Each screen in Minitab has a help button you can click for additional help. This document is intended to get you to the right place in Minitab, not be a comprehensive guide of how to use it.

Used to summarize numerical data, find the mean, standard deviation, variance, etc. The *by variables* is an optional variable used to group the data. Use the *statistics* button to change what values are displayed. In particular, we will often turn on the variance.

Almost all of the graphs except for the graphical summary are found here.

- Histogram is the most common. There is a
*with fit*option to see if the 68-95-99.7 rule applies. - Dot plots, stem-and-leaf plots, and box plots are other options

- Pie charts are the most common graph for categorical data. The
*category*variable represents the labels on the slices of the pie. - A bar chart is another option for categorical data when the values don't make 100% of the choices or you can select more than one option.

This is a graph with multiple plots that show a lot of the information about a numerical variable. It includes a numerical summary, histogram, confidence interval, and test for normality.

Used to transform the data or generate new variables.

Create a list of numbers without having to type them all in.

Create a sequence using text. Separate your words with spaces. If there are spaces within your phrases, enclose them in quotes (Alien "American Pie" Jaws).

Useful for simulation purposes and sampling.

Find binomial probabilities. The cumulative probability is the area to the left of the value and the probability is the chance of getting that value.

Used to graph a distribution.

Find normal probabilities. This is easier done with the online probability calculator, but Minitab does allow you to specify a different mean and standard deviation.

Almost everything we do in this unit is a choice under basic statistics.

- 1 Sample t for confidence intervals for μ and hypothesis tests of form μ = 20.
- 2 Sample t for hypothesis tests of form μ
_{1}= μ_{2} - Paired t for hypothesis tests of form μ
_{d}= 0

- 1 proportion for confidence intervals for p and hypothesis tests of form p = 0.23. Be sure to check "use normal approximation" under options.
- 2 proportions for hypothesis tests of form p
_{1}= p_{2}. Be sure to check "use pooled proportion" under options.

Used to graph a distribution. Choose view probability when creating hypothesis testing graphs.

Don't confuse this with the probability distribution plot. This is used to see whether a group of data has a particular distribution. We are usually checking for normality. The data has the distribution you're testing for if the points basically fall along the line with no systematic patterns or outliers.

Use this to perform the chi-square goodness of fit test.

Use this to perform a test for independence. Click the *chi-square* button and select the chi-square analysis.

Use this to perform a one-way ANOVA test. There should be two columns, one for the factor (grouping) variable and one for the response (numerical) variable. If you created a variable for each group, then use the unstacked test.

Use this to perform a two-way ANOVA test. There should be three columns, two grouping variables for the row and column factors and one for the response (numerical) variable.

Used to graph two paired numerical variables.

Used to perform correlation between two or more variables.

Used to perform both simple and multiple linear regression.