Experimentation for Improvement

We are constantly using experiments to tweak and find improvements in our personal lives, our communities, and in our work. But are you doing it efficiently? Or are you changing one thing at a time and hoping for the best? In this course, you'll learn how to plan efficient experiments using statistical methods - enabling you to test for many variables that lead to better results.

About The Course

Would you like to:
  • improve the quality of drinking water;
  • make a stronger concrete or brick;
  • increase the sales from your store;
  • find the right combination of settings for your favourite recipe;
  • improve the quality of your company's product;
  • reduce waste;
  • minimize energy use?

No matter what your area of interest (and there are no limits to the applications!), it is clear: better experiments save time and money, and lead to improvement.

In this course we will learn to use efficient factorial experiments, fractional factorials and response surface methods. If these terms sound intimidating, don’t fear! We include a gentle introduction to basic statistical concepts, where required, throughout the course.

By the end of this 6-week course you will be able to design your own experimental program, changing multiple variables, and interpret the experimental data using simple tools, based on sound statistical principles. And you will quickly see how to make those improvements.

These tools and methods can be beneficial to solve the challenges you set for yourself above.

Frequently Asked Questions


Can I get a Verified Certificate after completing this class? 
Yes. Students who successfully complete the class requirements will receive a verified certificate. We are working on our certificate design, and will post that really soon.

Will I get a Statement of Accomplishment after completing this class?
Yes. Students who successfully complete the class will receive a Statement of Accomplishment signed by the instructor.

What are the social media channels for this course?
Hashtag your Twitter posts with #efi2014 and tweet us @2014efi 

What resources will I need for this class?
For this course, all you need is an Internet connection, copies of the text and slides (all of which are for free), and the time to read, write, and complete the assignments and project. All the materials for this course are freely available under a Create Commons license, so you are free to share them with friends and colleagues.

What is the coolest thing I will learn if I take this class?
You'll learn how to approach an optimum in any process in a few experiments, and be able to guarantee you are at the optimum. I will also show you how to make sure you remain at the optimum.

Anything else?
Every week I will give several tips and pieces of advice. These are things I have learned when doing data analysis and consulting with companies. Tips on how you should manage data, practical ideas on using statistical software, and so on. I can't wait to share these.

Recommended Background

We recommend you have a basic understanding of least squares models, but we will also have extra (optional) review videos, if you have not used these models before. We will only use this background from week 3 in the course, giving you some time to evaluate the course. In fact, this entire course can be done without knowing about least squares models, but these models do make concepts and explanations a little easier.

There are 4 different ways you might review this topic before the course starts:

  1. Coursera's Statistics 1: check out lecture 7 and 11, or
  2. Lectures from week 7 of Coursera's Making Sense of Data, or
  3. Sign up for Coursera's Regression models course, or
  4. Read chapter 4 of the free textbook Process Improvement using Data