Do illiterate mothers learn from their literate kids? Evidence from maternal nutritional knowledge

This study provides empirical evidence on whether literate kids augment the nutritional knowledge of their illiterate mothers. A mother's ability to correctly diagnose her child's height and weight for age is used to measure maternal nutritional knowledge. To mitigate the potential endogeneity of living with literate kids, we use an instrumental variable (IV) approach and estimate unobserved effect probit IV models using longitudinal data from the 2002 and 2006 Young Lives Ethiopian household survey. The results from our preferred specification suggest that an illiterate mother who lives with a literate kid is 6.2 and 4.4 percentage points more likely to correctly diagnose her child's height and weight for age, respectively, relative to an illiterate mother who lives with no literate household member. Falsification tests suggest that our results are not confounded by other factors. © 2020 John Wiley & Sons Ltd