Table 5.

Summary of Significant Associations and Pair-wise Differences, Regressions of HAZ(1), and ucHAZ with and without Interactions

	HAZ(1)	ucHAZ(1:5)	ucHAZ(5:8)	HAZ(1)	ucHAZ(1:5)	ucHAZ(5:8)	HAZ(1)	ucHAZ(1:5)	ucHAZ(5:8)	HAZ(1)	ucHAZ(1:5)	ucHAZ(5:8)
	(No interactions)			(No interactions)			(Sex interactions)			(Urban interactions)
	Pair-wise comparisons1			p<0.05			p<0.05			p<0.05
Consumption quintile	(4th):V≠I (5th):P≠I, E≠P, P≠V			I,P,V	E,P,V	E	I,P,V	E,V	E	I,V	E,V	E
Caregivers’ schooling	P≠I		E≠I, E≠P	E,P,V	V	I,P	E,P,V		P	E,P		I
Fathers’ schooling		E≠P, E≠V		E,I,P,V	E		I,V	E		I
Mothers’ height	E≠I, E≠V			E,I,P,V	E,I,P,V	E,I,P	I,P,V	E,I,P,V	E,I	E,P,I	E,I,P,V	E,I
Female	E≠I	P≠I, V≠I, E≠P, E≠V	E≠I, V≠I	E,I,P,V	P,V	E,V	V			E,I,P,V	P, V	E,V
Moved after R1	E≠I, E≠P			E,V			E,V			E
Time between measurement					I,P,V			I			I
Age at R1				E,I,P,V			E,I,P,V			E,I,P,V
Hospital presence	V≠I, E≠I, E≠P, E≠V	P≠I, V≠I, E≠V	V≠I	E,V	E,I	I	E,V	E,I	I	E,P,V	E,I,P
Site is urban											V
Community wealth				P			P		I		I
Sex Interactions: Fathers’ schooling							P	E
Urban Interactions:
Consumption quintile										E,V	V	V
Mother’s schooling
Fathers’ schooling												V
Mothers’ height											I
Female										E,P		I
Moved after R1
Time between  measurement											V
Hospital presence										P	E,I,P,V	I
Community wealth										V
Test: interactions jointly significant at p<0.05							E,P,V	E	I	E,P,V	E,I,P,V	E,I,P,V

Notes: E = Ethiopia, I = India, P = Peru, V = Vietnam;

¹Coefficient estimates for consumption quintiles may differ across countries due to differences in absolute per capita consumption levels in these four countries. Indeed we do find significant differences across countries for a number of the coefficient estimates for the consumption quintiles, which is consistent with the impacts of different absolute consumption per capita across the four countries. See etable 2 for analysis using PPP-adjusted consumption per capita.