacities. Practical recommendations include investing in evidence-based governance, faculty development in active pedagogy, structured mentorship programs to build academic resilience, and strategic quality assurance reforms aligned with ESG 2015 standards. Future longitudinal and multi-group SEM research is needed to establish causality and test measurement invariance across broader institutional typologies

Hali tarjima qilinmagan