The Chilean Early Childhood Longitudinal Survey was used to model the choices made by households with respect to day care or preschool (DC/PS) attendance. We also use georeferenced data on the location of centers that could take care of children. We present a detailed analysis of the correlates of DC/PS attendance choices, giving special attention to the distance between the household and the center as predictors. For these purposes, the joint decision of child attendance to DC/PS and mother's employment is modeled. The measurement of this association is relatively new to the literature because it requires georeferenced data, which only recently have begun to be collected systematically by public policy agencies. The association we find between distance and attendance to DC/PS centers is significant but smaller than the association to the age of the child. Separate analyses for 0- to 14-month-olds and 25- to 54-month-olds show some heterogeneous effects on the joint decision of attendance and education.