The type difference returns to the square root of the variance.
It allows you to clarify the proportion above or below a given value between 2 values.
A weak type difference characterises a homogeneous distribution. A non-existent type difference indicates that the values are identical.