The Impacts of the Multiscale Built Environment on Commuting Mode Choice: Spatial Heterogeneity, Moderating Effects, and Implications for Demand Estimation

Table 4

Parameter estimation results of BM.

Variables	Nonmotorized		Public transit
	Parameters	t-test	Parameters	t-test
City level variables
GDP	−0.131	−0.78	−0.364	−1.84
Urbanization rate (reference: <50%)
50%–75%	0.210	0.79	0.926	2.87
≥75%	0.188	0.48	0.578	1.28
City population density	−1.289	−1.15	−2.087	−1.81
Number of public transportation vehicles	0.029	1.57	0.087	4.29
Metro operating mileage (reference: 0 km)
0–100 km	0.384	1.05	1.163	2.56
≥100 km	1.089	1.37	2.268	4.21
Road area per capita	−0.043	−3.31	−0.033	−2.21
Neighborhood level variables
Neighborhood population density (reference: <1000 persons/km^2)
1000–5000 persons/km^2	0.114	0.54	0.774	2.91
5000–10000 persons/km^2	0.215	0.81	1.039	3.35
≥10000 persons/km^2	0.240	0.92	0.757	2.44
Land-use diversity	−0.172	−3.08	−0.161	−2.58
Green coverage	0.011	3.73	0.012	3.38
Air quality (reference: not good)	0.535	3.29	0.934	4.18
Neighborhood location (reference: suburban)	0.550	2.84	1.171	5.31
Individual level variables
Age	0.034	9.91	0.009	1.83
Gender (reference: female)	−0.728	−9.27	−0.561	−5.10
Hukou type (reference: nonlocal hukou)	−0.514	−4.24	−0.034	−2.22
Income	−0.070	−6.09	−0.033	−2.90
Car ownership (reference: without car)	−0.810	−9.23	−0.955	−7.79
Separation of workplace and residence	−0.613	−6.94	0.630	5.18
Commuting time	−0.014	−9.59	0.017	12.61
Constant	2.594	2.41	0.619	0.49
Statistical indicator
City-level variance	0.262
Neighborhood-level variance	0.486
Log likelihood	−3639.338
AIC	7386.675