ABSTRACT
This chapter focuses on identifying the "spatial footprint" of potential cluster areas (PCAs) in a three state region: Indiana, Michigan, and Ohio. It illustrates the use of Moran's I in the identification of PCAs. The chapter discusses the benefits and limitations of using Local Indicators of Spatial Association (LISA) to identify clusters. Floriculture was selected because of an ongoing cluster-based project, known as "Maumee Valley Growers". The project has been developing since 2003 to assist greenhouse growers in north-western Ohio cope with increasing Canadian imports and other competitive challenges. The chapter shows that how Local Index of Spatial Autocorrelation (LISA) can be used to identify PCAs. In order to determine the level of spatial autocorrelation locally, a spatial autocorrelation value is computed for each areal unit, which enables the user to determine how similar or dissimilar each county is to its surrounding neighbours. Identifying regions that have potential for industrial clustering is often a laborious sort through county data tables.
