Examining Spatial Statistics within the Western Greater Toronto Area

The purpose of this project was to examine and apply spatial statistical methods. The data set used was educational attainment divided by dissemination area (DA) for the Western Greater Toronto Area. The population with a bachelors degree or higher and no certificate diploma or degree were examined. These variables were normalized to total population aged 15 or older.

Several statistical methods were examined. Spatial autocorrelation was examined using Moran’s I. This method can be used to examine the similarity of nearby continuous features. This method calculates the difference between the target feature and the mean for all features, and the difference between each neighbour and the mean. The results can then be used to determine if the pattern is dispersed or clustered.

The General G-Statistic examines for clusters and then identifies if they are low or high values. The result does not identify the location of clusters, only that high or low values tend to cluster near another.

The hot spot analysis was also used in order to identify clusters and outliers. The tool used for this analysis was the Anselin Local Morans I tool. This tool identifies the location of clusters as well as if the clusters are of high or low values. The ArcMap Hot Spot Analysis (Getis-Ord Gi*) and the Optimized Hot Spot Analysis tool was also used. These tools identify both clusters as well as identifying if the clusters are of high or low values.

Geographic distribution was also measured using the standard distance and standard deviational ellipse. These tools determine the dispersion fo values around the mean centre. These tools show directional trends within the data.

Click here to view the pdf poster presentation of this project.

A03 Mississauga_SpatialStatistics