Table 5.

Wajir estimated intent-to-treat effects at two-year follow-up, by study arm

	(1)	(2)	(3)	(4)	(5)	(6)	(7)
	V-Only two-year follow-up Mean	VE Estimate	VEH Estimate	VEHW Estimate	VEH vs VE (3)–(2)	VEHW vs VE (4)–(2)	VEHW vs VEH (4)–(3)
Violence Prevention
Experienced violence by a male in the past year (=1) [n = 1878]	0.038	−0.006	0.015	− 0.022	0.021	−0.017	− 0.037*
95% CI		[−0.04, 0.03]	[−0.02, 0.05]	[− 0.05, 0.01]	[− 0.02, 0.06]	[− 0.05, 0.01]	[− 0.07, − 0.01]
Gender equitable attitudes z-scorea [n = 1903]	0.000	−0.171	− 0.208*	− 0.018	− 0.037	0.153	0.190†
95% CI		[−0.39, 0.05]	[−0.39, − 0.02]	[− 0.26, 0.22]	[− 0.23, 0.15]	[− 0.09, 0.39]	[− 0.02, 0.40]
Positive gender schooling attitudes z-score [n = 1878]	0.160	0.156*	0.099	0.068	−0.057	−0.088	− 0.031
95% CI		[0.00, 0.31]	[−0.05, 0.25]	[− 0.08, 0.21]	[− 0.22, 0.10]	[− 0.23, 0.06]	[− 0.18, 0.12]
Education
Grade attainment	4.492	0.259**	0.194*	0.181*	−0.065	− 0.079	− 0.013
95% CI		[0.10, 0.42]	[0.02, 0.36]	[0.01, 0.35]	[−0.23, 0.10]	[− 0.24, 0.09]	[− 0.19, 0.17]
Primary school complete (=1)	0.130	0.010	0.021	−0.029	0.011	−0.039	− 0.050
95% CI		[−0.05, 0.07]	[−0.05, 0.09]	[− 0.08, 0.03]	[− 0.06, 0.08]	[− 0.10, 0.02]	[− 0.11, 0.01]
Enrolled in current school year (=1)	0.808	0.144***	0.070*	0.084**	−0.074*	−0.060*	0.014
95% CI		[0.09, 0.20]	[0.01, 0.13]	[0.02, 0.14]	[−0.13, − 0.02]	[− 0.11, − 0.01]	[−0.05, 0.08]
Health
Knows most fertile period during menstrual cycle (=1) [n = 1878]	0.049	−0.031†	−0.021	− 0.013	0.010	0.018	0.008
95% CI		[−0.06, 0.00]	[−0.05, 0.01]	[− 0.04, 0.02]	[− 0.01, 0.03]	[−0.01, 0.04]	[− 0.01, 0.03]
Knows method of modern contraceptiona (=1) [n = 1848]	0.390	−0.097†	− 0.103*	− 0.096†	− 0.007	0.001	0.007
95% CI		[−0.21, 0.02]	[−0.19, − 0.01]	[−0.20, 0.01]	[− 0.10, 0.09]	[− 0.11, 0.11]	[−0.08, 0.10]
SRH knowledge z-scorea,b [n = 1400]	0.000	0.344*	0.433**	0.234†	0.090	−0.110	−0.199
95% CI		[0.08, 0.60]	[0.17, 0.70]	[−0.03, 0.50]	[−0.17, 0.35]	[− 0.37, 0.15]	[− 0.47, 0.07]
General self-efficacy z-score [n = 1878]	0.990	− 0.054	−0.014	− 0.030	0.040	0.024	−0.016
95% CI		[−0.25, 0.14]	[−0.22, 0.19]	[− 0.22, 0.16]	[− 0.17, 0.25]	[−0.17, 0.21]	[− 0.21, 0.18]
Wealth creation
Financial literacy z-score [n = 1878]	0.439	−0.069	0.030	0.185	0.099	0.254*	0.155
95% CI		[−0.31, 0.17]	[−0.19, 0.25]	[− 0.06, 0.43]	[− 0.08, 0.28]	[0.04, 0.46]	[− 0.04, 0.35]
Saved money in the past 6 months (=1) [n = 1878]	0.011	0.030	0.047*	0.409***	0.017	0.379***	0.362***
95% CI		[−0.01, 0.07]	[0.01, 0.09]	[0.31, 0.51]	[−0.03, 0.06]	[0.27, 0.48]	[0.26, 0.47]
Household-level outcomes
Expects girl to complete secondary school (=1) [n = 2007]	0.861	0.058*	0.050†	0.043	−0.008	−0.015	−0.007
95% CI		[0.01, 0.10]	[0.00, 0.10]	[−0.01, 0.10]	[−0.05, 0.04]	[− 0.06, 0.03]	[− 0.06, 0.05]
Household wealth quintile [n = 2011]	3.226	− 0.434*	−0.247	− 0.211	0.188	0.223	0.035
95% CI		[−0.83, − 0.04]	[− 0.64, 0.14]	[−0.54, 0.12]	[− 0.25, 0.62]	[− 0.16, 0.60]	[−0.34, 0.41]
Summary index z-scores
Violence prevention outcomes summary index z-scorec [n = 1878]	0.000	0.030	−0.101	0.104	−0.131†	0.073	0.205**
95% CI		[−0.15, 0.21]	[−0.29, 0.08]	[− 0.07, 0.27]	[− 0.29, 0.02]	[−0.06, 0.21]	[0.06, 0.35]
Education outcomes summary index z-score	0.000	0.302***	0.189*	0.122	−0.114	−0.180*	− 0.067
95% CI		[0.16, 0.44]	[0.02, 0.36]	[−0.03, 0.27]	[−0.29, 0.07]	[− 0.33, − 0.03]	[−0.25, 0.12]
Health outcomes summary index z-score [n = 1878]	0.000	−0.026	0.066	−0.014	0.092	0.012	−0.080
95% CI		[−0.22, 0.17]	[−0.13, 0.26]	[− 0.22, 0.19]	[− 0.06, 0.24]	[−0.15, 0.17]	[− 0.24, 0.08]
Wealth creation outcomes summary index z-score [n = 1878]	0.000	0.167	0.349*	2.852***	0.182	2.686***	2.504***
95% CI		[−0.14, 0.47]	[0.03, 0.66]	[2.14, 3.56]	[−0.15, 0.51]	[1.97, 3.40]	[1.79, 3.22]

The table reports two-year follow-up means for V-only in column 1, the OLS estimated ITT effect for each study arm relative to V-only in columns 2–4 and differences in the estimated ITT effects across study arms in columns 5–7. For binary outcomes (=1) these are linear probability models. Column 5 compares the estimates for VEH to VE, column 6 compares VEHW to VE, and column 7 compares VEHW to VEH. For example, the estimate in column 7 for ‘Experienced violence by a male in the past year’ (−0.037) is the difference between the estimate for VEHW (− 0.022) in column 4 and the estimate for VEH (0.015) in column 3. Minor differences in the reported differentials compared to the estimates presented in columns 2–4 are due to rounding. Numbers in square brackets indicate 95% confidence intervals. Regressions were estimated with standard errors clustered at the village level and included controls for 2009 district per the stratified randomization, age and the outcome measured at baseline unless otherwise noted. N = 1909; sample is smaller for some individual outcomes due to missing data as indicated

*** p < 0.001, ** p < 0.01, * p < 0.05, † p < 0.1

^a No baseline control for outcome variable available

^b Non-response for one or more items on the scale ranged from 24 to 30% across study arms and was higher for younger girls

^c Violence prevention variable is reverse coded prior to inclusion in summary variable