Assessing the Effects of Centroid Assignment Methods on Measuring Accessibility

Keywords: Centroid, Accessibility, Travel cost, GIS
Abstract Type: Paper Abstract

Abstract

Presumably, the center of a given polygon can vary depending on how the centroid is defined for that polygon. In an abstract sense, researchers have assumed that a population-based centroid better represents a given polygon than a geometric centroid because it accounts for population distribution. In applications, such as when measuring place-based accessibility, using a geometric centroid might be misleading because this practice could be overestimating travel costs in large polygons; however, this assumption has not been quantitatively tested. In this paper, we examine the role of centroid assignment methods by comparing the accessibility values of three different centroid assignment approaches. The analysis revealed that differences in accessibility among centroid methods are negatively related to the residential area ratio, and that lower residential area ratios exhibited greater differences in accessibility than higher ones. In addition, the geometric centroid method was more likely to overemphasize accessibility disparities than other assignment methods. The findings confirm that the geometric method overestimates travel costs more than the population-based methods and suggest that accessibility researchers can still use the geometric method in lieu of population-based methods when polygons have dense residential areas and other population data are unavailable.