Difference between median and mean

On the right is Amazon's quarterly revenue from their first year - 1996 - and a more recent year - 2017. Let's compare the mean and the median revenue between the two quarters to see how the data is distributed. The median is more robust to outliers, unlike the mean. So if your dataset has a lot of outliers, the mean will be pulled in the direction of the outliers.

Recall that for an odd number of observations, a median is the middle value in the data. With an even number of observations, the median is the mean of the two middle values. To manually calculate a median, the data needs to be sorted (unlike if you are using the MEDIAN() function). This data is naturally ascending, so no sorting is necessary.