Statistical Distributions

Introduction

Statistical calculations often appear mystical as though plucked from mid-air. Statistical calculations are in fact based on the mathematical pattern of events. These mathematical patterns are called distributions.

Example of statistical distributions are:

• Binominal
• Normal
• Poisson’s

There are many other statistical distributions; however, here let us here look at the binominal and Normal distributions.

Binominal distribution

When you flip a coin there are two possible outcomes; they are either heads or tails. So, when you fillip a coin and take the tally of all the ‘heads’ and ‘tails’ what you have is a binominal distribution.

Normal distribution

Normal distribution is also called Gaussian distribution.  Many natural phenomena follow a Normal distribution. For example, if you plot the height of all the boys in a class on the x-axis and the frequency on the y-axis, you will get a bell-shaped curve. The x-value of the peak of the bell-shaped curve corresponds to the mean (i.e. average) height. The mathematical pattern of the bell shape gives rise to many of the statistical calculations we often undertake in medical research.

The type of data determines whether normal distribution applies. Normal distribution applies to numeric data but not to categorical data.