Alongside classical determinants of education, there is a growing literature of social interactions in education which seems to be particularly concentrated in developed countries. This seems paradoxical as norms, culture and social capital appear to play a more important role in everyday life in Africa. We use a rich data set collected in Niakhar in rural Senegal, between 2001 and 2008 to study whether the school attendance of a child depends on the school attendance of other children in the same social group. Social groups are defined using geographical proximity and caste groups. While it is particularly difficult to empirically identify the impact of social group behavior, we take advantage of the temporal structure of the data to deal with a number of endogeneity issues. We rely moreover on different empirical strategies and placebo tests to argue that our results are not subject to confounding interpretations. Results show evidence of a strong and positive effect of social interactions on school attendance and the impact is greater for members of the highest caste.