Background:
Walking to school can be an important source of physical activity for school-aged children. It is also an affordable and environmentally clean mode of travel that can serve health, equity and mobility purposes. In reality, however, it is underutilized due to personal barriers, safety concerns, long distances to schools, and poor infrastructure conditions.
Objectives:
This cross-sectional study examines the prevalence and correlates of elementary school children's walking to school behaviors, and their implications for health and equity. It focuses on low-income, Hispanic children, who may have no means to get to school other than walking, while being exposed to polluted and unsafe environments.
Methods:
A "Safe Routes to School" survey was sent out in April, 2007 to 4,759 parents/guardians of students from eight elementary schools in Austin, Texas. The selected schools are from low-income neighborhoods (over 90% of the students receiving free or reduced-price lunch) with high percentages (66-98%) of Hispanic students. The survey collected data about the children's typical school travel modes and the parental considerations in choosing their child's travel mode. A total of 1,281 completed surveys were returned and analyzed.
Descriptive statistics were used to analyze travel mode share and travel time. Bivariate analyses explored disparities in environmental supports for walking. Factor variables were extracted for perception and attitude variables. Binary logistic regressions were used to predict the odds of walking as a typical mode, with personal, social and physical environmental variables. Included were eight individual models (one for each school) and one pooled model with each school entered as an indicator variable.
Results:
Walking accounted for 28% and 34% of mode shares for the trip to and from school, respectively, and was mostly accompanied by an adult. Biking was rarely used as a travel mode (0.6%). Most walking, biking, school bus and private vehicle trips (63-88%) took less than 15 minutes. Public transit trips were the longest.
Parent's education level was associated positively with car ownership and safety concerns and negatively with the quality of overall walking environment. Parents' safety concerns showed no significant difference across schools, although the objective crash and crime rates were significantly different. This implies that perceived safety is not necessarily dictated by objective data.
From the pooled regression model, parent's education level, car ownership and personal barriers were negative correlates while children's and parents' affirmative walking attitude was a positive correlate of walking. Students from Blanton Elementary or those eligible for school bus service were less likely to walk to school. Positive social influence (frequent walking by other children and parents) was positively associated with walking. Parents' perception of proximity to school was the strongest predictor of walking (OR = 4.918, p < 0.01). Parents' safety concerns and presence of highway/freeway, convenience stores or office buildings along the route to school were negatively associated with walking. Sidewalk quality and overall walking environment were insignificant, possibly due to their lack of variability.
Similar patterns were revealed in the eight individual models. Negative correlates were parent's education level, car ownership, parental barriers, school bus service, safety concerns, and the presence of highway/freeway and convenience stores. Positive correlates included positive walking attitude and close distance to school. Variables that became significant were age (positive), single-parent status (negative) and busy roads (negative). Sidewalk quality was a negative correlate in Harris Elementary School, possibly because more parents are concerned about sidewalk conditions when a high percentage (45%) of their children walk to school and are often forced to do so (93% with no school bus service and 16% of families with no cars). In addition, peer influence became insignificant in all individual models.
Conclusions:
Findings have important implications for policy, practice and research. School sitting emerged as the key issue because it determines the travel distance and the presence of highway/freeway along the route, which were the strongest predictors of walking to school. Current policies such as minimum school size and incentives for new school construction may be pushing schools away from where children live, and thereby deter walking to school. Secondly, parents' personal barriers and perceived safety were critical, suggesting the importance of educational and other assistance programs in addition to physical improvements. Third, the likely differences in the correlates of walking among adults and children call for tailored efforts responding to the specific needs of the target population. Finally, findings suggest that underprivileged children walk more but do so in unsafe environments with poor pedestrian infrastructure. Health and disparity issues of school transportation should be incorporated into future policy and intervention efforts.
Support:
This study is supported by Active Living Research Dissertation Grant.
