Research Article
Identifying High-Risk Intersections for Walking and Bicycling Using Multiple Data Sources in the City of San Diego
Table 1
Negative binomial regression model for pedestrians.
| Variable | Estimates | t-stat | Sig. |
| (Intercept) | 4.865 | 23.24 | 0.000 | Transit stop density (0.5 miles) | 8.782 | 3.81 | 0.000 | Percentage of regular transit rider, pedestrian, or bicyclist population (0.25 miles) | 3.717 | 1.95 | 0.051 | Employment density (0.25 miles) | 0.051 | 2.41 | 0.016 | Maximum speed limit within the intersection less than 40 mph | 1.135 | 5.37 | 0.000 | Percentage of vacant housing units (0.5 miles) | −3.517 | −2.99 | 0.003 | Total commercial or mixed-use land area (0.1 mile) | 0.190 | 5.01 | 0.000 | If the area contains a higher crime count than the average crime counts among the buffers (0.25 miles) | −0.292 | −1.65 | 0.098 |
| N | 45 | R-squared | 0.70 | RMSE | 1633.24 | MAE | 1147.41 |
|
|