This paper studies two experiments of early childhood development programs in The Gambia: one increasing access to services, and another improving service quality. In the first experiment, new community-based early childhood development (ECD) centers were introduced to randomly chosen villages that had no pre-existing structured ECD services. In the second experiment, a randomly assigned subset of existing ECD centers received intensive provider training. We find no evidence that either intervention improved average levels of child development. Exploratory analysis suggests that, in fact, the first experiment, which increased access to relatively low quality ECD services, led to declines in child development among children from less disadvantaged households. Evidence supports that these households may have been steered away from better quality early childhood settings in their homes.