The extent to which individuals within a population live in groups is often variable, as group formation is most likely to occur when the benefits of group-living outweigh the costs. The distribution of resources can underpin the likelihood of group formation by altering costs associated with individuals aggregating. By influencing the extent of groupliving, resources may also influence relatedness patterns. To address this, we examined relatedness in raccoons (Procyon lotor) by combining spatial and genetic data in an experimental framework wherein resources were manipulated. We tested for differences in patterns of genetic relatedness among 12 sites subject to three treatments: no supplemental food; supplemental food added in a dispersed, non-predictable fashion; and clumped, predictable food supplements resulting in aggregations. We measured mean relatedness for >500 raccoons using blood and hair samples genotyped at 14 nuclear microsatellites and tested for differences between sites and sexes. For two sites, we radio-tracked raccoons and calculated volume of intersection (VI) values, which measure the probability of co-occurrence between individuals' home ranges, and calculated the relationship between VI and relatedness. Telemetry data indicated that raccoons exposed to clumped food shifted their home ranges, in turn perturbing patterns of relatedness between individuals, especially among females. This was exhibited through a positive association between VI and relatedness for females at the site with dispersed food, which was not observed among males or at the site with clumped food. At the landscape level, sites differed significantly in mean relatedness, but there were no clear patterns that corresponded to treatments. Overall, our observations support expectations of the resource dispersion hypothesis, wherein individuals modify ranges to include clumped resources if the costs of tolerating other individuals are low.