TABLE 2.
Robustness of Diffusion via Roads: Other “Proximity” Measures (Hazard ratios)
| Model 1 Trade diffusion | Model 2 Regional news coverage | Model 3 Develop- mental emulation | Model 4 Civiliza- tional emulation | Model 5 Emulation among legal families | |
|---|---|---|---|---|---|
| Vulnerability to externalities (Neighbor’s Policy Weighted by Sum of Roads) (s) | 1.02*# | 1.03*# | 1.03*# | 1.02*# | 1.03*# |
| U.S. Pressure | 1.33 | 2.38* | 2.27* | 2.29* | 2.25* |
| Rule of Law (s) | 1.74*# | 1.44*# | 1.50*# | 1.22# | 1.48*# |
| Policy Weighted by Trade Partners (s) | 1.34# | -- | -- | -- | -- |
| Regional Press Stories on Human Trafficking (s) | -- | 0.96# | -- | -- | -- |
| Criminalization Density within Developmental Level (s) | -- | -- | 0.98# | -- | -- |
| Criminalization Density among Civilizational Group (s) | -- | -- | -- | 1.04*# | -- |
| Criminalization Density among Legal Family (s) | -- | -- | -- | -- | 1.01# |
Note: Results are from a Cox-Proportional Hazards Model. Variable names followed by (s) are fit using smoothing splines. The reported values are either the hazard ratios, or the best linear approximation to the hazard ratios. All results based on two-tailed tests.
*
= linear effect significant at the .05 level.
#
= non-linear effect significant at the .05 level.