Exploring The Impact Of Digital Skills On Subjective Well-Being Using Hybrid Models In Synthetic Panel Data Analysis

Subjective well-being (SWB) is a fundamental indicator of quality of life, reflecting individuals’ perceptions of satisfaction and happiness. However, research is often constrained by the limited availability of cross-sectional or panel datasets that capture both SWB and digital skills indicators. Moreover, developing effective policy requires models that differentiate both within-cluster and between-cluster effects, showing variations among population groups. This study examines how digital competencies influence SWB across different age cohorts, using hybrid models to account for the heterogeneity between groups in synthetic panel. In the absence of a comprehensive unified dataset, we constructed a synthetic panel by integrating data from four Rosstat databases covering 76 Russian regions from 2016 to 2023, enabling analysis at both cohort and regional levels. Fixed effects models enhance causal inference for time-varying factors, such as skill acquisition and income fluctuations. To further differentiate these influences, we isolate both contextual (between-cohort) and individual (within-cohort) effects by employing a hybrid regression model. Findings indicate that digital skills impact SWB through multiple channels mediated by economic conditions and health dynamics. Specifically, work satisfaction is primarily influenced by economic conditions. Health satisfaction is affected by inequality and access to digital services. Active lifestyle is closely linked to digital interactions and education. The findings suggest that policymakers should prioritize advanced digital training programs to mitigate unintended lifestyle trade-offs.