Table 3.

Main impacts on early marriage and pregnancy among youth aged 14–21 at baseline, by sex and country

	Male		Female
	Ever married or cohabited	Ever married or cohabited	Ever married or cohabited	Ever married or cohabited	Ever pregnant	Ever pregnant	Currently pregnant	Currently pregnant	Ever given birth	Ever given birth
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
	Unadjusted	Adjusted	Unadjusted	Adjusted	Unadjusted	Adjusted	Unadjusted	Adjusted	Unadjusted	Adjusted
Panel A: Malawi (Social Cash Transfer Program)
Treatment status	−0.0135	−0.0179*	−0.00983	−0.00428	−0.00137	0.00507	−0.0177	−0.0171	0.0196	0.0231
	(0.0111)	(0.00878)	(0.0187)	(0.0176)	(0.0238)	(0.0169)	(0.0174)	(0.0172)	(0.0234)	(0.0205)
Observations	1,023	1,023	878	878	917	917	917	917	917	917
R‐squared	0.371	0.403	0.364	0.454	0.403	0.463	0.094	0.100	0.379	0.447
Endline control mean	0.06		0.27		0.41		0.06		0.35
Panel B: Zambia (Multiple Category Targeted Grant)
Treatment status	0.000861	−0.00114	0.0112	0.0117	0.000860	0.000716	−0.0117	−0.0106	0.00375	0.00382
	(0.00779)	(0.00770)	(0.0145)	(0.0141)	(0.0187)	(0.0198)	(0.00957)	(0.0116)	(0.0174)	(0.0173)
Observations	1,296	1,296	1,070	1,070	1,210	1,210	1,210	1,210	1,210	1,210
R‐squared	0.140	0.167	0.398	0.415	0.362	0.398	0.117	0.133	0.386	0.424
Endline control mean	0.02		0.07		0.22		0.04		0.18

NOTES: Estimations of equation (1) use ANCOVA modeling among panel individuals (follow‐up after 30 months in Malawi and after 36 months in Zambia). Robust standard errors in parentheses corrected for clustering. ^***p < 0.01, ^**p < 0.05, ^*p < 0.1. All controls are measured at baseline and include stratifying indicators used for the randomization (traditional authorities dummies in Malawi and districts in Zambia). Inconsistent observations, namely those individuals reporting ever being married/pregnant at baseline but never being married/pregnant at endline were excluded from the analysis. Sampling weights have been applied in Malawi.