Theoretical Background: The COVID-19 pandemic was associated with various demands and challenges, especially for children and adolescents, including fears of a potentially life-threatening infection, contact restrictions, and school closures. This study examines possible changes in children's and adolescents' stress experience during the COVID-19 pandemic and the mediating role of emotion regulation. Objective: We hypothesized (1) that children and adolescents report higher levels of (a) stress experience and (b) somatic stress symptoms during the pandemic. We examined changes over 2 years exploratively. Furthermore, we expected (2) that functional and dysfunctional emotion regulation strategies mediate the potential changes over time. Method: In the fall of 2020 (NT1 = 173) and 2022 (NT2 = 121), we collected self-reports from German high school students from grades 5 to 8 (T1) and 7 to 10 (T2), respectively; a subsample (n = 71) participated in both surveys. In both assessments, we administered two subscales (stress vulnerability and somatic stress symptoms) of the German Stress and Coping Questionnaire for Children and Adolescents (SSKJ 3 - 8-R; Lohaus et al., 2018) and two subscales (functional and dysfunctional emotion regulation strategies) of the Process-Oriented Emotion Regulation Measure for Children and Adolescents (POEM-CA; Ruth & Lohaus, 2023). In 2022, we conducted an additional survey (N = 83 German high school students) to cross-validate the results on stress vulnerability and somatic symptoms. Results: To test Hypothesis 1, we compared data from T1 (2020) with norm data on the SSKJ 3 - 8-R (2018) using multivariate analysis of variance. The multivariate effect was significant, indicating higher scores for the 2020 survey. At the univariate level, the difference was attributable to stress experience and persisted longitudinally, whereas we found no differences for physical stress symptoms. Longitudinally, we found no significant differences between T1 and T2. The cross-validation with a second high school sample replicated increased scores for stress experience as well as for physical stress symptoms. To test Hypothesis 2, we conducted mediation analyses with functional and dysfunctional emotion-regulation strategies as mediators for longitudinal changes of the stress indicators (stress experience and symptoms). Changes in somatic stress symptoms were fully mediated by dysfunctional emotion regulation. Discussion and Conclusion: The results indicate a significant increase in the experience of stress (and partially also somatic stress symptoms) in children and adolescents during the COVID-19 pandemic compared to prepandemic data from the norm sample. In combination with a lack of emotion-regulation skills, this may contribute to an increase in psychological problems. We discuss the consequences and limitations of the study.