Ever seen a graph like this? I bet you have. And it's not very difficult to interpret it, either. If I tell you that this part represents how often I go to work by bike, and this part is how often I go to work by public transport and by car, you can easily conclude that I bike a lot. Which is unsurprising if you know that I live in Amsterdam, where everybody bikes. In this video, I will present to you how to visualize categorical data, like my mode of transport. I will show you how to make and interpret the bar chart, the pie chart, and the tally table. We already saw that the appropriate technique to graphically display your data depends on the type of variable. Okay, let's start with an example. Remember that we talked about international money transfers? If a mistake was made in such a transfer, a client submitted a reclaim, leading to an investigation as to what went wrong and how to correct it. For a sample of reclaims, we have recorded the type of reclaim, the employee that handled it, and the processing time that was needed. There are several questions you can have about this data set. For instance, which type of reclaims occur most often? Or, which person handles the most reclaims? Let's visualize these variables to answer these questions. But first, pause the video to load your data before you continue. This is your data main tab with reclaims in one column, the person handling the reclaim in another, and the processing time in the third column. Let's make a pie chart and a bar chart, and for that you go to the menu Graph, and then you find the Pie Chart and the Bar Chart over here. Let's start with a Pie Chart. Well, we want to make a pie chart for the Reclaim type, OK. And next we can also make a bar chart, go back to Graph and select Bar Chart. And well, we have to have a count for unique values, so the first one, the Simple one, OK. And then you want to make the bar chart for the Reclaims, OK. These are the two charts, let's study this output. Let's take a look at the pie chart. The total number of reclaims in the sample formed a pie. Then the size of a colored piece of pie represents the number of times the category occurred relative to the total amount of reclaims. For example, this orange part is roughly one-fourth of the pie. Then we know that the type matching, which corresponds to the orange color, is also roughly one-fourth of all reclaims. A pie chart is a popular tool. However, for some questions, it's not the perfect graph. Can you, for example, see whether Status info or Matching is a larger category? For these kind of questions, we have a so-called bar chart. This is a bar chart for our data. The horizontal axis shows us the categories. The height of each bin represents how often each category of reclaims occurs. The difference between Matching and Status Info is much more clear. You see easily that Matching is the most occurring category. So, we have answered our first question using a pie chart and a bar chart. Let's have a look at the second question. Which person handles the most reclaims? We can also use a pie chart to answer this question, but for now let's have a look at another technique, the tally table. We can ask Minitab for a tally table, which also shows the frequencies, just like the pie and the bar charts. However, it shows the exact percentages. Let's have a look. To make a tally table, you can go to Stat, and then you to the menu Tables and it's the first option, the Tally Individual Variables. For which variable do we want to have a table? Well, for Person. Let's display the Counts and also the percentages. Okay, your tally table is given in the Session window. Let's study it. The table gives us the same information as the pie chart and the bar chart, only now we get numbers. Apart from the count, we also know the percentage. It is easy to spot that Margriet and Marcel both handled the most reclaims. 22 per person, or 18.03% of the total. In summary, a pie chart shows us how often each category occurs. And a bar chart gives us the same information, but it is easier to spot which category occurs most often. A tally table gives the exact numbers. And each of these are visualization techniques for a single categorical variable.
