Dispersion or variation can be expressed either in terms of the original units of a series or as an abstract figure like a ratio or percentage.
Absolute measures of dispersion measure the extent of dispersion of the item values from a measure of central tendency. They are expressed in terms of the original units of the series. These measures only give an idea about dispersion with reference to a given measure of central tendency. So they are not suitable for comparing variability among distributions.
Relative measures of dispersion are known as ‘Coefficient of dispersion’. They are obtained as ratios or percentages. They are pure numbers independent of the units of measurement. Variability or dispersion among different distributions are compared by these relative measures.
One example will illustrate the point. Supposing the average mark of a group of students is 20 and the absolute measure of dispersion is 10 and the average mark of another group of students is 60 and the absolute measure of dispersion is also 10. This does not mean that variability in the two series are the same. For this, relative measure of dispersion should be calculated.