The first map shows four chloropleth maps of Escambia county using different data classification methods: Equal Intervals, Quantile, Standard Deviation and Natural Breaks.
Which Classification best represents the data and why?
To answer the question I looked at the way the data was spread using the Classify button under the Symbology tab and noted the leftward skew of the data indicating that the majority of census tracks in Escambia county have low population percentages of Africans (36 of 58 are roughly 25% or less) while three have very high percentages (86%, 88.1%, 95.6%). As such, the best map of the data should highlight these two extremes, especially the high percentage tracks.
Examining the distributions created by each of the four Data Classification methods helped to identify which classification best displayed those aspects. The Quantile method was rejected because it spanned too great a range of values, 44.6 – 95.6 (over 50% of the total range), into one Class break. The Standard Deviation method was rejected because the distribution of the data was not normally distributed. The Natural Break method was rejected because it did not isolate the high percentage values, even after adding up to 13 breaks.
The Equal Interval classification seemed the best fit because its legend isolated most of the tracks under 25% into one interval and had a higher starting percentage for the upper range, much closer in value to the high percentage tracks, making them easily identified on the map and providing a better indicator of their outlying nature.
Very nice Sean! Good logic! Try and reserve blue for water in future maps.
ReplyDelete