A Reliability Generalization Meta-analysis of the Spanish Burnout Inventory (SBI)

Abstract

Burnout is a response to chronic work stress that occurs when the individual feels overwhelmed and powerless to cope with difficulties at work, that affects a wide range of professional fields. Epidemiological data concerning this syndrome reflect the seriousness of the problem. Those who suffer from the syndrome usually show health problems of a psychosomatic, emotional, attitudinal, and behavioral nature. In addition, burnout has negative effects on organizations (sick leave, reduced performance) and for the users of the service. The Spanish Burnout Inventory (SBI) is a 20-item 5-point Liker-type scale that allows a rapid assessment of the prevalence of this regarding four dimensions: enthusiasm toward the job, psychological exhaustion, indolence, and guilt. The extensive administration of the SBI gives us the opportunity to synthesize and support the conclusions regarding some of its psychometric properties with samples of teachers from a wide variety of countries. The goal of the present work is to study the reliability of the SBI based on the model that better fits its internal structure. To do so, the One-Stage meta-analytic structural equation modeling (MASEM) technique is applied to the empirical Pearson correlation matrices collected in 32 independent studies. First, a combined correlation matrix is obtained following a fixed-effect and random-effects model to study the evidence of heterogeneity. Then, the four-factor model is fitted to the combined matrix. Based on the goodness of fit indices and the heterogeneity evidence, a reliability index is computed according to the four-factor structure and potential moderator variables are examined.