Statistical methods

You can use the following statistical methods for cell, neighborhood, and zonal functions:

• Majority: determines the value that occurs most often.
• Minority: determines the value that occurs least often.
• Maximum: determines the highest value.
• Minimum: determines the lowest value.
• Range: computes the difference between the highest and lowest values.
• Mean: computes the total of the values divided by the number of observations.
• Median: computes the central value, where there are an equal number of values above and below it.
• Standard Deviation: computes the spread or dispersion of values from the mean.
• Sum: computes a total value by adding all of the values.
• Variety: determines the number of different values.

To be evaluated, the cell values must be numbers, either a measurement or a code representing a feature or the ranking of a feature.

If the values are measurements, like temperature, all of the statistical methods described above are valid. For example, you can find the mean temperature from a range of different temperatures. However, if the values are codes, such as unique numbers representing features, some of the statistical methods may not be useful. If you had, for example, a raster dataset with values ranging from 1 to 20 and each value was code indicating a specific vegetation type, then finding the sum of those values would not make sense. On the other hand, you could use variety to find out how many types of vegetation occur.