Table 7. Associations of psychosocial and environmental variables and the interaction terms with passive transport.
| Passive transport to work | Passive transport to other destinations | |||
|---|---|---|---|---|
| Logit model: OR of being non-participant in passive transporta (95% CI) | Negative binomial model: minutes/week passive transport useb (95% CI) | Logit model: OR of being non-participant in passive transporta (95% CI) | Negative binomial model: minutes/week passive transport useb (95% CI) | |
| Psychosocial | ||||
| social norm | 2.13 (1.08, 4.23)*, c | 1.18 (1.09, 1.28)*** | 0.63 (0.42, 0.96)* | 1.21 (1.09, 1.34)*** |
| Modeling | 0.66 (0.44, 0.99)* | |||
| perceived benefits | 0.60 (0.33, 1.08) | 0.43 (0.23, 0.79)** | ||
| perceived barriers | 0.59 (0.31, 1.14) | 0.74 (0.56, 0.99)*, c | ||
| Evironmental | ||||
| residential density | 1.78 (1.02, 3.12)* | |||
| land use mix diversity | 0.83 (0.74, 0.94)** | |||
| land use mix access | 3.06 (1.29, 7.24)* | 0.92 (0.71, 1.19) | ||
| safety from traffic | 6.30 (1.19, 33.31)*, c | 5.26 (1.01, 27.54)*, c | ||
| safety from crime | 4.15 (1.82, 9.45)*** | 1.35 (1.01, 1.80)*, c | 0.72 (0.58, 0.90)** | |
| facilities at work | 2.98 (0.62, 14.41) | 0.64 (0.43, 0.95)* | ||
| Distance | 0.98 (0.95, 1.00) | 1.00 (0.99, 1.00)c | ||
| Interaction terms | ||||
| social norm*educational level | 0.30 (0.13, 0.69)** | |||
| perceived barriers*educational level | 1.66 (1.17, 2.35)** | |||
| safety from traffic*educational level | 0.02 (0.00, 0.16)*** | 0.08 (0.01, 0.60)* | ||
| safety from crime*educational level | 0.59 (0.42, 0.83)** | |||
| distance*educational level | 1.02 (1.02, 1.03)*** | |||
OR = odds ratio
CI = confidence interval
* p<0.05
** p<0.01
*** p<0.001.
a Logit model: ZINB models evaluate the correlates of the odds of non-participation in passive transport to work or to other destinations.
b Negative binomial model: simultaneously, among participants who did use passive transport to work or to other destinations, ZINB models evaluate the correlates of weekly minutes passive transport use to work or to other destinations. Negative binomial model parameters represent the proportional increase in minutes/week passive transport use to work or to other destinations with a one-unit increase in the predictor. The model for commuting was adjusted for distance to work.
c Reference category is non-college educated