Probit Regression Results, Average Marginal Effects Reported: Municipal and First Nations Subsets (Equation [3]) and Full Sample with Interaction Effect (Equation [4])
| Variable | First Nations Water Systems (N=145) | Municipal Water Systems (N=565) | Full Sample with Interaction Effect (N=710) |
|---|---|---|---|
| System is supplied through a water-sharing arrangement (WSA) | −0.438*** | 0.034 | −0.075** |
| (0.086) | (0.027) | (0.030) | |
| System is located in a First Nations community (FN) | — | — | 0.218*** |
| (0.054) | |||
| System is located in a First Nations community and supplied through a water-sharing arrangement (FN × WSA) | — | — | −0.331***a |
| (0.086) | |||
| Host community is located in northern Ontario (NORTH) | 0.090 | 0.056* | 0.069* |
| (0.136) | (0.031) | (0.038) | |
| Host community is ≤5 km from closest neighbor with water infrastructure (dumdis1) | −0.139 | −0.018 | −0.051 |
| (0.129) | (0.029) | (0.036) | |
| Host community is >5 km and ≤10 km from closest neighbor with water infrastructure (dumdis2) | −0.080 | −0.019 | −0.034 |
| (0.121) | (0.026) | (0.034) | |
| Natural log of host community population density (100s/km2) (lnPOPDEN) | −0.022 | 0.003 | −0.002 |
| (0.037) | (0.006) | (0.009) | |
| Natural log of regional (census division) median income (CA$1,000s) (lnINC) | 1.580*** | −0.184 | 0.138 |
| (0.601) | (0.118) | (0.170) | |
| System is supplied exclusively by groundwater sources (GW) | −0.334*** | −0.069*** | −0.127*** |
| (0.094) | (0.026) | (0.029) | |
| System is classified as “large residential” (100+ connections) (LARGE) | −0.040 | 0.034* | 0.012 |
| (0.100) | (0.019) | (0.027) | |
| Pseudo R-squared | 0.1398 | 0.1538 | 0.3338 |
Note: Standard errors are in parentheses and are clustered by census subdivision (i.e., community housing the water system). The dependent variable equals 1 if the drinking water advisory was in effect at some point during the study period (2009–2010), 0 otherwise.
↵a Stata does not generate marginal effects for interaction terms (in our case, i.WSA##i.FN) using the standard “margins” command. The marginal effect for this interaction term was calculated separately using the following command: margins WSA, dydx(FN) pwcompare(effects). This marginal effect compares First Nations water systems supplied through water-sharing arrangements to First Nations water systems that are independently supplied.
↵* p < 0.1;
↵** p < 0.05;
↵*** p < 0.01.