BACKGROUND: Diarrhoea and respiratory infections are the two biggest causes of child death globally. Handwashing with soap could substantially reduce diarrhoea and respiratory infections, but prevalence of adequate handwashing is low. We tested whether a scalable village-level intervention based on emotional drivers of behaviour, rather than knowledge, could improve handwashing behaviour in rural India. METHODS: The study was done in Chittoor district in southern Andhra Pradesh, India, between May 24, 2011, and Sept 10, 2012. Eligible villages had a population of 700-2000 people, a state-run primary school for children aged 8-13 years, and a preschool for children younger than 5 years. 14 villages (clusters) were selected, stratified by population size (<1200 vs >1200), and randomly assigned in a 1:1 ratio to intervention or control (no intervention). Clusters were enrolled by the study manager. Random allocation was done by the study statistician using a random number generator. The intervention included community and school-based events incorporating an animated film, skits, and public pledging ceremonies. Outcomes were measured by direct observation in 20-25 households per village at baseline and at three follow-up visits (6 weeks, 6 months, and 12 months after the intervention). Observers had no connection with the intervention and observers and participant households were told that the study was about domestic water use to reduce the risk of bias. No other masking was possible. The primary outcome was the proportion of handwashing with soap at key events (after defecation, after cleaning a child's bottom, before food preparation, and before eating) at all follow-up visits. The control villages received a shortened version of the intervention before the final follow-up round. Outcome data are presented as village-level means. FINDINGS: Handwashing with soap at key events was rare at baseline in both the intervention and control groups (1% [SD 1] vs 2% [1]). At 6 weeks' follow-up, handwashing with soap at key events was more common in the intervention group than in the control group (19% [SD 21] vs 4% [2]; difference 15%, p=0·005). At the 6-month follow-up visit, the proportion handwashing with soap was 37% (SD 7) in the intervention group versus 6% (3) in the control group (difference 31%; p=0·02). At the 12-month follow-up visit, after the control villages had received the shortened intervention, the proportion handwashing with soap was 29% (SD 9) in the intervention group and 29% (13) in the control group. INTERPRETATION: This study shows that substantial increases in handwashing with soap can be achieved using a scalable intervention based on emotional drivers. FUNDING: Wellcome Trust, SHARE.