5.1 Common Methods of Data Classification
equal intervals (equal steps)- each class occupies an equal interval along a number line
determine class interval (class width)
determine upper limit
determine lower limit
specify class limits (in your legend)
determine which data values fall into each class
quantiles- data are rank-ordered and equal numbers of observations are placed in each class
mean-standard deviation- data classification to consider how the data is distributed
maximum breaks- raw data is ordered from high to low, the differences in adjacent values are computed, the largest breaks serve as the class breaks
natural breaks- consider the natural grouping of the data. (via picture: histogram, dispertion graph)
optimal- is done by placing similar data values in the same class to minimize classification error. First quantiles are used to classify the data, then placing similar values in other classes if it is appropriate.
n-class- there are no classes of the data, the scale is a gradient. People cant pin-point data values.