Sharpening Your Sword

These lessons are designed for you to relearn what you have forgotten, reinforce what you have just learned or learn a new tool or technique.

All Lessons are:

  • Accessable 24hrs a day for length of subscription
  • Narrated lessons are  developed and presented by Peter Bersbach and are based on “The Six Sigma Handbook” by Thomas Pyzdek
  • Email address available to ask questions of Peter Bersbach
  • Microsoft Excel w/ Free Data Analysis Add-in used in lessons designated

We do use, extensively, Microsoft Excels and it’s Data Analysis add-in. This add-in is free to all Excel owners. Other spreadsheet software may have the same capabilities.

  • Statistical Software: Minitab is used in some lessons

Although there are alot of  statistical software packages out there, we use Minitab where statistical software is needed. You can purchase Minitab or SigmaXL® . You also can get a free 30 day trial copy of Minitab to get started.


Six Sigma Overview
Video Length: 33 minutes

This introductory seminar provides an overview of Six Sigma and is designed for CEO’s, owners, and managers to learn how to think in new ways about possible breakthrough improvements in their organization or business. Using the Six Sigma process, the seminar will demonstrate results from several small and mid-size businesses that have had great success by implementing this critical top-down, buy-in process. This seminar will detail what Six Sigma is and is not. Upon completion of this seminar, you will be able to make an informed decision as to whether or not you want to incorporate the Six Sigma process into your business.

Variation and Measurement
Video Length: 22 minutes

You will learn about the how we measure variation using different measurement scales, and data types. We will define, show examples and inform you of the correct statistic to use for each of the four measurement scales (Nominal, Ordinal, Interval and Ratio)  We will also discuss the two data types (Attribute and Continuous) and give you examples of each. Last we explain the important of knowing, before you start collecting data, what type a data you will be collecting and on what scale as that will define the statistical tool that you will use to “picture’ or summarize what you collect.

Descriptive Statistics
Video Length: 25 Minutes

Data are just a list of numbers, in this lesson we will take that list and turn it into “Statistical Pictures”  called Descriptive Statistics so we can understand more about the process the data is from. In this lesson we will learn about the descriptive statistics called mean, median, mode, range, standard deviation, skewness, Kurtosis. During this lesson besides learning about these descriptive statistics you will get a chance to use them yourself.

Prerequisite: The information in Variation and Measurement.

Decision Theory
Video length: 17 Minutes

This lesson on Decision Theory talks about how we make a decision and how that applies to Statistics.  This is important to understand because when we look at process and make a hypothesis , or guess ,  of what it is doing, we will eventually have to decided whether our hypothesis,  or guess,  is correct or not . So in this lesson we’re going to  talk about decision theory, point and interval estimates, error types, confidence intervals, and mean estimating .

Prerequisite: The information in Descriptive Statistics.

Cause and Effect Diagrams
Video Length: 5 Minutes

This lesson on Cause and Effect Diagramming talks you through all the step in building a cause and effect diagram starting with the definition of the issue to creating the diagram itself.  I use Visio to create most of mine but Powerpoint will work just as well.


t Test (also know as the Student t or t statistic)

This hypotheses test is used to compare two means. You can use it to compare a small sample mean to a known target or population mean. You can also use it to compare the means of two small samples.  The data for this test is assumed to be normal and one should check this before using this test. Below is the equation.

There are several comparisons that one might make and each is handled a little differently when using software to run the t test.


  • Comparing two “paired” samples. Pairing happens when you have two sets of data that each measurement in one set is matched to another. This could happen when you take a measurement before and after some type of treatment on the same object. Pairing reduces the variance or variation and can detect smaller shifts. Examples of Pairing are: Weights of individuals before and after a exercise program, Pre and Post test scores of students taking a class, or measuring the same product with two different devices.

Just a note; as the sample size gets larger, the t distribution becomes the normal distribution. If you have large samples (something collected electronically) you might want to use the z statistic instead.