CLASSIFICATION METHODS IN ARCVIEW
Vicki Drake
Santa MonicaCollege
ArcView’s five classification methods are Natural Breaks, Quantiles, Equal Interval, Equal Area, and Standard Deviation. Each has its own strengths and weaknesses to show features in the data clearly for easy understanding.
Natural Breaks Classification
This is the default classification method in ArcView, using an algorithm, Jenks Optimization, to group values within a class, resulting in classes of similar values separated by breakpoints. This method works well with data that is not evenly distributed and not heavily skewed toward one end of the distribution.
Although Natural Breaks classification can be used to discover patterns in the data, it can frequently result in legends that are difficult to understand and strange number ranges can appear in the legend. It is possible to “tweak” the number ranges (using Legend Editor table), however, too much “tweaking” may result in an uneven mess. If there are peculiar number ranges, changing the Labels in the Legend Editor from numeric to text labels is a better option. The number ranges can be labeled “high, medium, or low” – categories more easily comprehended by the average map reader.
Quantile Classification
In a Quantile Classification, equal numbers of features (and the corresponding records in the attribute table) are placed in each class. For example, if there are 20 features in a theme, the ArcView (using Quantile classification) will place four features into each of five classes. This is a good method of classification for evenly distributed data.
Default Quantile Classifications can result in very misleading legends, however. In order to have an equal number of features in each class, features will greatly different values can be placed in a single class. One solution is to manually increase the number of classes – this may separate dissimilar values.
Equal Interval Classification
The Equal Interval Classification divides the total range of feature values, from maximum to minimum, into five equal subranges. This creates an easy to understand legend and works best with continuously distributed data, such as precipitation amounts across North America. The downside of this method is that if it is applied to certain data types, the legend can end up with classes with few, if any map features contained within them.
Equal Area Classification
The Equal Area Classification is a very specialized ArcView method that divides the range of features in a polygon theme so that each subrange contains basically the same area. This is a useful method for only a limited number of situations, such as apportioning sampling areas so that every observer gets an equal area.
Standard Deviation Classification
The Standard Deviation Classification method calculates the mean value of a field in a theme, and then divides the range of feature value by several deviations above and below the mean. This method works well with normally distributed data; that is, data that fits a standard bell curve, with most feature values falling in the center, with a few high and low values appearing above and below.
© Vicki Drake
Santa MonicaCollege
GIS lectures