Appendix E.

Regression Analysis of Child’s BPI by Migration Status and Child’s Age and Sex (Standard Errors in Parentheses).

	Internalizing BPI			Externalizing BPI
	Model 2	Model 3	Model 4	Model 2	Model 3	Model 4
Boys, Age 3-9 (N=1,484)
Migration status (ref. rural local, both parents)
Left behind, both away	0.137 (0.462)	0.090 (0.468)	0.353 (0.449)	0.731 (0.772)	0.322 (0.776)	0.057 (0.762)
Boys, Age 10-15 (N=890)
Migration status (ref. rural local, both parents)
Left behind, both away	1.298* (0.588)	1.007 (0.600)	0.478 (0.585)	2.016 (1.056)	1.484 (1.104)	1,012 (1.134)
Girls, Age 3-9 (N=1,202)
Migration status (ref. rural local, both parents)
Left behind, both away	1.328** (0.472)	1.094* (0.476)	0.761 (0.447)	0.987 (0.818)	0.725 (0.824)	0.421 (0.810)
Girls, Age 10-15 (N=762)
Migration status (ref. rural local, both parents)
Left behind, both away	0.838 (0.619)	0.642 (0.637)	0.004 (0.601)	2.067* (1.014)	1.737 (1.006)	1.291 (0.984)

^*p < .05.

^**p < .01.

^***p < .001.

Note: For migration status, only coefficients for children left behind by both parents are shown. Other covariates and other categories of the migration variable are the same as those in Table 3.