It is only recently that the fuzzy-set theoretic approach to spatial objects and their concepts has joined the mainstream of geographic information system (GIS) and science (GIScience). Several reasons account for this. Recent research on spatial objects has revealed that spatial vagueness is inherent in some geographic features (Burrough and Frank 1996). For instance, the boundary between a mountain and a valley is not sharply defined. Furthermore, even if a geographic phenomenon is best described as crisp, humans tend not to reason in a precise manner, but rather in an approximate manner (e.g. they live near Chicago). Moreover, perception and cognition vary widely between individuals. Furthermore, information can be incomplete or imprecise due to rough measurements or to our incomplete ability to grasp the scope and detail of spatial objects. In other words, there always exists a gap between the reality and its representation. We use fuzzy set theory (Zadeh 1965) as a mean to reconcile discrepancies existing between reality and its representation. We discern three representational different levels at which fuzzy set concepts can be applied, namely the ontology, perception, and implementation levels. The (spatial) ontology level pertains to generic concepts inherent in spatial objects. The perception level concerns the mental models used to perceive the environment. The implementation level encompasses the errors that have propagated during system implementation. The combination of spatial vagueness, diverse human perceptions, and implementation errors account for the gap existing between reality and its representation. In general, fuzzy set concepts preserve details (Robinson 2002) whereas traditional (crisp) GIS data models overlook the loss of information by forcing reality into a coarse (in the sense of low resolution) representation. Fuzzy set theory can overcome the gap by providing mechanisms for ontologically and cognitively plausible (Worboys 2001) and error-sensitive (Duckham et al. 2001) representation of the reality. In sum, fuzzy set theory provides a means to address various kinds of uncertainty such as spatial vagueness, human perception, and imperfect information. This study is part of a larger project aimed at geographically referencing the fatal accident data. Our task is to pinpoint the location where a traffic crash is most likely to have occurred given the limited and imprecise information available on this crash. In our study, georeferencing can be roughly defined as the conversion of the linguistic description of a location to a quantitative specification. As Goodchild (2000) pointed out, effective georeferencing can be a matter of life and death in the case of communication between a caller and an emergency dispatcher. The linguistic description of location is sometimes not clear-cut, not only because many alternate names are used to refer to the same location, but also because the location itself is not well defined. We focus on the problem of determining the location of a certain locality. We hypothesize that location indeterminacy of localities is caused by spatial vagueness, interpersonal differences in perception, and imperfect information. We compute the value that quantifies location indeterminacy by modeling the indeterminate part of localities by a fuzzy set membership function. We examine the relationship between the value of location indeterminacy and attributes of localities in order to test the stated hypotheses. The purpose of this research is to show how fuzzy set theory can be properly applied in modeling localities. Also the result will assure whether there exists fuzziness in determining the location of locality. This study develops a fuzzy set membership function for indeterminate boundaries of localities. By testing our hypotheses on the relationship between location determinacy and characteristics of locality, we examine whether fuzzy set theories can capture various kinds of uncertainty at the ontology, perception, and implementation levels. Modeling localities by fuzzy sets has a definite advantage over a crisp set in that it makes best possible use of sparse information to reconstitute detail. More specifically, fuzzy-set-based localities constitute a closer depiction of reality, such as overlapping memberships of localities. Next, fuzzy set provides a conservative representation tool for individual differences in the perception. Finally, allowing the soft processing (fuzzy set modeling) over the hard data (reference data) can minimize the problems caused by the imperfection of source data. The remainder of this chapter is organized as follows. In Section 4.2, the specific georeferencing motivating this study is described. We formulate research hypotheses and specify the assumptions on which this study is based. In Section 4.3, we give a brief overview of related research, such as the ontology of spatial objects, the representation of fuzzy regions, and the notion of nearness. In Section 4.4, we define the fuzzy set membership function of localities. The implementation steps in GIS are described in Section 4.5. The analyses of results are given in Section 4.6. We examine if fuzziness is substantial in identifying localities. By looking at the cases that are georeferenced by a fuzzy set modeling, we may or may not find evidence of fuzziness in locality. The hypothesis is examined also. Finally, Section 4.7 concludes this study.