Table 5.

Underweight at birth and Stunting – differences between Africans and Coloured

	(1)	(2)	Sample
	Basic	Balanced
African only:
low birth weight	0.021 (0.05)	0.036 (0.05)	213
current HAZ	−0.161 (0.26)	−0.318 (0.42)	230
stunted	−0.025 (0.07)	0.046 (0.11)	230
Coloured only:
low birth weight	0.125*** (0.05)	0.128* (0.08)	162
current HAZ	−0.699 (0.434)	−1.254 (0.88)
stunted	0.086 (0.08)	0.053 (0.15)	190

Marginal effect of being born to a teen mother from weighted probit regression for dichotomous outcomes. Weighted OLS regression used for continuous outcomes. Standard errors presented in parenthesis, bootstrapped for the Balanced estimates. All regressions include a male indicators. Height outcomes include an additional quadratic in age in days and mother’s height. The Balanced models include controls for maternal grandparent education, wave 1 household income, whether the household owns five or more books and the mother’s score on the literacy numeracy test. The same sample is used for each regression on a common outcome. Basic weighted using the original CAPS sample weights, Balanced represents the double robust specification using the inverse probability weight. African only restricts the sample to Africans resulting in matching within the African population group. Similarly, Coloured only matches individuals within the coloured population group only. Estimates marked with three asterisks (***) are significant at the 1% level, those marked with two (**) are significant at the 5% level, and those marked with one (*) are significant at the 10% level.