Data Classification

Q: Which classification do you think best represents the data and why?

I think the Natural Breaks classification fits the data best.

Equal Intervals doesn’t make sense because the distribution of African Americans should vary percentage-wise across the county. Some geographical areas will have a higher percentage of African Americas.

Quantile classification also fails to consider how the data (percentage of African Americans) is distributed. The enumeration units will not be of the same size for percent of population data, and although the map looks great, it is stretching and squeezing the data to give equal map area.

Standard deviation only works when the data is normally distributed.  This data is not.

Natural Breaks classification works well to represent this data.   Determining the breaks is fairly arbitrary (what really differentiates high percentage from medium-high percentage in practical, street-level terms?) but this classification is sensitive to distribution.