Given the importance of considering context in development, the goal of the present study was to develop and provide initial validity evidence for the Kids’ Church Survey (KCS), a new measure of children's church‐based social support. Data were collected from 1253 children ages 6–14 attending mainline Protestant, evangelical Protestant, and Catholic churches. Parallel and exploratory factor analyses supported a three‐factor solution: received (emotional) church support, perceived church support from peers, and perceived church support from adults. Confirmatory models conducted with independent samples provided an excellent fit for the data. All three scales evidenced acceptable internal (.78–.92) and test–retest (.88–.95) reliability. Measurement invariance was demonstrated across genders and age groups, with the exception of the perceived peer support scale, which was not invariant across ages. The KCS was sensitive to between‐church differences in children's programs and incrementally predicted self‐esteem, prosocial behavior, and spirituality. Applications for researchers, mental health practitioners, and clergy are discussed.