Development and validation of the BASE-66 inventory for comprehensive academic stress measurement.

Academic stress (AS) is a multidimensional process expressed through emotional, cognitive, physiological, and behavioral responses to academic demands. The COVID-19 pandemic accelerated the digitalization of higher education and restricted in-person social interaction, increasing the salience of socioeconomic constraints, home-based learning conditions, and interactional disruptions. These changes exposed measurement gaps in widely used AS instruments, particularly in their coverage of contextual stressors and in the psychometric consistency of coping-related content. To address this gap, we developed and provided initial psychometric validation for the BASE-66 (Broad Academic Stress Evaluation), an inventory designed to assess AS through an integrated framework that includes both academic and non-academic sources of strain. We used an instrumental psychometric design implemented in three phases: item development, main administration, and post-test evaluation. BASE-66 was adapted from the SISCO-II-AS framework through qualitative focus groups conducted during emergency remote teaching, followed by pilot testing and iterative item refinement. The main administration included 501 university students (71.7% women; mean age 22.19 years, SD=3.13) recruited from three universities in Concepción, Chile. Factor structure was examined with exploratory factor analyses (EFA) based on polychoric correlations using WLSMV estimation in a stratified split sample (EFA n=254; CFA n=247). Confirmatory factor analysis tested the full retained solution. Reliability was assessed with Cronbach's alpha and McDonald's omega in the full sample, and temporal stability was evaluated via test-retest correlations in a follow-up subsample (n=85). Stressor items supported a five-factor structure: Academic Workload Stressors, Academic Performance Stressors, Stressors associated with General Social Interaction, Socioeconomic Stressors, and Stressors associated with Classroom Interaction. Reaction items were best represented by a bifactor model with a dominant General Reaction factor and three specific domains capturing residual variance: Emotional-Cognitive Reactions, Physical Exhaustion Reactions, and Social Conflict Reactions. Coping items suggested two factors (Restorative Coping and Distraction-Reappraisal Coping), but their loadings and reliability were comparatively weaker, supporting a cautious interpretation. The confirmatory model showed acceptable fit (χ²/df=1.28; RMSEA=0.037; SRMR=0.077; CFI=0.93). Internal consistency was adequate to excellent across stressor and reaction domains, and test-retest correlations indicated solid stability for several scales, with lower stability for social and classroom interaction stressors consistent with higher contextual sensitivity. Women scored higher on multiple stressor and reaction domains, whereas classroom interaction stressors and several coping indices did not show gender differences. In conclusion, BASE-66 provides a context-sensitive instrument for comprehensive AS assessment in Chilean university students, capturing academic, socioeconomic, and interactional stressor domains alongside differentiated reaction profiles. Future work should evaluate measurement invariance, strengthen coping measurement, and incorporate longitudinal and multi-method validation designs.