Each location was then plotted by highest correlation frequency, allowing for visualisation in the spread of locational correlations for each location. It was hence decided that locations that did not have all neighbouring locations present were excluded, as this lack in data could cause a potential skew in our results, as potential correlations could not occur. Therefore locations 7, 8, 9, 12, 13, 14, 17, 18, 19 were chosen to continue with analysis.
Sample data for the spread of values in correlations and degree of noise were created to generate larger datasets, ensuing confidence in our resulting data. These were validated using a t-test. A Shapiro-Wilks test was performed to assess distribution. Both variables indicated normal distribution, however original correlation values, when plotted, showed to have a slight skew to the left, and so the degree of noise was accepted as our variable for analysis, due to its visual confidence and clear statistical normal distribution.