Table A.4. Robustness to Alternate Age Controls.

Effects of Treatment on Early-life Investments with Varied Age Controls

Panel A: No Age Controls
	Vaccinations				Breastfed for more than 6 mos.
	No. of polio doses	No. of DPT doses	Measles	No. of total vaccinations
Own Treatment Exposure	0.492*** (0.153)	0.544*** (0.163)	0.191*** (0.0608)	1.224*** (0.356)	0.161*** (0.0538)
Observations	456	456	454	454	453
Mean of Dependent Variable	2.645	2.629	0.756	6.026	0.872
Panel B: Smooth Quadratic
	Vaccinations				Breastfed for more than 6 mos.
	No. of polio doses	No. of DPT doses	Measles	No. of total vaccinations
Own Treatment Exposure	0.372** (0.137)	0.425*** (0.152)	0.104* (0.0562)	0.899*** (0.322)	0.0842* (0.0484)
Observations	456	456	454	454	453
Mean of Dependent Variable	2.645	2.629	0.756	6.026	0.872
Panel C: Quarter of Birth × Integer Age
	Vaccinations				Breastfed for more than 6 mos.
	No. of polio doses	No. of DPT doses	Measles	No. of total vaccinations
Own Treatment Exposure	0.214* (0.122)	0.257* (0.136)	0.0550 (0.0554)	0.523* (0.290)	8.25e-05 (0.00943)
Observations	456	456	454	454	453
Mean of Dependent Variable	2.645	2.629	2.629	6.026	0.872

Notes: Robust standard errors in parentheses (*** p<0.01, ** p<0.05, * p<0.1), clustered at the level of DHS cluster (primary sampling unit). Treatment exposure calculated using month and year of birth, and years in which the program was rolled out in each district (see Table A1). Specifications also control for: district fixed effects; urban dummy; rainy season dummy; percent of household's salt that is iodized (0, 25, 50, 75, or 100); minimum and maximum sibling ages; dummies for median sibling age; household size (deciles); number of females in household (quintiles); gender-specific birth order dummies; dummies for number of younger and older siblings; dummies for mother's integer age; and dummies for mother's years of completed schooling. Sample is restricted to households with at least 1 child < 5 years in 1999 DHS.