This paper studies the decision made by a family to invest in student migration. We propose an empirical structural decision model which reflects the importance of both the return to the investment and the budgetary constraint in the choice of the family. We circumvent the problem of endogeneity of the educational attainment by deriving sharp bounds and conduct inference for the parameters of interest. The data are collected on students from Cameroon, using a new snowball sampling procedure, which allow the inclusion of both migrants and non-migrants in the sample. We propose bias corrected estimators for this procedure. We study the characteristics of potential candidates to migration that increase or decrease their probability to migrate, accounting for a potential helper in the diaspora. Among the interesting results we find that a choice to complete a Master's degree doubles the odds of migration, there is little evidence of gender preference, students migrants are positively selected on their previous academic results.