Binomial Distribution

If you click Statistics and then Binomial Distribution, you get a screen like this.

The Binomial Distribution calculates the chance when a simple thing (flipping a coin or rolling one dice)

is done repeatedly and you are interested in how many times some thing happens.

The thing might be Heads on the coin, or three on the dice.

 

n is the number of times you flip the coin (or whatever),

k is the number of times you want and

p is the chance the thing happens if you only did it once.

Here I flipped a coin (so p, the chance of Heads, is  0.5). I flipped it 10 times, so n is 10. I wanted 4 heads, so k is 4.

 

I clicked     and I got the two answers and the graph.

The Probability  is 20.5078%,  the chance of 4 heads in 10 flips.

The Cumulative Probability is the total chance of getting 4 heads or 3 heads or 2 heads or 1 head or no heads; that is, I have a 37.6953% chance of getting 4 heads or less.

The graph shows bars, one for each possible value of k. The diagonally shaded bar is for  k = 4; the solid bars are for values less than 4. The chance of 0 heads is so small that it looks like there is no bar there, but there really is.

 

Binomial Probability is calculated using the formula  

 

CalGraph will calculate probability for  n up to 99.

For larger values of n, use the Normal Approximation to the Binomial Distribution.

 

CalGraph will calculate the other way too: given a probability, CalGraph can tell you what k cuts off that amount. Click Critical Values at the top of the screen, and then Alpha. You get this window:

 

Here I put in n = 40 and p = 0.5  I put the chance, the Alpha, equal to 0.05

Then I clicked One-tail lower and then Calculate.

You see I got  k = 14 as the Critical Value.

This means the Cumulative Probability for k = 14 is about 0.05

In other words, the chance k is 14 or less is about 0.05.

Of course this is not exact. If I calculate the Cumulative Probability, I get 0.040345.

  

If I click One-tail upper, I get 26, that is, the chance of k is 26 or more is about 0.05

Two-tailed gives k = 13 and 27, with the chance k is 13 or less or 27 or more being very roughly 0.05