Not long ago, a reader called to that she was using the “Word of the Week” to help educate her team and orient them to statistics, but that some of the terms went over their heads and, for some, unfortunately reinforced an aversion to statistics. “Instead of the negative binomial, how about just starting with the plain-vanilla binomial?” she pleaded. So here goes.
A Binomial distribution is used to describe an experiment, event, or process for which the probability of success is the same for each trial and each trial has only two possible outcomes. If a coin is tossed n number of times, the probability of a certain number of heads being observed in n tosses of the coin is given by the Binomial distribution.
Yes/no (0/1) data are ubiquitous — purchase or don’t purchase, respond to treatment or stay ill, click on a link or don’t click, win or lose, etc. The binomial distribution is used in determining whether observed proportions differ significantly from what chance might produce.