Determinants of active school travel levels among children: A case study in a mountainous city

Abstract

Introduction

Children's active school travel (AST) is crucial for their daily physical activity, especially in mountainous cities where terrain complexities such as slope and road density are closely associated with safety and efficiency. Understanding these relationships is essential for preventing childhood obesity, reducing depression, and enhancing independence. However, AST in mountainous cities has received relatively little attention.

Methods

This study explores the factors associated with children's AST levels in Guiyang, a typical mountainous city in China. Utilizing data from 2021 resident travel survey, we calculate the daily cumulative time spent walking or cycling to and from school as a quantitative measure of AST levels by analyzing children's complete travel chains. This variable serves as the dependent variable in our analysis. A Gradient Boosting Regression Tree (GBRT) model is employed to capture the nonlinear and interactive relationships between children's AST levels and their socioeconomic characteristics, built environment, and topographical features.

Results

The results indicate significant nonlinear associations and threshold effects among the examined factors. In particular, topographical features show strong associations with variations in children's AST levels.

Conclusions

By revealing critical nonlinear relationships and threshold effects related to built environment and topographical factors, this study provides actionable, data-driven insights. These findings can inform more targeted infrastructure and road network strategies to enhance AST among children in mountainous cities.