### Introduction

The Binomial Distribution describes the probability of having exactly k successes in n independent Yes/No (success/failure) experiment with probability of success p. The Yes/No (success/failure) experiment is also called a Bernoulli experiment or Bernoulli trial.

Binomial probability can be calculated as

`Binomial probability = number of scenarios × P(single scenario)`

The probability of single scenario, P(single scenario), is simply

```P(single scenario) = pk(1 - p)(n - k)

pk means the probability of success to the power of number of successes.
(1 - p)(n - k) means the probability of failure to the power of number of failures. ```

If there were many more, say we are looking for how many scenarios for 4 success in 100 trials, above method would be very tedious, and very error prone. Therefore, we usually use an alternative approach, namely the choose function which is useful for calculating the number of ways to choose k successes in n trials.

To evaluate this choose function, we divide n factorial by k factorial times n-k factorial.

#### Choose function - examples

How many scenarios yield 1 success in 4 trials

```n = 4, k = 1
4!          4 × 3 × 2 × 1
(4 1) = ------------- = ---------------- = 4
1! × (4-1)!     1 × 3 × 2 × 1```

In R, the associate function is also called choose and it takes two arguments n and k. So:

``````> choose(4,1)
[1] 4``````

How many scenarios yield 2 success in 9 trials

```n = 9, k = 2
9!          9 × 8 × 7!
(9 2) = ------------- = ------------- = 36
2! × (9-2)!     2 × 1 × 7! ```

In R, the associate function is also called choose and it takes two arguments n and k. So:

``````> choose(9,2)
[1] 36``````

### Definition of Binomial Distribution

Put all above together, if p represents probability of success, (1-p) represents probability of failure, n represents number of independent trials, and k represents number of success:

where:

#### Binomial Conditions

What does it take for a random variable to follow a binomial distribution? Here are the 4 conditions:

1. The trials must be independent.
2. The number of trials, n , must be fixed.
3. Each trial outcome must be classified as a success or a failure.
4. The probability of success, p, must be the same for each trial.

### Examples 1

According to a 2013 Gallup poll, worldwide only 13% of employees are engaged at work (psychologically committed to their jobs and likely to be making positive contributions to their organizations). We are interested in finding the probability that among a random sample of 10 employees, what is the probability that 8 of them are engaged at work?

We can find this probability answer using the binomial distribution, because it actually meet the conditions required for the binomial distribution.

#### Solution 1 - Calculate by hand

```n = 10
p = 0.13
1 - p = 0.87
k = 8

Therefore,
P(k = 8) = (10  8) × 0.138 × 0.872

10!
= ----------  × 0.138 × 0.872
8! × 2!

= 0.00000278```

#### Solution 2 - Use R `dbinom()`

We can also calculate the same probability using R, `dbinom()` and takes 3 arguments. The 1st argument is the number of successes, the 2nd argument is sample size or the number of trials, and the 3rd argument is the probability of success.

``````> dbinom(8, size = 10, p = 0.13)
[1] 2.77842e-06``````

#### Solution 3 - Use web app

The app link is available from https://gallery.shinyapps.io/dist_calc/

### Standard Deviation of Binomial Distribution

Among a random sample of 100 employees, how many would you expect to be engaged at work?

Worldwide only 13% of employees are engaged at work, so p = 0.13

The expected number of engaged employees is:

`n × p = 100 × 0.13 = 13`

More formally, the Expected Value (mean) of binomial distribution is:

`Expected Value (mean) of binomial distribution = μ = np`

However, this doesn't mean that in every random sample of 100 employees exactly 13 will be engaged at work. In some examples, this value will be fewer, and in others, more. So how much would we expect this value to vary?

As usual, we can quantify the variability around the mean using the standard deviation. And for Binomial Distribution, the standard deviation is defined as the square root of np(1-p):

So, plug in the values from the survey, we would expect:

`sqrt(100 × 0.13 × 0.87) = 3.36. `

This means that 13 out of 100 employees are expected to be at engaged at work, give or take approximately 3.36.

• N/A