A Clustering Approach to Identify High-Risk Taxi Drivers Based on Self-Reported Driving Behavior

Table 9

Second Logistic regression model.

Model	Step	Predictor	B	value	Odds	-2LL	CHI2
High/medium risk drivers vs. low-risk drivers	1	Annual mileage	0.46	0.001	1.63	650.31	81.07
	2	Lapses	0.76	0.460	0.46	464.99	266.39
		Inattention errors	1.49	0.000	4.47
		Ordinary violations	1.01	0.068	2.76
		Inexperience errors	−0.54	0.120	0.57
		Aggressive violations	1.58	0.000	4.85