BackgroundThe role of children and staff in SARS-CoV-2 transmission outside and within households is still not fully understood when large numbers are in regular, frequent contact in schools.MethodsWe used the self-controlled case-series method during the alpha- and delta-dominant periods to explore the incidence of infection in periods around a household member infection, relative to periods without household infection, in a cohort of primary and secondary English school children and staff from November 2020 to July 2021.ResultsWe found the relative incidence of infection in students and staff was highest in the 1-7 days following household infection, remaining high up to 14 days after, with risk also elevated in the 6-12 days before household infection. Younger students had a higher relative incidence following household infection, suggesting household transmission may play a more prominent role compared with older students. The relative incidence was also higher amongst students in the alpha variant dominant period.ConclusionsThis analysis suggests SARS-CoV2 infection in children, young people and staff at English schools were more likely to be associated with within-household transmission than from outside the household, but that a small increased risk of seeding from outside is observed.Key messagesQuestion: With respect to incidence of SARS-CoV-2 infection before and after household member infection, is within-household transmission more likely than community transmission amongst school children and staff?Findings: In this self-controlled case-series analysis, the relative incidence of infection in students and staff was highest in the 1-7 days following household infection, remaining high up to 14 days after, with the highest relative incidence found in younger, primary school-aged children.Importance: Within-household transmission is more likely than from outside the household, but a small increased risk of seeding from outside the household is observed as well as variation by age and variant dominant period.