Different neighborhood walkability indexes for active commuting to school are necessary for urban and rural children and adolescents

Table 5 Performance of regression models between walkability index and ACS for urban and rural adolescents, controlling for gender, age, and SES

No. variables	Urban adolescents				Rural adolescents
No. variables	Walkability index^a	MSE	RMSE	r-value	Walkability index^b	MSE	RMSE	r-value
2	[Crossing quality + Distance to school]	7.28	2.7	0.35	[Intersection density + Distance to school]	14.77	3.84	0.35
3	[Residential density + Distance to school + Crossing quality]	7.21	2.69	0.36	[Distance to school + No. crossings + Intersection density]	14.79	3.84	0.35
4	[Distance to school + No. four-way intersections + Land use mix + Crossing quality]	7.13	2.67	0.38	[Intersection density + Distance to school + No. crossings + Residential density]	15.06	3.88	0.33
5	[Residential density + Distance to school + No. four-way intersections + Land use mix + Crossing quality]	7.07	2.66	0.39	[Intersection density + Distance to school + No. crossings + Positive streetscape characteristics + No. traffic lanes]	15.21	3.9	0.31
6	[Residential density + Distance to school + No. four-way intersections + Land use mix + Positive streetscape characteristics + Crossing quality]	7.08	2.66	0.38	[Residential density + Distance to school + No. traffic lanes + Positive streetscape characteristics + No. four way intersections + Crossing quality]	15.38	3.92	0.30

Notes. Normalized values (range normalization) of the variables were used. MSE Mean square error, RMSE Root mean square error. ^aThe value of the variable Distance to school has been inverted according to the Beta values found in Table 2. ^bThe value of the variables Intersection density, No. four way intersections, Residential density, Land use mix, Distance to school and No. traffic lanes have been inverted according to the Beta values found in Table 2

ISSN: 1479-5868