We present a multivariate regression model of nutritional indicators, including interactions between education levels and overall community development, as well as estimates of community fixed effects, for two data sets: the first one is from a 2001 nationwide household survey in Nicaragua while the second set was collected in the Western regions of Honduras in 2002, as part of an evaluation process for a family entitlement program. Maternal stature, age difference with an older sibling, household size and income are seen as the main determinants of anthropometric development. Within the more homogeneous communities of Western Honduras, woman's education is also related to the child's nutritional status. The importance of individual community variables is much lower and the inclusion of community fixed effects does not alter the other parameters. However, in the Honduran sample, overall community development (assessed through the community fixed effects) has a small but significant and synergistic effect on the impact of child and household variables. The characteristics of this study and the complex design of entitlement programs prevent an adequate evaluation of those interventions within the poverty reduction strategies. The improvement of intra-community targeting seems to be a key orientation for redesign. An improved socio-economic situation, sound population and family policies, and an appropriate preventive public health care still are safe investments towards improvement of child health and nutrition.