There are multiple hypothesis tests available in Minitab and should be utilized as per data type and the objective.

Let’s take an example of one sample t-test.

**One sample t-test**

One sample t-test is used to compare population mean with a defined value using sample data.

The test uses sample standard deviation to calculate population standard deviation. If there is a large difference between the sample mean and the specified test mean (given value), then the test concludes that it is highly unlikely that population means will be anywhere near test mean.

**Assumptions**

1. Sample data follows normal distribution

2. Sample data is random

**Analysis in Minitab**

Let’s assume sample data of average handling time:

**To validate if the goal of reducing population mean to 4 minutes is significant, we can use one-sample t-test.**

H0: Hypothesized Mean (4 min) = Population Mean (µ)

H1: Hypothesized Mean > Population Mean (µ)

**Step 1: Check if data follows normal distribution to ensure we can use t-test.**

**To check do to Stat -> Basic Statistics -> Graphical Summary**

If p-value is greater than 0.05, that means data follows the normal distribution.

**Step 2: To conduct one sample t-test, go to Stat -> Basic Statistics -> 1-sample t**.

By default, the alternate hypothesis for 1-sample test in Minitab is ‘Hypothesized Mean ≠ Population Mean (µ),’ to change, go to ‘options’ while performing the test on Minitab as shown below.

**Minitab result interpretation:**

The following illustration shows the typical result of the t-test:

**Data Interpretation: As P is 0.000 i.e. less than significant level of 0.05, hence reject null hypothesis which means set target is significant.**

**Below shows the flow charts that can be used to select right hypothesis testing:**

**1. Continuous Normal Data (Parametric Test)**

**2. Continuous Non-Normal Data (Non-Parametric Test)**

**Hope this helps!**

**You ****can buy my book Lean Six Sigma: Process Improvement Using Minitab – Green Belt Certification on Kindle.**