Data Classification

In this week's lab we learned about different types of classification methods. We learned about several different methods and took a deeper focus into four classification methods, such as Equal Interval, Quantile, Standard Deviation, and Natural Breaks. One of our tasks was to review different types of data classifications and then review and prepare data that was provided to us of the Miami Dade census. Then we were tasked with creating two maps presenting the data using four different classifications. One map was to be done in percentage without the use of normalizing and the other was done in amount with normalizing graduated color symbology. Once the maps were completed, we then were tasked with taking a closer look and analyzing what we thought best displayed the data for our audience being the Miami Dade County Commissioners. Here is a short description of the four classification methods that were used. 

 Equal Interval                                                                                                Equal Interval classification divides data into equal size classes. Which works best when you have more continuous data spread across your area of reference. This method is more cut and dry, with just dividing classes into equal groups. Equal Interval can result in classes that do not have data or classes that have significantly more data than others. This might lead to not representing data well.

 

 Quantile

Quantile classification places an equal number of observations in each class. The quantile classification counts the quantity in each group and then arranges them as close to the average as possible. The Quantile classification can never have an empty class or a low number value class. This results in the method having classes with a large difference of value in the same class or different classes with similar values.

The Quantile map can look misleading if you are not paying attention to the Legend. It may not reveal subtle patterns.

 

 Standard Deviation

Standard Deviation classification finds the mean of the data and then there are positive and negative deviates from that mean and then each standard deviation becomes a class. This kind of method needs a good amount of clarification for the reader on what is being shown. This classification is one that should be normalized.

 Natural Break

Natural Break classification finds class breaks that will minimize variance by grouping similar values together and maximize between class differences based on the data. These kinds of breaks can cause data clustering of a large number of values into a limited number of categories. It’s a way of making sense of numerical data by identifying where natural divisions or clusters occur. This method is good for finding underlying patterns in the data                                     

The map that I chose to share here with you is the Percentage of the population above the age of 65 in Miami Dade County. I felt that the Percentage map without the normalizing did a better job displaying the data that I wanted to be displayed. I felt that the Natural Breaks did the best presentation of the data out of the four classification you see displayed on the map. When looking at the map I feel that the Natural Break does a better job at distribution of the classes. I feel that the map is visually easier for the reader to understand. I also feel that it more accurately displays the data. therefor less of a chance of misleading the reader.