GIS5007 - Module 4 - Data Classification

         This module focus on learning the 4 common data classification methods: Quantile, Equal Interval, Standard Deviation, and Natural Breaks. By using one dataset but 4 different classification method, the dataset can be visualized differently. Below is how I interpreted the 4 classification methods.

        Equal Interval divides a range of data values into equal-sized intervals (based on how many classes are given). It is excellent for uniform distributed data where values are spread out evenly. Extreme values are placed in the lowest or highest class. This can hide some values if some are not evenly distributed. 

        Quantile classifications divide a dataset into classes with the same number of data points, regardless of actual values in those data points. This ensures each class has the same number of data points, which can be beneficial for data that is evenly distributed. By dividing the data into equal portions, quantile classification provides a clear ranking of the data from lowest to highest. If the data is not evenly distributed, the resulting classes may have very different numerical ranges, potentially creating misleading maps.

        Standard Deviation is a method that classifies data based on how far each value is from the mean (average), using the standard deviation as the unit of measurement. It is useful for identifying values that are significantly above or below average. It is ideal for normally distributed data (Bell curve shape). It can be misleading with skewed data or non-normal distributions. 

        Natural Break - Rather than forcing equal intervals or equal group sizes, Natural Breaks uses an algorithm to minimize the variance within classes and maximize the difference between classes. This makes each class as internally similar as possible while being different from other classes. Natural breaks excel at highlighting natural groupings or clusters within data. This makes it useful for mapping data where there are distinct areas of high or low values. If the data is heavily skewed, natural breaks may not effectively group values, leading to a distorted representation of the data.

       For this module, I honestly didn't understand how the maps were normalized by the square mile. I might need to do more reading on this. Another difficult aspect of the lab was fitting 4 maps into 1 letter-size layout. I found that creating guides in the layout helped me with my neatline and even distribution of the maps. I found that using the properties panel to individually set the size of each map frame into one uniform size across all map frames.

        

                    Figure 1: Percent of population over 65 in Miami-Dade County using equal intervals,                            natural breaks, quantile, and standard deviation methods.

                        Figure 2: Percentage of population over 65 normalized by square miles using equal                                intervals, natural breaks, quantile, and standard deviation. 

        Looking at the two maps, I would personally choose the equal interval maps from Figure 1 if I wanted to quickly understand what area in Miami-Dade County has the highest percentage of senior citizens. The legend is easy to understand for this map compared to unlike standard deviation.