Table 8.

Robustness of the effect of abortion centers to including all interviewed women aged ≥25 instead of restricting 2006 and 2011 surveys to ever-married women

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	Any Method	Modern Method	Female Sterilization	Male Sterilization	Modern Method Other Than Sterilization	Traditional Method	Ever Had an Abortion	Share of All Pregnancies Aborted
Number of Abortion Centers	0.977**	0.979**	0.979†	1.001	0.977**	0.988	1.007	0.002**
	(0.0051)	(0.0060)	(0.0106)	(0.0099)	(0.0044)	(0.0078)	(0.0092)	(0.0003)
Number of Observations	25,174	25,174	24,843	25,150	25,174	24,848	24,454	24,268
Number of Clusters	75	75	72	74	75	70	69	75
Pseudo-R 2	.0665	.0618	.1607	.1177	.1006	.0828	.1259	.0736
Mean Y	0.479	0.435	0.196	0.085	0.155	0.045	0.050	0.016

Notes: Data are omitted for the following variables: three dummy variables indicating DHS waves; district fixed effects; and controls for urban location, age at interview, religion, and education. Columns 1–7 report odds ratios from a logit model. Column 8 presents coefficients from a linear regression including a constant (coefficient not reported here). District-correlated robust standard errors are shown in parentheses. Regressions are weighted using survey weights. The sample is women aged 25–49 who usually reside in the household, irrespective of their marital status. The excluded religious category is “Hindu,” and the excluded education category is “No education.” Some observations are dropped in columns 3, 4, 6, and 7 because of lack of variation in the value of the dependent variable within district. Observations for women who have never had any pregnancy are dropped in column 8 because the share of aborted pregnancies is not defined for these women.

Sources: Authors’ calculations using Demographic and Health Surveys of Nepal (1996–2011) and Technical Committee for Implementation of Comprehensive Abortion Care (2010).

^† p < .10; **p < .01