In our example, the data points approximately follow a straight line that falls mainly between the 95% confidence interval limits, and so it can be concluded that the data is normally distributed. The normal distribution is a good fit if the data points approximately follow a straight line. Here is a screenshot of the example result for our previous product_weight example: To create a normal probability plot in Minitab, select Graph > Probability Plot > Single, specify the column of data to analyze, leave the distribution option to be normal, and then click OK. All points for a normal distribution should approximately form a straight line that falls between 95% confidence interval limits.
This provides a more decisive approach for deciding if a data set is normally distributed. We can also construct a Normal Probability Plot to test the assumption of normality. Notice how the data is symmetrically distributed and concentrated in the center of the histogram. The figure below shows a histogram which suggests that the data is normally distribution. To create a histogram in Minitab, select Graph > Histogram > With Fit, specify the column of data to analyze, in this case ‘product weight’, and then click OK. Remember to copy the data from the Excel worksheet and paste it into the Minitab worksheet. For this example, we may use the product_weight worksheet. He takes a random sample of 40 products and measures their weights. He is interested in the weight of a food products with a target of 50 grams per item.
Normality test using Minitab and beautiful graphs.Suppose that a line manager is seeking to assess how consistently a production line is producing. One is that basic stats, normality test and the other is graph probability plot and you can choose accordingly. To summarize we have two commands we can use. You remember the null hypothesis abnormality test states that the data follows in normal distribution. This is because the P value is for B and C are less than 0.05. In this case, we can see the A's normal whereas B and C are not. You have the mean standard deviation and number of data points and the symbolic statistic and the P values for the Anderson-Darling test. And that you have it all three graphs and the P values shown here. And here for the variables we can do all three at once against the normal distribution. To do that go up to graph, probability plot and since we have three columns it's multiple Y variables each Y is a column and want them to be displayed by overlaying each other on the same graph, click okay. However, there is a shortcut I like to share with you where you can do all three graphs at once. If it was perfectly normal the blue dots would reside exactly on a red line. Therefore A is data follows the normal distribution. Remember, the null hypothesis in a normality test is that it follows a normal distribution. Let's skip the auto-save and we have the probability plot for A, the P value for Anderson-Darling test shows a 0.751 which is larger than 0.05 our alpha.
And let's use the Anderson-Darling test which is the most common one. To do that go up to stat, basic statistics, normality tests and we have to do one variable at a time, start with A. Here we have three sets of data A, B and C and we want to check whether the data follows in normal distribution.
In this movie I will show you how to run the normality tests using Minitab.