Although there is a great deal of literature detailing changes in the responses of small mammals at habitat edges, few studies measure spatial use, and even fewer agree on a definition of edge. This lack of consistency in edge definition likely results from a difficulty in defining edge habitat. Many edge studies have arbitrarily defined edge habitat by using trap spacing, sampling method, or some other single vegetation characteristic to delineate edge from interior. The objective of this study was to determine whether space use by the white-footed mouse (Peromyscus leucopus) varied in forest adjacent to agriculture along a habitat gradient from forest edge to forest interior. In this way, edge was not artificially defined, but rather linked spatial changes exhibited by the white-footed mouse to existing vegetation characteristics along an edge-to-interior gradient. We also compared the results from our gradient model to results from a model of different artificially defined edges. Two and three dimensional space use by female P. leucopus increased along a gradient of changing edge-to-interior habitat variables while male space use did not change. Our measured artificial definitions of edge yielded differing results for each edge definition emphasizing this particular method’s measurement bias. Modeling species variables along an edge-to-interior gradient may provide a more accurate method for measuring edge effects.