Another strength of kernel density estimation is
that it can address problems in geocoding large
amount of criminal incidents, at least to some
extent. As the number of geocoded incidents get
larger, it becomes increasingly difficult to visually
examine the spatial patterns of the incidents with a
point-pattern map, since ‘‘dots conceal other dots,
giving a false impression of the true crime density’’
(McLafferty et al., 2000, p.80). Kernel density
estimation correctly takes such overlapping ‘‘dots’’
into account in producing the patterns of offense
densities.