Despite housing ∼18% of the world's population, India does not yet have an estimate of prevalence of autism. This study was carried out to estimate the prevalence of autism in a selected population of school‐children in India. N = 11,849 children (mean age = 5.9 [SD = 1.3], 39.5% females) were selected from various school types from three boroughs in Kolkata, India. Parents/caregivers and teachers filled in the social and communication disorders checklist (SCDC). Children meeting cutoff on parent‐reported SCDC were followed up with the social communication questionnaire (SCQ). SCQ‐positive children were administered the autism diagnostic observation schedule (ADOS). Teacher report on SCDC was available on all 11,849 children. Parent‐report SCDC scores were obtained for 5,947 children. Mean scores on teacher SCDC were significantly lower than parent SCDC. Out of 1,247 SCDC‐positive children, 882 answered the SCQ, of whom 124 met the cutoff score of 15. Six of these children met criteria for autism, autism spectrum disorder (ASD), or broader autism spectrum on the ADOS. The weighted estimate of supra‐threshold SCQ scores was 3.54% (CI: 2.88–4.3%). The weighted prevalence estimate of positive scores (for broader autism spectrum + ASD + autism) was 0.23% (0.07–0.46%). As ∼20% children in this state are known to be out of the school system, and ASD prevalence is likely to be higher in this group, this estimate is likely to represent the lower‐bound of the true prevalence. This study provides preliminary data on the prevalence of broader‐spectrum autism and supra‐threshold autistic traits in a population sample of school children in Eastern India. Autism Res 2017, 10: 1597–1605. ©2017 The Authors Autism Research published by Wiley Periodicals, Inc. on behalf of International Society for Autism Research