The association between sidewalk length and walking for different purposes in established neighborhoods

Table 3 Probit and linear regression estimates for the association between demographic, built environment, reasons for moving to a new neighborhood (preferences), and attitude and neighborhood-based recreation walking

	Probit models for participation in recreation walking in the neighborhood (n = 1681)			Linear models for weekly minutes of recreation walking in the neighborhood (walkers only n = 892)
	Probit (95CI) ¹	Probit (95CI) ²	Marginal (%) ²	B (95CI) ¹	B (95CI) ³	Heckman-corrected B (95CI) ³
Built environment
Walkability index	0.009 (−0.027, 0.046)	0.013 (−0.024, 0.051)	0.48	−1.58 (−5.40, 2.24)	−0.96 (−4.75, 2.82)	−1.04 (−6.08, 4.01)
Sidewalk length (per 10 km)	−0.002 (−0.050, 0.046)	−0.013 (−0.063, 0.037)	−0.47	2.65 (−2.43, 7.73)	1.61 (−3.46, 6.67)	1.64 (−2.17, 5.45)
Attitude and neighborhood preferences
Attitude towards walking		0.076 (0.059, 0.093)*	2.79		3.19 (1.25, 5.13)*	2.79 (−1.01, 6.58)
Access to recreation		0.253 (0.150, 0.356)*	9.29
Access to schools		−0.075 (−0.145, -0.005)*	−2.75		−8.29 (−15.19, -1.39)*	−7.96 (−15.31, - −0.61)*
Access to services		−0.073 (−0.182, 0.035)	−2.69		−11.97 (−22.77, -1.17)*	−11.75 (−22.63, - −0.87)*
Streets pedestrian/cycle friendly		0.198 (0.091, 0.306)*	7.27		13.65 (4.18, 23.11)*	12.59 (−0.15, 25.32)^Ω
Housing affordability/variety		−0.151 (−0.270, −0.032)*	−5.54
Inverse Mills Ratio						−8.04
Rho						−0.08

¹Adjusted for gender, age, education, walkability and sidewalks.
²Adjusted for gender, age, education, walkability and sidewalks, neighborhood preferences, and attitude correlates.
³Adjusted for gender, age, education, walkability and sidewalks and for preference and attitude variables with statistically significant bivariate associations with walking.
B: Unstandardized regression coefficient. 95CI: 95% confidence interval. *p < .05; ^Ωp < .10.