What shapes teachers’ turnover intentions? An integrative perspective from the job demands–resources model

Teacher turnover undermines school stability and student outcomes. Guided by the job demands–resources model, this study used random forest analysis to examine 36 potential antecedents of turnover intentions. Data were collected from 830 Czech primary and lower secondary school teachers with a mean age of 45.9 years (SD = 10.9) across 44 schools. Twenty-four predictors exceeded the random-noise benchmark, with physical fatigue and job satisfaction emerging as the strongest. Work engagement, cognitive weariness, work ability and work–life conflict also played major roles, alongside personal resources such as resilience, teacher self-efficacy and locus of control. Several predictors exhibited non-linear or threshold patterns. Conservation of resources (COR) theory provided a complementary lens for interpreting U-shaped resource-related patterns in terms of resource conservation and investment. By evaluating a broad set of antecedents, the study extends existing research beyond isolated predictors and provides a more integrated account of the factors shaping teachers’ turnover intentions. It also reveals patterns that conventional linear analyses may obscure. Together, these findings provide school leaders and policymakers with a basis for retention strategies. Such strategies should prioritise early burnout detection, strengthen work engagement and job satisfaction, and be tailored to teachers at different career stages.