The mean deviation of a set of measurements is the average distance of each measurement from a central value. It can be calculated from the mean or median of the data. The standard deviation is a better estimator of the population standard deviation for normal distributions, but mean deviation is more accurate for certain types of data with outliers.