Minitab Cookbook (2014)

Preface

Minitab^® Statistical Software is a program with a long history. Its beginnings were at Penn State University in 1972, where three professors, namely, Barbara F. Ryan, Thomas A. Ryan, and Brian L. Joiner set about creating a statistics package to help their students learn and use statistics easily. This emphasis on trying to make statistics more accessible to everyone has continued through every iteration. Barbara Ryan still owns Minitab, the company that continues to create new versions of Minitab to make the use of statistics easier for everyone.

Over the years, Minitab has grown, each version adding new features and functionalities. Along with more advanced techniques that are added, there are also new easy-to-use features. In Minitab 13, the StatGuide™ was added to give quick references to the terminologies. In Version 16, the Assistant was added to help guide users to the right graph or statistical tool, continuing the trend of making statistics accessible.

After I obtained my Masters in Physics, I started working at Minitab. Most of my work has been concentrated on teaching how to use the software and how to understand the results, or when to use which statistical tool. The move from physics to statistics was made very easy by using Minitab. Its pedigree in being a teaching tool shows throughout, and it is still a powerful tool that is being used in many sectors of industry or business.

Part of the success of Minitab can be put down to the world's growing realization that understanding data and using data-driven decisions has become essential to success. This is epitomized with different business improvement programs such as Six Sigma and Lean Six Sigma. No matter which name is used to describe the improvement plan, the days of saying, "It looks like that made it better", or "If we do this it should work" are over. Increasingly, the questions are "Can we prove what should be changed?", or "Have we successfully improved the process". Minitab provides the tools that can be used to understand those variations and prove these differences if they exist.

In this book, I have attempted to try and find as much real data as possible to illustrate the use of each tool. This meant many nights of searching for different datasets and different data stores. Some data has just appeared at the right time, a serendipitous question on how to run some test or the other; others I have found from open source locations. Websites that keep a track of public data for use as examples, such as the Data and Story Library (DASL), have been invaluable sources. Quandl, for instance, is a website that holds a massive amount of data for financial, economic, and social information.

In a few places, it was not possible to provide real data. Of those datasets, most are based on real examples that are carefully recreated to hide the real study or to tidy up the example.

I wanted to show how varied the use of both Minitab as a tool and statistics can be. With this in mind, data has been picked from a wide variety of topics. This also provides another benefit for us. One problem new users of Minitab can face is how to insert the data correctly. What format should we use to enter our results? The worksheet does bear a similarity to an Excel spreadsheet, but anyone trying to use the worksheet like Excel will end up in a mess. The key is to enter data in columns. In each chapter, there are a few examples that show the formatting, right from getting this data into Minitab and into the right layout for use with that tool.

I hope you find this book useful. We want you to be able to pick a recipe and jump to that page and follow the example of interest to you.

What this book covers

Chapter 1, Worksheet, Data Management, and the Calculator, shows how to manage your datasets. We look at getting data into Minitab and at formatting tools, such as transposing or stacking data.

Chapter 2, Tables and Graphs, covers examples of creating graphs, and using some of the tabulation tools. The examples use bar charts, pareto charts, Tally, scatterplots, and more.

Chapter 3, Basic Statistical Tools, looks at the statistics in the basic statistics menu. We cover the use of the hypothesis test tools and look at chi-square tables.

Chapter 4, Using Analysis of Variance, covers the use of ANOVA from a simple one-way ANOVA, to general linear models, and to mixed effect models.

Chapter 5, Regression and Modeling the Relationship between X and Y, looks at how to use the regression tools. This covers the basic fitted-line plots before going into the more complex general regression tools using several predictors, model reduction tools, and binary logistic regression.

Chapter 6, Understanding Process Variation with Control Charts, shows how control charts are used to monitor the stability of a process. Here, we look at the use of the familiar Xbar-R, I-MR charts, and also go on to look at the more complex Laney control charts and rare event charts.

Chapter 7, Capability, Process Variation, and Specifications, looks at the tools used to assess a process to its specifications. We cover the use of normal and nonnormal data along with acceptance plans.

Chapter 8, Measurement Systems Analysis, covers the tools used to assess the quality of the measurement system. We look at the Gage R&R tools, including the expanded Gage R&R and attribute measurement studies.

Chapter 9, Multivariate Statistics, looks at the use of principal component analysis and factor analysis for reducing the number of variables or understanding associations in the data. Also, it covers cluster analysis tools, correspondence analysis, and discriminant analysis.

Chapter 10, Time Series Analysis, covers tools to fit to trends, seasonality, and then looks at what to use when no trends or seasonalities exist in the data.

Chapter 11, Macro Writing, looks at how to create simple macros and execs before it looks at the more complicated local macros.

Appendix, Navigating Minitab and Useful Shortcuts, lists navigating tools and useful shortcuts to be used in Minitab.

What you need for this book

Ideally, if you are using this book, you have Minitab available with you. This was written with users of Minitab 17 in mind. Datasets that are provided here can be opened in Minitab 16 and higher. The strategy for new versions of Minitab is to be least disruptive in user experience as possible. Anyone using earlier versions should find that a lot of the commands still run true; however, they may find certain tools here that are not available in previous versions of Minitab.

For instance, the Assistant appears in Minitab 16 and higher. Also, with Minitab 17, there have been big changes to the linear model tools.

Also, note that Laney control charts appeared in Version 16.2. For anyone using Minitab 16.1, you can update your version to the latest 16 version by using Check for Updates under Help or by talking to your IT department.

Who this book is for

The focus of this book is instructions on how to use Minitab. We do not explain how to interpret the statistics nor do we dig deep into the statistical formulas.

While it is not expected that the reader has in-depth knowledge of all the areas of statistics covered in this book, you should have a basic understanding of the tools that we want to use.

This book is for anyone who wants to know how Minitab likes to have data set up and how we can get Minitab to run those functions to get the most from the software. It is for anyone who feels a bit lost while looking at the worksheet or session folders and wonders what to do next.

If you find yourself asking, "How can I run a binary logistic regression to see the significant effects and present this output in a useful way?" or you run back to Excel to reformat a worksheet and then go back to Minitab, then we have the instructions for you here.

Minitab has become a very powerful and all-inclusive statistical package covering a lot of statistics. This is intended as a guide to help us find our way to the right menu and the right tool.

Conventions

In this book, you will find a number of styles of text that distinguish between different kinds of information. Here are some examples of these styles, and an explanation of their meaning.

Code words in text, database table names, folder names, filenames, file extensions, pathnames, dummy URLs, user input, and Twitter handles are shown as follows: "We can include other contexts through the use of the include directive."

When we are pointing to a URL then this is indicated as follows:

http://lib.stat.cmu.edu/DASL/DataArchive.html

A block of code is set as follows:

TSPLOT Data;

Index;

Connect:

Symbol;

Any command-line input or output is written as follows:

%Glayout

New terms and important words are shown in bold. Words that you see on the screen, in menus, or dialog boxes for example, appear in the text like this: " Any worksheet is suitable, but be aware that the Open Worksheet… command".

Files to open use the code format as follows:

Oxford Weather.txt

Columns in the worksheet are referred to as follows:

Year

Note

Warnings or important notes appear in a box like this.

Tip

Tips and tricks appear like this.