Module 4 introduces us to Data Classification methods, specifically, Equal Interval, Quantile, Standard Deviation and Natural Breaks. Equal Interval and Quantile are best used with data that is more uniform mainly because of the way they both derive the classes for mapping. Natural Breaks takes the distribution of the data into account when creating classes, and Standard Deviation uses statistics to find the average (mean) and creates classes based on the standard deviation from the mean.
I elected to present the four classification methods via a four panel map, all at 1:600,000 scale for uniformity and the ease of only having to make one scale bar. As stated previously, Equal Interval and Quantile use a derived method of classification and it clearly shows in the mapped data. The equal interval creates classes that all have the same range value. With this data, there is no way to determine how the information shows groups or clusters of persons in the census tracts. The same is true for the Quantile method, which uses observations as the metric for creating classes. This method indicates more population density than the Equal Interval method, but because the classes are derived from an arbitrary number of observations, the range in each class is not standard and it requires the map user to see the actual census tract data to understand what information is being displayed concerning the density of senior citizens.
The Natural Breaks method by design aligns with the range of data, and uses algorithms to find the natural breaks as well as make the values in each class as similar as possible, unlike the Quantile method. This provides the map reader with a more accurate representation of the density of senior citizens in Miami-Dade County.
The Standard Deviation method is based on statistical measurements, that determines the average and standard deviation and creates classes based on the distance from the average. This method, when using normalized data, provides not only accurate information, it provides a useful and intuitive way to understand the mapped data. The text and lecture recommended using a diverging color scheme to visually see which areas are less than or equal to the average, and which areas are greater than or equal to the average value. I elected to add the mean and standard deviation value to aid the map reader with data comprehension.
While creating the legend, I discovered that ArcGIS Pro layout format is not WYSIWYG (whiz e wig, or What You See Is What You Get) when attempting to clean up the Legend by removing the title, MiamiDadeCensus, from being listed in the legend. When I removed the title, I discovered that the software replaced it with the word Layer, when it was exported, but not when it was viewed. I was able to go into the legend properties and deselect the layer level from displaying.
Creating this map helped me to get a better understanding of how important it is to plan the map based on the end user. Data classification is not a simple plug and play measure, it requires some understanding of which method works best to convey the desired message to the map reader.
Comments
Post a Comment