You are looking at 1 - 1 of 1 items

  • Author or Editor: Andrea De Montis x
Clear All Modify Search
You do not have access to this content

Andrea De Montis, Simone Caschili and Daniele Trogu

The aim of spatial autocorrelation is to inspect the organization and structure of spatial phenomena. This statistical method allows one to scrutinize spatial patterns and relationships that generate relevant phenomena in territories under consideration. In fact, spatial phenomena are often self-determining and may positively or negatively influence adjacent units. Autocorrelation analysis pinpoints the mutual influence among spatial units as well as any polarizing effect around a specific area. In this chapter we use spatial autocorrelation analysis to scrutinize the level of spatial dependency of accessibility for commuters in continental US counties on the theoretical basis described above. Counties’ accessibility is computed according to gravity theory (spatial interaction models with impedance function obeying to exponential and power decay impendence functions). The scope of our study is to understand whether space influences counties’ accessibility. We first apply global and local univariate spatial autocorrelation analysis in order to test the hypothesis that counties with higher accessibility positively influence adjacent counties. Finally we test whether accessibility is spatially correlated with socio-demographic variables, such as residential population and income per capita. We develop a bivariate autocorrelation analysis to assess spatial dependencies between the accessibility of US counties and socio-economic variables both at the global and at the local level.