The bivariate spatial relationship between perinatal deaths
and livebirths was studied using Moran’s bivariate. Local spatial
autocorrelation was analysed in this study. Moran’s diagram
shows the types of spatial relationship between a unit of
place and the neighbouring units. This allows visualization of
four types of local spatial associations between an observation
point and its neighbours, each located in a quadrant of the
scatterplot. GeoDa can create a local indicator of spatial autocorrelation
(LISA) cluster map for identifying local clusters and
spatial outliers of a geo-referenced variable. For each region, the
local Moran statistic is computed and tested for significance by
randomization. Significant local clusters and spatial outliers
are portrayed in different colours. Areas with negative local
Moran coefficients originating from high and low surrounding
values are termed highelow outliers. They are coloured light