. 2021 Nov 4;36(1):153–169. doi: 10.1007/s11269-021-03018-8

Table 2.

Regression Results, Marginal Effects

		(1)			(2)			(3)			(4)
	ME		Std Err.	ME		Std Err.	ME		Std Err.	ME		Std Err.
Capacity
ln(# service connections), 2005	0.027	***	0.007	0.024	***	0.007	0.031	***	0.007	0.027	***	0.007
% Unmetered service connections	-0.546	***	0.069	-0.509	***	0.070	-0.577	***	0.070	-0.514	***	0.070
ln(Median household income)	0.223	***	0.029	0.196	***	0.031
Motivation
Governance: Special district	0.050	**	0.020	0.051	***	0.020	0.059	***	0.020	0.060	***	0.020
Governance: Private owner	0.011		0.026	0.008		0.027	-0.002		0.027	-0.003		0.027
South-of-Delta water importer							0.064	***	0.020	0.043	**	0.020
# Customer complaints per connection	3.387	***	0.929	3.192	***	0.926	3.477	***	0.949	3.159	***	0.940
% Neighbors with PCWR^a				0.204	***	0.077				0.326	***	0.075
External
Avg. annual max. temperature, °C	0.0003		0.002	0.001		0.002	-0.002		0.002	0.000		0.002
Year fixed effects		Yes			Yes			Yes			Yes
Log likelihood		-1,477			-1,473			-1,500			-1,490
LR χ²		338			345			293			311
Prob> χ²		0.000			0.000			0.000			0.000
McFadden's R²		0.103			0.105			0.089			0.095
N		2,584			2,584			2,584			2,584
# Community Water Systems		323			323			323			323

Delta method standard errors are reported for marginal effects

The dependent variable in all models is an indicator for having a PCWR; the unit of analysis is the utility-year

Delta method standard errors are reported

** Statistically significant at the 5% level; *** Statistically significant at the 1% level

^aRepresents % Neighbors with PCWR in each of the eight hydrologic regions in a given year