Using data from the National Survey of Children, this paper examines the hypothesis that the amount of time children spend viewing television and the extent to which the content viewed is sexual in nature is related to the initiation of sexual activity. Several theories that would lead to this hypothesis are reviewed. The data do not provide any strong or consistent evidence for such links. However, some aspects of the context in which television is viewed are related to sexual activity. The authors suggest ways in which the design and measures could be strengthened to provide a more rigorous test of the hypothesis.