Skewness measures the asymmetry of the probability distribution compared with the normal distribution. In other words, compared to a normal distribution (bell curve), is the data skewed to one side or the other in terms of having a longer tail to the right or left? While we must be careful when interpreting skewness, a positive skew generally means that the tail is on the right, while negative skew indicates that the tail is on the left. This is compared to a normal distribution with two equal tails.
This recipe demonstrates how to calculate Karl Pearson's coefficient of skewness (the second method), which is given by the following formula:
In plain English, to find the skewness you subtract the median (the middle value) from the mean (average), multiply by three, and then divide by the standard deviation.
...