Many studies showed a global acute increase in stress hormones at kindergarten entry, but inconsistencies among them due to small samples with varying methodologies and very few measurement points, preclude robust conclusions. The current study aimed to describe the pattern of morning cortisol concentration in children during the transition from preschool to school by examining whether we could identify certain stress responses such as stress reactivity, but also anticipatory stress and stress recovery. We further tested the robustness of this pattern across several characteristics. Participants were 384 children recruited from two cohorts of a pregnancy study, and followed across their transition from preschool to kindergarten. Five morning salivary cortisol samples were collected over time from the children: twice before school entry and three times after school entry. Although no anticipatory stress was observed two weeks before school entry, latent growth curve models showed that most children’s morning cortisol concentrations increased during the first two weeks of school, and was not associated with any sociodemographic characteristics, supporting the hypothesis that school entry is a universal social stressor. In contrast, the rate of recovery measured two months after school entry, however, revealed between children variance, which could not be explained by any specific sociodemographic characteristics. This study showed that stress reactivity at school entry is sustained for at least two weeks in most children. However, the observed variability in the rate of recovery might be associated with adaptation to the school transition and will allow further testing of functional correlates.