Enhancing Nurses' Clinical Decision-Making Confidence Through Dual Pathways of Self-Directed Learning: A Structural Equation Model

This study aimed to explore how emotional intelligence influences nurses' clinical decision-making confidence through the dual mediating pathways of cognitive and relational self-directed learning. A cross-sectional survey design was employed, and 1,478 registered nurses were recruited via convenience sampling from medical schools and affiliated hospitals in eastern, central, and western China between June and August 2025. Data were collected using the Wong and Law Emotional Intelligence Scale, a restructured two-factor Self-Directed Learning Self-Rating Scale, and the Clinical Decision-Making Confidence Scale. Structural equation modeling combined with the bias-corrected bootstrap method (5,000 resamples) was used to test the mediation hypotheses. Results revealed that emotional intelligence significantly predicted cognitive self-directed learning (β = 0.643, p < 0.01), relational self-directed learning (β = 0.601, p < 0.01), and clinical decision-making confidence. In the model with cognitive self-directed learning as the mediator, the indirect effect was 0.339 (95% CI [0.273, 0.409]), accounting for 71.8% of the total standardized effect. In the model with relational self-directed learning as the mediator, the indirect effect was 0.241 (95% CI [0.185, 0.302]), accounting for 50.8% of the total standardized effect. Both pathways were statistically significant. The findings indicate that both cognitive and relational self-directed learning serve as significant mediators in the relationship between emotional intelligence and clinical decision-making confidence, supporting the integration of emotional intelligence training and self-directed learning development in nursing education programs.