How to Perform a Normality Test on Minitab?

It’s crucial to determine if the provided data follows a normal distribution before doing any statistical study on it. You can utilize parametric tests (test of means) for further statistical analysis if the provided data has a normal distribution. You would therefore need to apply non-parametric tests if the presented data did not have a normal distribution (test of medians). Parametric tests are more effective than non-parametric tests, as is common knowledge. Checking the normalcy of the provided data becomes much more crucial as a result.

1. Make the hypothesis.

Writing the hypothesis at the outset of each statistical investigation is a wise strategy. “Data follows a normal distribution” is the null hypothesis for the normality test, in contrast to the alternative viewpoint.

2. Decide on the data.

The spreadsheet’s data should be copied if you want to perform the normality test on it.

3. The data is then copied and pasted into the Minitab spreadsheet.

Copy and paste the data into the Minitab spreadsheet after launching Minitab.

4. Press on “Stat.”

The data entry cells for the Minitab spreadsheet are displayed. Then click “stat” in the menu bar and choose a Normality test submenu in Basic Statistics.

5. Choose data.

On the screen, a little window with the title “Normality Test” will show up. Choose something useful by clicking inside the white box, then click “Select.” 

Be careful that the name of the chosen data will appear in the “Variable” tab. 

Also, keep in mind that the box next to “Tests for Normalcy” is already ticked for “Anderson Darling.” The Anderson-Darling normality test is the most popular one. As a result, “AndersonDarling” is the preselected option for Tests for Normality in Minitab.

6. Select “Ok.”

7. Understand the p-value displayed in the Normal Probability Plot.

• On the screen, a normal probability plot will appear. 
• If the p-value shown in the normal probability plot is greater than or less than 0.05, please verify this.

8. Infer the outcomes.

The conclusion will be “Data doesn’t follow a normal distribution” if the assumption is incorrect. Then link the p-value to the stated thesis. You may anticipate a decent approximation to the normal distribution given the dependence on the results of the two normality tests.

9. If the p-value is larger than 0.05, do not reject the null hypothesis.

If the p-value observed in the normal probability plot is more than 0.05, the null hypothesis is not excluded. Thus, the statement “Data follows a normal distribution” is drawn.

10. If the p-value is less than 0.05, the null hypothesis should be rejected.

We reject the null hypothesis if the observed p-value in the normal probability plot is less than 0.05. “Data does not follow a normal distribution,” is the conclusion that follows.

Example:

The scientist at a protein shake firm tries to figure out how much protein is in the protein powder. The quoted percentage is 60.25 percent. The scientist determines the rate of protein in 35 randomly selected samples. The scientist wants to be sure the assumption of normality is true before doing a hypothesis test.

Conclusion

The normality test evaluates whether data exhibit characteristics of a normal distribution. An asymmetrical bell-shaped arc around its mean characterizes a normal distribution. The normal distribution is the most popular statistical distribution because it naturally occurs in many physical, biological, and social measurement situations.

Generally speaking, the best methods for assessing normally were the normality test and probability plot. The normality test is actually a covert hypothesis test, though. The data are not unusual, according to the null hypothesis (Ho). Additionally, our alternate hypothesis (HA) is that the data is unusual.