So, we've looked at the numerical displays of information. And just like we had for single categorical variables, where we looked at bar charts and pie charts, we can report this information in other formats. So rather than saying, let's look at a pivot table and let's look at the numerical values, we can put this into a side-by-side bar charts to look at the comparison. So, in this chart we're going back to the movie ratings by studio. For each studio, what fraction of movies come from a particular rating? Within 20th Century Fox, the three bars present add up to 100%. Within Buena Vista, the four bars add up to 100%. Just a way of communicating this that's a little bit easier to process than having to read all of the numerical values. Another way that this can be formatted is into a segmented bar chart. So this is taking the exact same format we had previously, but rather than putting the bars side-by-side for each studio, it's stacking them on top of each other. So you see that the red bar for 20th Century Fox comes up to around 20%. All right, so that 20% is the PG movies. The green bars starts at around 20% and goes up to just shy of 80%. And so if I take that difference, we're talking about around 60% for the PG-13 movies. And then the purple bar, the rated R movies, start at around 80%, go up to 100%. That's also making about 20%. So, if we look roughly at this, we're looking at 20, 60 and 20. That's the challenge with these segmented bar charts. The more categories you have that you're trying to stack on top of each other, it can be harder to parse out how much is in any particular segment. Because I have to look at, where does it start, where does it end? So it's one thing to keep in mind. As you have a lot of categories you may want to stick with the side-by-side chart rather than stacking them up. We'll touch on this idea of independence versus there there being a relationship from a quantitative standpoint. We'll touch on that a little bit later on in this course. But one way that we can look at this visually, would be to say, if we ask the question, is the distribution of movie rating across studios the same? Well, if that's the case, what we should observe is for all of the studios, we should see roughly the same cut points. And you'll see right away that that's not the case. Buena Vista looks very different from 20th Century Fox, looks very different from Warner Brothers. So there is some relationship in this data between the movie ratings and the studios. So again, we'll get into kind of, how do we numerically assess these relationships, a little bit later on in the course. All right, so as far as the techniques that we've looked at, if we're only looking at one categorical variable at a time, we can look at frequency tables, we can summarize that using bar charts and pie charts. If we're trying to understand relationships among two or more categorical variables, that's where the contingency tables, which we produced using the pivot table feature came into play. We'll look at that in more detail for survey data, how can we use the pivot table to extract some insights to drive some business decisions? So, what we're going to make sure we're very comfortable using that pivot table tool from a visualization standpoint. The segmented, or the side-by-side bar charts, are going to be the way of demonstrating the relationship among those that may exist across those different categorical variables. So then in the next module that we're going to work on, the next piece of this course is going to provide you with survey data formatted in Excel. So it's a Microsoft Excel workbook. And the pivot table is going to be our go-to tool as far as formatting that data, summarizing that data, and trying to tie that to the specific business questions that we have. And if you have any questions based on the material that we've covered so far, as you gain some more familiarity using that pivot table functionality, be sure to post those questions onto the course website.
