Table A2.

Covariate balance in estimating the average treatment effect of child migration on the word test score.

		Girls (Mean)		Boys (Mean)
		Treated	Control	Treated	Control
Age (years)	Unmatched	12.24	12.56 †	12.44	12.60
	Matched	12.25	12.33	12.42	12.27
Father’s education (years)	Unmatched	7.17	6.16 *	7.73	6.23 ***
	Matched	7.16	7.29	7.71	6.80
Mother’s education (years)	Unmatched	5.85	4.29 ***	5.80	4.23 ***
	Matched	5.85	5.00	5.77	4.70 †
Had brother(s)	Unmatched	0.63	0.66	0.36	0.36
	Matched	0.62	0.67	0.36	0.42
Had sister(s)	Unmatched	0.36	0.45 †	0.40	0.51 *
	Matched	0.35	0.32	0.40	0.40
Paternal grandparent(s) alive	Unmatched	0.81	0.86	0.90	0.86
	Matched	0.83	0.88	0.90	0.93
Paternal grandparent(s) alive	Unmatched	0.93	0.91	0.93	0.88
	Matched	0.92	0.91	0.93	0.92
Born in a hospital	Unmatched	0.58	0.38 ***	0.64	0.41 ***
	Matched	0.57	0.58	0.63	0.60
Ever attended a kindergarten	Unmatched	0.82	0.48 ***	0.84	0.51 ***
	Matched	0.82	0.76	0.83	0.80
% Agricultural population (county)	Unmatched	0.74	0.79 *	0.70	0.79 ***
	Matched	0.76	0.73	0.71	0.77
% Agricultural population squared	Unmatched	0.60	0.66 *	0.57	0.66 **
	Matched	0.62	0.60	0.57	0.65 †
Region of Birth (ref: East)
North	Unmatched	0.13	0.11	0.09	0.11
	Matched	0.13	0.12	0.09	0.02 †
Northeast	Unmatched	0.09	0.08	0.05	0.08
	Matched	0.10	0.05	0.06	0.03
Central	Unmatched	0.14	0.17	0.16	0.18
	Matched	0.14	0.05 *	0.17	0.17
South	Unmatched	0.21	0.16	0.20	0.16
	Matched	0.19	0.23	0.19	0.16
Southwest	Unmatched	0.21	0.16	0.19	0.18
	Matched	0.22	0.29	0.19	0.31 †
Northwest	Unmatched	0.12	0.23 **	0.10	0.20 *
	Matched	0.12	0.13	0.10	0.11

Note: The control group include non-migrant and left-behind children.

^†p < .1;

^*p < .05;

^**p < .01;

^***p < .001 (two-tailed tests).