Normal Distribution
Known as the “bell curve,” it is symmetrical distribution, with values gradually rising to the mean value, which is the highest value as well, and gradually falling after it.
Positive Distribution
The distribution where data are skewed to the left (to the smaller value): its maximum is smaller than it would be in the normal distribution. Its values grow drastically and then slowly decrease.
The examples are situations with fixed lower boundaries; their values tend to rise sharply and then slowly decrease. If most of the orders are delivered within three days, but some of them can be delivered much later. It gives a negative distribution, with most orders delivered within three days, but with a few of them delivered much later than this value.
Negative Distribution
The opposite of the positive distribution: here, data are skewed to the right (to the bigger value), and its maximum is larger than it would be in the normal distribution. Its values increase slowly and then decrease quickly.
The examples are situations with upper boundaries fixed, contrary to fixed lower boundaries of the positive distribution. One of them is the case of the company that guarantees that all its orders will be delivered no later than one week after the order. The orders are delivered gradually in one week, but most of them will be delivered near the end of that term, giving a negative distribution.
Bimodal Distribution
The distributions of that type have two peaks: the value grows, then falls to some degree, grows again, and falls ultimately.
An example is the level of some substance in a human body released in the morning and the evening: the concentration of the substance in the blood will give a bimodal distribution with two peaks, corresponding to the morning and the evening increases.
Inverted U-shaped Distribution
It is the symmetric bimodal distribution that has two peaks on the extremes; its mean value is equal to its median and is the smallest value in the distribution.
An example can be the app’s rating having very polar opinions among people. If most rates are 5/5 and 1/5, then the rates’ distribution will get inverted U-shaped distribution, with peaks at 1/5 and 5/5 values.
