O'Reilly logo
  • Ankur Sawhney thinks this is interesting:

In most cases, the standard deviation, the average absolute deviation, and the interquartile range agree in their rankings of the amount of variation in different data sets. Exceptions can occur with outliers because the squaring of deviations can have a large effect on the standard deviation. Just as the median is more resistant to outliers than the mean, so the average absolute deviation and the interquartile range are more resistant to outliers than the standard deviation. Nonetheless, because of its mathematical tractability and importance in probability and statistics, the standard deviation is usually used to gauge the variation in a set of data.


Cover of Essential Statistics, Regression, and Econometrics


Average absolute deviation gives an idea of the 'amount of variation' in the dataset