Table 4 Effect of $1,000 in mean... | Duke University Press

Table 4

Effect of $1,000 in mean district income on pregnancy-related health and birth outcomes: OLS and DDIV estimates, individual birth sample

	OLS			DDIV
	All Births			All Births			Low-Poverty Districts			High-Poverty Districts
	Coef.	SE	n	Coef.	SE	n	Coef.	SE	n	Coef.	SE	n
A. First-Stage Estimates
PA HYPE Status_d × Post_t	—			1,825.4	768.1^*	282	1,551.2	1,194.3	131	2,140.6	974.0^*	151
F statistic				16.56			9.79			7.27
NY HYPE Status_d × Post_t	—			−469.7	1,021.6	181	—			—
F statistic				40.78
PA Periphery_d × Post_t^a	—			2,020.3	1,856.7	289	—			—
F statistic				20.75
B. Estimates for Mean District Income
Pregnancy-related health
Smoking
Prepregnancy (=1)	0.00009	0.00008	76,737	0.003	0.004	76,737	0.008	0.010	35,151	0.000	0.004	41,586
Any trimester (=1)	0.00015	0.00007^*	76,584	0.004	0.004	76,584	0.012	0.011	35,081	0.000	0.003	41,503
Prepregnancy weight
Underweight (=1)	−0.00007	0.00009	61,765	−0.001	0.002	61,765	−0.001	0.003	27,351	−0.001	0.002	34,414
Overweight/obese (=1)	0.00005	0.00018	61,765	−0.001	0.005	61,765	0.002	0.007	27,351	−0.004	0.006	34,414
Gestational weight gain
Low (=1)	0.00014	0.00013	57,211	0.005	0.004	57,211	0.012	0.008	25,103	0.002	0.004	32,108
High (=1)	0.00002	0.00011	57,211	−0.004	0.003	57,211	−0.003	0.005	25,103	−0.005	0.004	32,108
Prenatal care utilization
First trimester (=1)	−0.00021	0.00018	58,629	−0.001	0.006	58,629	−0.002	0.006	25,679	0.000	0.009	32,950
Adequate or more (=1)	0.00025	0.00019	56,863	0.018	0.007^**	56,863	0.022	0.014	25,020	0.014	0.007^*	31,843
Birth outcomes
Birth weight
Very low (=1)	0.00000	0.00001	77,161	−0.001	0.001	77,161	−0.002	0.002	35,304	−0.001	0.001	41,857
Low (=1)	−0.00005	0.00005	77,161	−0.001	0.002	77,161	−0.007	0.007	35,304	0.002	0.002	41,857
Weight (grams)	0.08801	0.12753	77,161	−6.228	5.190	77,161	1.289	11.181	35,304	−9.340	5.129^†	41,857
Gestation length
Preterm birth (=1)	0.00004	0.00006	77,457	0.000	0.003	77,457	−0.001	0.005	35,437	0.001	0.003	42,020
Gestation (weeks)	−0.00009	0.00047	76,730	−0.011	0.024	76,730	0.007	0.043	35,099	−0.017	0.024	41,631

	OLS			DDIV
	All Births			All Births			Low-Poverty Districts			High-Poverty Districts
	Coef.	SE	n	Coef.	SE	n	Coef.	SE	n	Coef.	SE	n
A. First-Stage Estimates
PA HYPE Status_d × Post_t	—			1,825.4	768.1^*	282	1,551.2	1,194.3	131	2,140.6	974.0^*	151
F statistic				16.56			9.79			7.27
NY HYPE Status_d × Post_t	—			−469.7	1,021.6	181	—			—
F statistic				40.78
PA Periphery_d × Post_t^a	—			2,020.3	1,856.7	289	—			—
F statistic				20.75
B. Estimates for Mean District Income
Pregnancy-related health
Smoking
Prepregnancy (=1)	0.00009	0.00008	76,737	0.003	0.004	76,737	0.008	0.010	35,151	0.000	0.004	41,586
Any trimester (=1)	0.00015	0.00007^*	76,584	0.004	0.004	76,584	0.012	0.011	35,081	0.000	0.003	41,503
Prepregnancy weight
Underweight (=1)	−0.00007	0.00009	61,765	−0.001	0.002	61,765	−0.001	0.003	27,351	−0.001	0.002	34,414
Overweight/obese (=1)	0.00005	0.00018	61,765	−0.001	0.005	61,765	0.002	0.007	27,351	−0.004	0.006	34,414
Gestational weight gain
Low (=1)	0.00014	0.00013	57,211	0.005	0.004	57,211	0.012	0.008	25,103	0.002	0.004	32,108
High (=1)	0.00002	0.00011	57,211	−0.004	0.003	57,211	−0.003	0.005	25,103	−0.005	0.004	32,108
Prenatal care utilization
First trimester (=1)	−0.00021	0.00018	58,629	−0.001	0.006	58,629	−0.002	0.006	25,679	0.000	0.009	32,950
Adequate or more (=1)	0.00025	0.00019	56,863	0.018	0.007^**	56,863	0.022	0.014	25,020	0.014	0.007^*	31,843
Birth outcomes
Birth weight
Very low (=1)	0.00000	0.00001	77,161	−0.001	0.001	77,161	−0.002	0.002	35,304	−0.001	0.001	41,857
Low (=1)	−0.00005	0.00005	77,161	−0.001	0.002	77,161	−0.007	0.007	35,304	0.002	0.002	41,857
Weight (grams)	0.08801	0.12753	77,161	−6.228	5.190	77,161	1.289	11.181	35,304	−9.340	5.129^†	41,857
Gestation length
Preterm birth (=1)	0.00004	0.00006	77,457	0.000	0.003	77,457	−0.001	0.005	35,437	0.001	0.003	42,020
Gestation (weeks)	−0.00009	0.00047	76,730	−0.011	0.024	76,730	0.007	0.043	35,099	−0.017	0.024	41,631

Notes: Each cell represents results from separate models. Across all models, data are restricted to PA districts with 51+% of land area above either the HYPE or the Periphery of the Marcellus Shale geological formation. Mean district income is adjusted for inflation to 2010 US$ using CPI-U-RS. Panel A presents the DiD estimator (π) for the interaction between location above the HYPE and postboom period (2011–2012) (i.e., HYPE_d × Post_t) in the prediction of mean district income using IPW and population-weighted district-level models that include period, district, and region fixed effects and robust and clustered standard errors. Panel B presents the simple and observed DiD OLS association between mean district income and each outcome, as well as the second-stage DDIV results. The second-stage DDIV coefficients (β) are for the interaction between instrumented mean district income and postboom period (2011–2012) (i.e., HYPE_d × Post_t) from individual-level models that include period, region, and district fixed effects and birthing parent control variables. Individual birthing parent control variables include race and ethnicity, marital status, educational attainment, age, and age squared.

All PA HYPE shale districts were dropped for this test.

^†

p < .10; ^*p < .05; ^**p < .01

View Large

This Feature Is Available To Subscribers Only