Data A: {4, 4, 4, 4, 50}; Data B: {4, 8, 15, 24, 39, 50} Range = 46
Same range value but they have very different distributions.
Also to avoid extreme observations, inter-quartile range is useful as a measure of dispersion. A larger IQR value indicates data are well spread.
Inter-quartile range is the range of the middle most 50% of the observations (distribution)
Inter-quartile Range = Q3 – Q1 = P75, - P25.
The lowest value to be (minimum non-outlier) = Q1 – 1.5IQR
The largest value to be (maximum non-outlier) = Q3 + 1.5IQR
Values which are smaller than the lowest value or the largest value to be in the boxplot are know as extreme values (outlier).
