There are four prominent measures of variability and they each represent a unique approach to estimating the spread of scores.
The Range is rarely used because it considers only the two extreme scores on either end. The range says nothing about the scores in the middle.
The Interquartile Range should be used if the data is of the ordinal type or if extreme scores skew the distribution in either or both directions.
The Variance is rarely used because by squaring the differences, the score achieved is not within the range of the original data.
The Standard Deviation is the most stable and is used with non-skewed distributions that are of the interval or ratio type. Since it relies on the mean, any distortion of the mean will also distort the standard deviation.
