A discrete data distribution describes the probability of occurrence for discrete events.

For example, a discrete distribution describes the probability of households in California having 0, 1, 2, 3, or 4 cars. The Department of Motor Vehicles may use this information to determine the probability of 2 or fewer cars per household. The discrete distribution is commonly used in service quality settings, such as customer call centers, hospitals, and financial institutions.

A discrete distribution is a list of the different numerical values of the variable of interest and their associated probabilities. The probabilities must sum to 1. For the car example, the probabilities of specific numbers of cars are listed below.

Number of cars per household | Probability |
---|---|

0 | 0.03 |

1 | 0.13 |

2 | 0.70 |

3 | 0.10 |

4 | 0.04 |

In Minitab, the values in a discrete distribution must be numeric. However, you can use numeric values to represent categorical variables. For example, a credit card company wants to understand the types of customer calls. You could use Minitab to create this distribution by coding the call types: 1 = annual percentage rate is too high, 2 = question about incentives (airline miles), and so on.

Use the discrete distribution for discrete data when none of the other discrete distributions (Bernoulli, binomial, hypergeometric, integer, and Poisson) are appropriate.