Correlates of non-concordance between perceived and objective measures of walkability

Abstract

Background: Objective and self-reported physical environmental attributes have been related to physical activity.

Purpose: We examined the characteristics of adults who are resident in objectively identified high walkable neighborhoods but whose perceptions of neighborhood attributes are not concordant with objective attributes relating to high walkability.

Methods: Neighborhood built-environment attributes relating to walkability (dwelling density, intersection density, land use mix, and net retail area) were determined objectively, using Geographic Information System databases; data on corresponding perceptions of local environment attributes (from the Neighborhood Environment Walkability Scale) were derived from a self-completion survey of a socially diverse sample of 2,650 adults aged 19 to 65. Objective and perceived walkability attributes were categorized using median splits, and correlates of non-concordance were determined using multiple logistic regression models.

Results: There was a fair overall agreement between objectively determined walkability and perceived walkability (Kappa = 0.35, 95% CI = 0.31-0.39). Among those resident in objectively assessed high walkable areas (n = 1,063), 32.1% perceived them to be low walkable; conversely, 32.7% (n = 1,021) resident in objectively determined low walkability areas perceived them to be high. For residents of objectively determined high walkable areas, the characteristics that differentiated those with perceptions of low walkability (non-concordant perceptions) from those with concordant perceptions of high walkability were: not being university-educated (OR = 1.47, 95% CI = 1.06-2.04); having lower household incomes (OR = 1.54, 95% CI = 1.09-2.17); being overweight (OR = 1.46, 95% CI = 1.03-2.07); and walking fewer days per week for transport (OR = 1.75, 95% CI = 1.11-2.70). Higher walking times and more positive cognitive variables were noted among participants who lived in a neighborhood with low walkability that was perceived as high compared to those who lived in a high walkable environment that was perceived as low walkable.

Conclusion: Adults with lower educational attainment and lower incomes, who were overweight, or who were less physically active for transportation purposes, were more likely to misperceive their high walkable neighborhood as low walkable. There is the potential for physical activity promotion and persuasion strategies to address non-concordant perceptions, especially among those who live in high walkable environments but perceive them to be low and also among those who are socially disadvantaged and are less active. Perceptions of environmental attributes may be more strongly correlated with cognitive antecedents and with behavior than are objective measures.