Hypothesis Testing

Checking for differences

Why use it?

Hypothesis testing is used to help determine if the variation between groups of data is due to true differences between the groups or is the result of common-cause variation, which is the natural variation in a process. This tool is most commonly used in the Analyze step of the DMAIC method to determine if different levels of a discrete process setting (x) result in significant differences in the output (y). An example would be “Do different regions of the country have different defect levels?” This tool is also used in the Improve step of the DMAIC method to prove a statistically significant difference in “before” and “after” data.

What does it do?

How do I do it?

  1. Collect and plot the data
A Hypothesis-Testing Data Plot
  1. Select the appropriate test
Which Hypothesis Test to Use?
  1. Analyze the data
Hypothesis tests often require you to gather a lot of data to observe a significant difference. Work with your Six Sigma Expert or Master Six Sigma Expert to determine the power of your test and to detect the size difference you are looking for. You might need to collect additional data to see the difference you desire.
Hypothesis tests are so named because they start with what is called a null hypothesis and set out to prove or disprove it. The null hypothesis states that there is no difference between the groups. This null hypothesis is assumed to be true until it is disproven with data. If the result of the test proves to be significant (p < .05), the null hypothesis is declared to be untrue.
When performing Analysis of Variance (ANOVA) testing, you should also perform an additional hypothesis test for equal variance in the subgroups. The ANOVA assumes equal variances. Your Six Sigma Expert or Master Six Sigma Expert can help you draw conclusions if this assumption is not met.


Many statistical procedures have built-in hypothesis tests. For example, in regression analysis (see page 214 for details), p-values are given on a slope. These values come from a test of the null hypothesis that the slope is zero (i.e., there is no difference in slope).

For more information on Hypothesis Testing see The Black Belt Memory Jogger®.

20. Interrelationship Digraph