A distribution is a way to represent the frequency of outcomes that occur in a sample. The shape of the distribution changes based on how many observations are within it. If we take our data set and create an empirical distribution, then as more observations are added to the data set, there will be less variation between each observation.
There will also be more observations in the area of the distribution closer to the mean. The shape of a distribution is determined by three things: how much variation there is within it, where most of its values are clustered, and which specific type that it falls under.
When looking at distributions as they increase in size (i.e., adding data points), we see two different shapes emerge-the “skewed” or fat tail distribution and a “normal” shaped distribution with no long tails on either side. A skewed or fat tailed distribution has outliers that live far from the average value; these outliers can cause steep drops off near one end of the curve when there are many observations being added because this