It is time to practice with visualizing your data. Remember, that we measured the caffeine percentage of coffee. We already made a boxplot and saw that there was a big difference between the caffeine percentage of coffee produced on three different machines. We also found that machine three did not comply with the maximum of 0.1% caffeine for decaf coffee. However, next to the machine we also measured the batch number. Now, my question to you is the following. Is batch also an influence factor for caffeine percentage, and if it is which batch is the best to produce decaf coffee? Now, pause this video, load your data into minutae, analyze your data, and answer the question before you continue. Are you ready for the solutions? First we have to see what our variables are. The Y variable or the CTQ is caffeine percentage and it's numerical. The X variable or the influence factor is the batch number. Don't be fooled by the fact that it's a number. It is a categorical variable. If you look at a tree diagram, you see that we can make a boxplot. Let's take a look a minute at how to make such a plot. I have copy-pasted the variables caffeine percentage, extractor number and batch number into mini tab. To make a boxplot we go to graph and then, you select boxplot. Now, which bottle should we make? We should make the one with groups, because we have caffeine percentage and we want to show it for each batch. Okay, our graph variable, your CTQ or your Y variable is caffeine percentage. Your categorical variable for grouping your influence factor is, in this example batch number. Okay and this is our boxplot. What do you see? Let's take a closer look. This is a boxplot. We see that there are no big differences between the medians and the boxes for each batch. This, we can conclude that probably the batch number is not an important influence factor for caffeine percentage. Furthermore, you can see that not a single batch is complying with the norm which was 0.1% for decaf coffee. Summarizing. If you have numerical Y and a categorical X variable, you can use a boxplot with groups to visualize the variables. The boxplot showed that the batch Is not a convincing influence factor for caffeine percentage.