Table 3.

Nonstandardized Path Coefficients for Direct, Indirect (where applicable), and Total Effects From Women's Empowerment to Children's Nutritional Status (N = 13,780).

Path and Dependent Outcomes	Latent Predictors	Direct Effect ($\mathit{\beta }$)	Indirect Effect ($\mathit{\beta }$)	Total Effect
Path 1: Intrinsic Agency	Assets	0.22 (0.20, 0.25)
Path 1: HAZ	Intrinsic Agency	0.05 (0.01, 0.08)
Through intrinsic agency	Assets		0.01 (0.003, 0.02)	0.06 (0.01,0.07)
Path 1: WHZ	Intrinsic Agency	0.09 (0.07, 0.12)
Through intrinsic agency	Assets		0.02 (0.015,0.03)	0.11 (0.04,0.13)
Path 1: Anemia	Intrinsic Agency	0.01 (−0.02, 0.04)
Through intrinsic agency	Assets		0.002 (−0.005, 0.009)	0.01 (−0.02,0.05)
Path 2: HAZ	Assets	0.04 (0.01, 0.07)
Path 2: WHZ	Assets	0.08 (0.06, 0.11)
Path 2: Anemia	Assets	0.06 (0.03, 0.08)
Path 3: Instrumental Agency	Assets	0.14 (0.12, 0.16)
Path 3: HAZ	Instrumental Agency	−0.02 (−0.06, 0.01)
Through instrumental agency	Assets		−0.003 (−0.008, 0.001)	−0.02 (−0.05,0.01)
Path 3: WHZ	Instrumental agency	−0.01 (−0.03, 0.02)
Through instrumental agency	Assets		0.000 (−0.004, 0.003)	−0.003 (−0.03,0.03)
Path 3: Anemia	Instrumental agency	0.09 (0.06, 0.13)
Through instrumental agency	Assets		0.01 (0.007, 0.02)	0.10 (0.05,0.14)

Note.

^a Model fit statistics: df = 78; CFI = 0.995; TLI: 0.993; RMSEA: 0.004 (0.001,0.006).

^b 95% confidence intervals based on bootstrapping methods.

^d Anemia model excludes Kenya (n = 4918).

^e Abbreviations: HAZ, height-for-age z-score; WHZ, weight-for-age z-score. df, degrees of freedom; CFI, comparative fit index; TLI, tucker lewis index; RMSEA, root-mean square error of approximation.