Table 2:

Predictors of neonatal mortality, 2015–2016

Dependent variable:	Neonatal death = 1,000; neonatal survival = 0
	Uttar Pradesh	Bihar	Other regionsa
Facility	3.997	−0.979	−6.580***
	(2.803)	(3.335)	(1.814)
Male	11.11***	13.99***	11.11***
	(2.291)	(2.568)	(0.997)
Birth order 2	−55.61***	−48.60***	−44.69***
	(4.426)	(5.443)	(2.011)
Birth order 3	−102.2***	−80.23***	−98.67***
	(7.037)	(8.666)	(3.794)
Birth order 4	−151.4***	−119.7***	−149.6***
	(10.38)	(12.03)	(6.135)
Sibsize of 2	50.50***	42.61***	40.91***
	(4.743)	(6.026)	(2.125)
Sibsize of 3	104.6***	75.20***	96.99***
	(7.267)	(9.125)	(3.806)
Sibsize of 4	164.6***	123.7***	151.7***
	(10.17)	(12.39)	(6.176)
Sibsize of 5	187.1***	135.4***	163.5***
	(12.12)	(14.17)	(7.155)
Sibsize of 6	198.3***	151.6***	180.9***
	(12.34)	(14.39)	(7.751)
Mother’s education (in years)	−2.089	−4.991*	−0.339
	(1.583)	(1.965)	(0.496)
Mother’s age at birth	−7.934***	−8.150**	−3.678***
	(2.188)	(2.688)	(0.985)
Mother’s age at birth2	0.132***	0.134**	0.0680***
	(0.0386)	(0.0492)	(0.0183)
PSU fixed effects	✓	✓	✓
CMC fixed effects	✓	✓	✓
SES indicators	✓	✓	✓
n	40,256	24,721	139,267

Note: The table reports OLS regression coefficients and standard errors (clustered by PSU) from regressions of NNM on facility birth and controls (see Equation 2 and Section 2.2 for a description of the controls).

^*p < 0.05

^**p < 0.01

^***p < 0.001.

^a“other regions” combines data for all states listed in Table 1 except for Uttar Pradesh and Bihar.