Examining the spatial dependency between shared accommodations and points of interest

Abstract

This study proposes a three-step analytical framework to examine the proximity between Airbnb listings and points of interest (POIs), and to evaluate the spatial dependence between them across dimensions. Grounded in Tobler's first law (TFL) of geography, this framework integrates standard deviational ellipse, average nearest neighbor, and geographically weighted regression methods. The framework is validated using data from 14 U.S. cities of varying sizes, encompassing 122,005 Airbnb listings and 113,169 POIs. Although the results reveal a notable proximity of Airbnb listings to POIs, the degree of spatial dependency varies across cities and property types. Notably, several frequently overlooked POIs, such as schools, post boxes, bakeries, and bicycle rental stations, demonstrate significant relationships with Airbnb listings. This research advances the literature on shared accommodations by elucidating the spatial relationship between Airbnb and POIs, introducing new location factors, and providing a robust analytical framework. The findings provide valuable practical implications for enhancing business and management strategies for Airbnb stakeholders, while balancing the growth of the shared accommodation sector against its effects on local communities.