Spatial Clustering and Locational Correlates of Coworking Spaces in Jeddah, Saudi Arabia: A Multiscale GIS and Logistic Regression Analysis

Coworking spaces have become an increasingly important component of contemporary urban economies, yet their spatial organisation remains poorly understood in rapidly developing Gulf cities. This study examines the spatial clustering and urban correlates of coworking space locations in Jeddah, Saudi Arabia. A dataset of 53 coworking spaces was integrated with points of interest, built-up land, and road-network data within a GIS framework. Spatial pattern was evaluated using Monte Carlo nearest-neighbour analysis, kernel density estimation, and Ripley’s L-function. A 1 km grid was subsequently used to model coworking space presence through binary logistic generalised linear regression, with a 2 km grid used for sensitivity analysis. Coworking spaces exhibited pronounced spatial clustering: the observed mean nearest-neighbour distance was 1023 m compared with 2667 m under complete spatial randomness (nearest-neighbour ratio = 0.384, Monte Carlo p = 0.001). The observed centred Ripley’s L-function exceeded the upper pointwise 95% CSR simulation envelope throughout the evaluated 0.25–10 km range, indicating multiscale spatial concentration. In the final 1 km model, business density (OR = 2.13, p < 0.001), built-up proportion (OR = 1.042 per percentage point, p < 0.001), and road density (OR = 1.075, p = 0.008) were positively associated with coworking space presence. Business density and built-up proportion remained significant at the 2 km scale, whereas the road-density association weakened. These findings suggest that coworking spaces in Jeddah are primarily embedded within business-intensive, highly urbanised, and accessible parts of the metropolitan area while also demonstrating sensitivity to spatial analytical scale.