Since walkability plays an important role in active commuting, several cities are actively promoting its integration into urban and environmental planning policies. This study examined the association between walkability and active commuting in Seoul, Korea. A multilevel logistic regression model was used to examine the correlation between Walkability Score and the probability of active commuting after controlling for individual variables. The analysis used 129,044 individual samples nested within 424 administrative districts (dongs). In this study, three models were tested: Model 1 contained only individual variables, Model 2 contained individual variables and Walkability Score, and Model 3 included neighborhood-level variables in addition to the variables of Model 2. The results showed that the Walkability Score was significantly correlated with the odds of active commuting. Specifically, every additional one-point increase in Walkability Score was associated with 0.3% higher odds of active commuting (Model 2: odds ratio (OR) = 1.003, 95% confidence interval (CI) = 1.001-1.005; Model 3: OR = 1.003, 95% CI = 1.001-1.006). Additionally, public transportation density was also positively correlated with the odds of active commuting. The odds of active commuting were positively correlated with younger age, female, lower-income, and having no car. Based on the findings, policy recommendations in urban planning and design, transport engineering, and environmental planning are provided.
Keywords: Seoul; Walk Score; Walkability Score, multilevel logistic regression model; active commuting; cycling; geographic information system (GIS); walking.