4. Our advice

4.1 Think about the necessary prerequisites

In this example, all the results are present, the design is complete and the data are assumed to be distributed according to a Normal distribution. A poorly established design, missing data and non-homogeneous variances are all problems when carrying out an analysis of variances.

4.2 Don't forget sampling

One point not to be overlooked is sampling. If you want to study the quality of a batch, this means you'll have to "select" parts from that batch....