The binomial distribution is one of the easier distributions to handle in terms of structure and calculation. Here’s what you need to know in order to choose a binomial approach in an experiment:
Rule
A random variable is represented by the binomial distribution if all of these points are fulfilled:
Rule
The probability of getting exactly successes out of trials is
where is the total number of trials, is the number of successes you want and is the probability of success in each trial.
Example 1
You roll four dice and look for the number of sixes. The random variable is then
When you roll a die, you either get a , or you don’t. Call each a success. The experiment then follows the binomial distribution with
Now you can calculate the probability of getting either , , , or sixes.
You do that by using the formula above. Then you can put the numbers you get into a table:
0 | 1 | 2 | 3 | 4 | |
This is the probability distribution of .
Note! The sum of the probabilities in a probability distribution is always .
Example 2
A bus company increases how often it inspects passengers’ tickets. They assume that 1 out of 5 passengers travels without a ticket. If they inspect the tickets of 20 random passengers, what’s the probability that
1 in 5 ride without tickets, which is what you’re being asked to examine. That means that
That makes this a binomial distribution with and .
The probability that exactly one of the doesn’t have a ticket is around %.
The probability that exactly five of the 20 ride without a ticket is around %.
To calculate , you have to add up and , making the first order of business to find them using the same formula you used in Items 2 and 3.
Example 3
Stephen King participates in a quiz with 50 questions where every question has four alternatives. Unfortunately, Stephen has forgotten to prepare at all, forcing him to guess on every question.
The probability of guessing the correct answer to a question with four alternatives is .
becomes
Sadly, Stephen is probably never going to Paris.
The sum of these probabilities is
If you insert this into the expression above, you get that
It seems like Stephen at least gets to go to the Museum of Natural History for free.
If the data set of a binomial distribution becomes very large, the set follows a normal distribution rather than the binomial distribution. In that case you can use these formulas:
Rule
The binomial distribution has an expected value, variance, and standard deviation as follows:
If , can be approximated with the normal distribution with expected value and standard deviation .
A rule of thumb is that if and are both greater than 10, can be approximated with the normal distribution.